LIBRARY.OF CONGRESS. (S|ap SopnMft 1» Slielfl..fi._ZiL>> UNITED STATES OF AMERICA. CLINICAL LECTURES GOWERS. OTHER WORKS BY W. R. GOWERS, M.D., F.R.S. MANUAL OF DISEASES OF THE NERVOUS SYSTEM. A Complete Text-book. Second Edition. Revised, Enlarged, and in many parts Rewritten. With many new Illustrations. Two Vol- umes. Octavo. Vol. I. Diseases of the Nerves and Spinal Cord. Cloth, $3.00 Vol. II. Diseases of the Brain and Cranial Nerves; General and Functional Diseases. Cloth, $4.00 SYPHILIS AND THE NERVOUS SYSTEM. Being a Revised Reprint of the Lettsomian Lectures for 1890, delivered before the Medical Society of London. i2mo. Cloth, $1.00 DIAGNOSIS OF DISEASES OF THE BRAIN. 8vo. Second Edition. Illustrated. Cloth, $1.50 MEDICAL OPHTHALMOSCOPY. A Manual and Atlas, with Colored Autotype and Lithographic Plates and Wood-cuts, comprising Original Illustrations of the Changes of the Eye in Diseases of the Brain, Kidney, etc. Third Edition. Revised, with the assistance of R. Marcus Gunn, f.r.c.s., Surgeon, Royal London Ophthalmic Hospital, Moorfields. Octavo. Cloth, $4.00 THE DYNAMICS OF LIFE. i2mo. Cloth, .75 CLINICAL LECTURES ON DISEASES NERVOUS SYSTEM DELIVERED AT THE NATIONAL HOSPITAL FOR THE PARALYSED AND EPILEPTIC, LONDON. BY W. R. GOWERS, M.D., F.R.S., PHYSICIAN TO THE HOSPITAL; CONSULTING PHYSICIAN TO UNIVERSITY COLLEGE HOS- PITAL ; FORMERLY PROFESSOR OF CLINICAL MEDICINE IN UNIVERSITY COLLEGE. A -5 29 1895 S^U/Ish^ QjC^ PHILADELPHIA : P. BLAKISTON, SON & CO., IOI2 WALNUT STREET. I8 95 . , ^ 1 i> Copyright, 1895, by P. Blakiston, Son & Co. PRESS OF WM. F. FELL & CC 1220-24 SANSON! STREET, PHILADELPHIA. PREFACE. The following Lectures have been delivered at the National Hospital for the Paralysed and Epileptic. They are reprinted from various English medical journals, with the exception of two lectures, and for permission to reproduce them I am indebted to the J. B. Lippincott Company. Queen Anne St., London, June, 1895. W. R. Gowers. CONTENTS. PAGE I. The Principles of Diagnosis of Diseases of the Nervous System, 9 II. Mistaken Diagnosis, 21 III. Argyria and Syphilis, 34 IV. Syphilitic Hemiplegia, 48 V. Bulbar Paralysis, 71 VI. Facial Paralysis, 89 VII. Facial Contraction after Palsy, 108 VIII. Acute Ascending Myelitis, 119 IX. Locomotor Ataxy, I, 129 X. " » II, 143 XI. The Foot Clonus and its Meaning, 1G3 XII. Syringo-Myelia, 175 XIII. The Treatment of Muscular Contraction, 188 XIV. The Infantile Causes of Epilepsy, I, 197 XV. " " « " II, 205 XVI. Neuralgia, 214 XVII. Lead Palsy, 232 XVIII. Saturnine Tabes, 242 XIX. Optic Neuritis, I, ...... 252 XX. « << II 266 CLINICAL LECTURES Diseases of the Nervous System LECTURE I SOME GENERAL PRINCIPLES OF THE DIAGNOSIS OF DISEASES OF THE NERVOUS SYSTEM. Gentlemen: — If we look back over the progress of medical science, three epochs of discovery stand out in special salience, contrasting, in the steepness of the rise in know- ledge they present, with the more gradual progress of other branches and of the same subjects at other times. These are, first, the revolution in the conception of diseases of the heart and circulation which Harvey's great discovery entailed; secondly, the penetrating extension of knowledge of all the thoracic diseases which was effected by the invention of the methods of auscultation and percussion, and the discoveries which followed that invention ; and, thirdly, the enormous advance in our knowledge of the nervous system and its diseases which the last quarter of a century has witnessed. This has been largely due to the development of microscopical research, in some degree to Post-graduate Lecture, Layicet, Feb. 20, 1892. 2 9 io PRINCIPLES OF DIAGNOSIS. the progress of experiment, but very much to the extra- ordinary increase in the capacity for investigation which the general progress of science has produced, and to the fertile field presented by these diseases for the exercise of that faculty. The results of the application of the chemical and microscopical investigation to the urine, and the trans- forming revelations made through the invention of the ophthalmoscope and the laryngoscope, are hardly less remarkable, but their range is more limited than those of the progress effected during the three great epochs in discovery that I have mentioned. The last of these con- cerns us in a special manner, because it has taken place in our own time and is still in active progress. It is to certain consequences of it that I desire to direct your attention to-day — consequences that are of the utmost importance to the practitioner in his daily work. How great, how wide and profound, has been the change that our knowledge of diseases of the nervous system has undergone cannot, indeed, be realized by all members of our profession. A large number of those who studied before the change took place have been unable to follow the new development of knowledge, — the successive stages of which have followed each other with a rapidity almost bewildering, and the difficulty has been enhanced by the freedom with which investigators have revelled in a novel nomenclature, embracing alike the new and the old, in part essential, but in still larger measure superfluous. The exigencies of practice, with its increasing demand on the time and energy of the practitioner, and its unceasing echo of the cry (shall I say of a colleague's daughter?) " Give, give," has made it often impossible for him to attempt, or, attempting to continue, to follow the successive discoveries, Without a knowledge of their scientific features he has generally found it impracticable to use them, and so he has either rested content with that which he learned of old, or CHANGED CONDITIONS. II has gained a fragmentary knowledge that has served only to mislead him. Those, on the other hand, who have learned the subject as we have it now, cannot realise the change, for they have but one of the terms needed for the comparison ; while the large number of practitioners who studied medicine during the period in which the change was taking place, learned a little of the new and a good deal of the old, and acquired a mixture containing many incompatibles, with the common consequence that may be described in terms of metaphor, not inexact, as " mental turbidity." To clear your minds and get rid, as far as may be, of all that prevents a clear vision of disease, to learn how most usefully to see disease through the clearer media, and, above all, to see it in such a way as shall enable you to treat it with such advantage as is attainable — these, I con- ceive, are the objects that bring together here the members of this post-graduate course. But I am not sure that the students of to-day, or at least many of them, who have acquired the latest and fullest knowledge of these diseases, have learned, in equal measure, how to use their knowl- edge, or are free from the need of that training in method which the practitioner feels so keenly. The reason for this, or at least the conditions that make it probable, you will perceive when I tell you the nature of the difficulty that is experienced and the character of the need that is felt. Let me, however, first give you an illustration of the change that has come over our knowledge of diseases of the nervous system during the last quarter of a century. I remember, twenty -five years ago, listening to a lecture on " chronic spinal meningitis," given by one of the most competent teachers of the day, which presented the latest and clearest opinions on the subject. The malady, as described in that lecture, had no existence. In the case on which it was given, the membranes of the spinal cord were 12 PRINCIPLES OF DIAGNOSIS. certainly in a perfectly normal state. The symptoms then thought to prove the existence of chronic meningitis are now known to depend solely upon changes within the spinal cord itself. The pathology had been inferred from analogy, but the new means of investigation we have since obtained, coupled with the more extensive and precise observation which I have mentioned as so salient a feature of recent progress, have proved that the analogy was false and the malady quite different from that which was sup- posed to exist. The symptoms which were then ascribed to chronic meningitis are, indeed, now known to manifest more than one distinguishable disease. But the chief source of practical difficulty is not the mere correction of erroneous inference, and the displacement of the work of fancy by ascertained fact; it depends on the vast increase in our knowledge of the anatomy and physiology of the nervous system and on the application of this to disease ; it depends, further, on the revelation, effected by micro- scopical research, of the extent and variety of degenerative processes in all parts of the nervous system. It depends, lastly, on our ever-widening perception of so-called " functional," really nutritional, diseases for which no anatomical cause has yet been discovered, which have been distinguished or recognised by means of (i) the increased exactness of observation, (2) the aid received through the differentiation of processes formerly confused with these, but now known to depend on structural change, and (3) in consequence of the special indications afforded by the discoveries now being made of the vast range of influence of toxic agents on the nervous system. Can we wonder, then, that our knowledge has been changed, even to the point of transformation, and that those familiar only with the old should be almost lost in the mazes of the new? The practical effect of these changes is to render the cus- tomary methods of diagnosis wholly inapplicable to a large INADEQUACY OF TYPES, 13 proportion of cases of disease of the nervous system, to render it useless, or more than useless in the case of a large number of the nutritional affections that are called " func- tional," and of the structural maladies that are called " degenerative," as well as many of the maladies that depend on the visible and conspicuous changes that are ca'led " organic." The student has to acquire his knowledge of all diseases by the method of ''types," and the same method is em- ployed in the practical diagnosis of most other maladies. Indeed, in the case of many, no other system of recognis- ing the nature of a case could be employed than that of observing the correspondence of the symptoms, in charac- ter and course, with those of a certain form which is known by a special name. But when we turn to the diseases of the nervous system that I have indicated, we meet with a series of maladies the variety of which is aimost infinite. When we consider the various structures comprised in the u nervous system," their diversity of function, the extremely numerous combinations of structure and function that may be deranged at the same time, all mutually acting and re- acting to produce the manifestation of disorder ; when we consider further the variety of nutritional and degenerative processes to which these structures are liable, and also the degree and extent to which they are influenced by toxic blood states — agencies that we are only beginning to per- ceive in their nature and diversity, — we can readily under- stand that it not only may be. but that it must be, true that the diseases we meet with are so numerous and so diverse as entirely to baffle an attempt to range them under definite types, that can be designated by definite names. But in the series they form, there are here and there aggre- gations, as it were, in which the morbid process corre- sponds, at least in its chief features, varying only in minor points. These we treat as types, and call by names, and 14 PRINCIPLES OF DIAGNOSIS. by them the student learns the general features of the classes of disease in which they occur; but they seldom stand alone or are marked off by any well-defined bound- ary. Intermediate cases connect adjacent groups — cases that combine the features of two or several, or present some characteristics of one and some of another, and very often symptoms that cannot be recognised as pertaining to any recognised type. Of these forms, numerous as they are, the student for the most part remains ignorant. He has enough to do to become acquainted with the more familiar groups, and the method by which he learns the features of these cannot take in the intermediate varieties. When he proceeds to practical work he soon finds himself confronted by cases the like of which he has never heard of, and which he cannot find described in any text-book of these diseases. If he attempts to identify them, to bring them under some recognised type, and to give them some definite name, he is at once and completely baffled. He is baffled of necessity, for the cases belong to no type, and no name can be given to them that does not involve more error than truth. It cannot be otherwise, for to constitute types and assign designations that would include even the chief of these varieties would be to multiply both types and names beyond the capacity of the human mind to learn, or its power to retain. In the case of these diseases, therefore — diseases that are so frequent as to make up at least half the cases of affections of the nervous system which the practitioner has to treat, — the method by which he has acquired his knowledge, and by which he is able to deal successfully with almost all other classes of disease, fails him entirely; and yet in most instances he has nothing to put in its place. " The old order changes " ; but it is " giving place to new " with a slowness which, at the present day, is a source of perplexity and difficulty widely felt, and constitutes a really serious obstacle to the practi- THE REMEDY. 15 tioner's work. That it is so — that I am not exaggerating the facts — will be, I am convinced, confirmed by most of those who bring to their work the earnest and thorough spirit which strives to understand the cases that have to be treated, which strives after light, and is not content merely to deal, in darkness, random strokes. What, then, is the remedy ? What new method of diagnosis can be adopted ? Happily, there is a means of meeting the difficulty, before which its formidable features will vanish, and the ever-growing accumulation of facts, the ever-increasing variety of definite morbid states of the nervous system, however numerous, however complex, however unfamiliar, may be dealt with as readily, as surely, as the diseases of any other class, or as the affections of the nervous system that do correspond to types, and may be recognised by such correspondence. Indeed, I had almost said that the method I am about to recommend enables you to deal with these diseases more readily and more surely than the other. At least, it has the advantage of placing the practitioner at once in the position needed for the adequate and wise treatment of the malady so far as such treatment is within our reach. The method you should adopt is this : Whenever you find yourself in the presence of a case that is not at once and completely familiar to you in all its details, forget for the time all your types and all your names. Deal with the case as one that has never been seen before, and work it out as a new problem, sui generis, to be investigated as such. Observe each symptom carefully, and consider its significance. Then put the several symptoms together, and consider the meaning of their combination, especially whether there is any one part of the nervous system at which disease might cause them all. Lastly, consider the way they came on, as indicating the nature of the lesion, comparing this with the evidence of their seat, and remem- 16 PRINCIPLES OF DIAGNOSIS. bering also that their character may in itself tell you some- thing of their probable nature. When described in the abstract, this may seem a lengthy process. It may even seem a formidable process. As a rule, it is neither. The common symptoms, even those presented by uncommon cases, are not numerous, so far as concerns their general character and actual nature. The question of localisation is only an application of the common physiology of the nervous system, of the facts that should be familiar to every student, and can be re-learned, if necessary, with ease, by every practitioner. All the knowledge needed for this method is that which every student gains, or ought to gain, in the course of his studies; it is only the mode of using the knowledge that is new and has to be acquired. But the student should remember the great importance of " keeping up " his physiology and anatomy of the nervous system, or at least those parts of it which are needed in practical work. There is no department of medicine that consists more largely of applied physiology and applied anatomy than these diseases. For this reason they should engage the attention of the student early in his hospital work, instead of, as is often the case, being relegated to a late period on account of their supposed difficulty — a period when his science has got " rusty," or has slowly vanished, until even its nomenclature awakens a mere echo from bare walls. As a matter of fact, much of the student's hindrance is due to this postponement. The application of his knowl- edge should be made to retain it. But for the successful use of this method it is essential that the knowledge, though neither extensive nor profound, should be firmly grasped and boldly used. Herein lies the chief practical difficulty. Timidity is almost a greater hindrance to diagnosis than is ignorance. You must feel sure of the meaning of symptoms, you must weigh the evidence with care, and then you may and must feel confi- FROM SYMPTOMS TO PROCESS. 17 dent that your conclusion is trustworthy. This confidence and boldness can only be acquired by familiarity with the process, by observing its use by others, and afterwards repeating it for yourselves, thus becoming so familiar with the language of disease that you can read it with ease, can see at once the meaning of its words, and perceive with readiness the significance of its sentences. It is by afford- ing an opportunity for this that the attendance on the practice, especially the outpatient practice, of special hos- pitals, is of peculiar value to the student, whether pre- graduate or post-graduate. A series of cases of the same class pass before him, in each of which he can observe the character of the symptoms and the process of diagnosis, and thus he gains, in a short time, a familiarity with the features of disease, and a quickness in perceiving their meaning, that he could not obtain in a long period of work at a general hospital, where such cases are few and far between, and the lesson of the one is more than half forgotten before another instance comes before him. In all this method of dealing with unfamiliar maladies there is nothing that is not within easy reach of the average student or practitioner. If any difficulty is felt, it is only, as in so many subjects, the first step that costs a conscious effort. Once learned, the method becomes more easy with each repetition, and is acquired the more speed- ily if repeated under slightly varying conditions. Soon, its steps cease to be consciously felt, and in a short time even a student who has been fairly grounded in the ele- mentary knowledge, becomes able to use the method with precision, and succeeds, in four-fifths of the cases that would otherwise baffle him, in arriving at a diagnosis in which a physician whose attention has been given for years chiefly to these diseases, can find little or nothing to change. Note, moreover, that the method indicated is that which the most experienced physician has himself to adopt 18 PRINCIPLES OF DIAGNOSIS. in the case of a large proportion of the cases that come or are sent to him. So infinitely various are the morbid states of the nervous system, so diverse their manifestations, that a very large number of the cases seen are practically new, even to a man of the largest experience, and if asked he has to confess that he has not seen before any case pre- cisely similar to, and often not one even approximately resembling, that before him. Yet, by employing this method, he is able to arrive without difficulty at a diagno- sis as sure as when the malady is of a common type. The only difference between his work and that of the student is that the latter has to adopt the method with scrupulous care, in a larger proportion of the cases that he sees. He has to deal with a larger number as essentially new pro- blems, to be worked out de novo. If he does this, as I have said, his errors will be few and rare, and only a small proportion of the cases will baffle his efforts. But never forget the essentials of the proceeding. Remember that you must for the time discard entirely your types and names. When you have made your diagnosis in the man- ner I have described, then, and then only, may you con- sider how far the case corresponds to a type and can be called by a familiar or unfamiliar name. The last point suggests a common difficulty. The desire for a name is strangely strong in the case of the majority of patients. Unless their disease is designated, they go away unhappy, discontented, distrusting. " But you have not told me what is the matter with me " is their parting plaint. What are you to do ? In a very large number of cases no recognised name can be given to the disease that does not involve more error than truth. A few patients are sufficiently sensible to endure the knowledge of the fact and to be content with it. In the case of most, the best plan is to give them a descriptive designation, and to write it down, that there may be no mistake, or it will come NAMES AND PROCESSES. 19 back to you some day in altogether altered form, so changed that you do not recognise your own production. The descriptive designation need not necessarily be short, although often it can be, and it then satisfies as much, pro- vided the words are unintelligible. " Cephalic dysesthe- sias " makes many a sufferer content. " Paroxysmal neural- gic pains, due to damage to the third and fourth posterior spinal roots, owing to a stationary lesion of the interver- tebral articulations," delighted one good woman whom I lately saw, and who earnestly desired to know what was the matter with her, no doctor, she said, having as yet informed her. But always adhere strictly to fact in these descriptions, and avoid words that are apt to mislead because used in various ways. If you use the term "scle- rosis," for instance, the chances are that the patient will look up the word in some medical dictionary, will identify his malady with insular sclerosis, read, forecast his future, and become utterly miserable. To sum up in a few words the necessary change. Dis- card in the first instance all attempts to identify or to name, and try instead to read the malady, tracing the symptoms to the seat of their cause, and discerning the nature of the morbid process by their character and course. The method has the great practical advantage of taking you at once to the elements that should guide your treat- ment, and of enabling you to treat wisely a case the like of which you have never heard of, and a name for which you may not know. Mind, I do not say that you need adopt this process in every case of disease of the nervous system that comes before you. There are many that do belong- to familiar types, cases that are characteristic and preceived at once to be such — cases of common chorea, for instance, of idiopathic epilepsy, many forms of hemi- plegia, and the like. But the method is needed in all unfamiliar maladies, and in all cases even of familiar aspect 20 PRINCIPLES OF DIAGNOSIS. that present some unusual feature; often, such a feature, in the end, is found to indicate that the case merely resem- bles superficially that to which it seemed to correspond, and is really quite different in nature. How real and how extensive is the need for the change in method I have described can only be realised by those who have to correct the mistakes or supplement the de- ficiencies of those who first attempt to make a diagnosis of such cases as I have indicated. In many cases a practi- tioner candidly confesses his inability. I often receive a letter saying, " I cannot make out what is the nature of the disease ; I have seen nothing like it before, and I can- not find its features described in any text-book." Yet the case, as a rule, readily yields to the application of the method I have described. More often, however, the prac- titioner has endeavored to make a diagnosis by the familiar system of types, and has, as it were, forced the case into the receptacle that seemed to correspond most nearly to it in form, but into which it did not really fit, and has given it a corresponding name, more or less erroneous, and often ludicrously wrong. I abstain from giving you instances ; the statement of the fact will suffice, and examples are so numerous that some one would probably find that the cap fitted him with unpleasing precision. I should like to have given you some instances of the working of this method, but for this there is not time. You will find it easy and more useful to work out such examples for yourselves, beginning with familiar diseases, but treating them, for the occasion, as unfamiliar. I hope, however, shortly to give you in another lecture a series of examples of its use, although not nominally such, by describing and analysing some of the more important di- agnostic signs, or " diagnostic guide-posts " as we may call them. Familiarity with these is of extreme value, and most of them will afford illustrations of the use of the method I have just described. LECTURE II. MISTAKEN DIAGNOSIS. Gentlemen: — It is always a pleasant thing to be right, but it is generally a much more useful thing to be wrong. If you are right, all that you do, as a rule, is to confirm your previous opinion, your previous habits of reasoning, and your previous self-esteem. But if you are wrong you gen- erally gain in knowledge and gain perception of the way in which your method of diagnosis needs improvement, and the influence on self-esteem is not likely to do you harm. At least that is my own experience, and I think I have observed it confirmed in others. But the result is depend- ent on deliberate effort. There is a strong temptation to smooth down error, and it is very easy not to gain from it its precious lesson. It is more easy to fancy that there is some accidental cause for the mistake than frankly to per- ceive that it is a fault But if you make a deliberate effort to realise and to face in your own mind the mistake you have made, to discern its cause, and to employ this percep- tion as far as you can to remove the cause and prevent a like mistake in the future — if you do this, almost every error becomes one of the precious experiences of your practical life. Yet you will note that I have not said this quite abso- lutely. I said it is " generally " more useful to be wrong, that " almost" every error is useful. As a matter of fact, there are errors you cannot avoid. Beware, however, of Post-graduate Lecture, British Medical Journal, July, 1894. 21 22 MISTAKEN DIAGNOSIS. thinking that any individual instance of error is of that character. The chances are great that it is not, that it might have been avoided, and they are considerable that it will be avoided in the future as a consequence of its occur- rence. At least I hope so. Still, there are cases in which error is inevitable. I should like to mention to you an instance which illustrates the conditions on which the relation depends. But this is not the chief subject I desire to bring before you. My chief object is to seize the opportunity afforded me by the courtesy of a private patient and his desire to be useful — the opportunity of illustrating avoidable mistakes by an example almost perfect in its character, almost unique in its features, and seldom equalled in the variety and import- ance of the instruction it affords. The case is indeed full of important lessons, and it is one the like of which, I ven- ture to say, not one of you is likely to have seen before or is likely to see again. But, first let me say a few words about inevitable errors. There is a patient in the hospital at this moment whom I cannot show you now, but whose case is an example of mistakes that must be. I saw him first about two months ago, at the commencement of his illness. I then made a diagnosis, of the correctness of which I felt assured, and which I am now certain was wrong. Yet if I saw a like case to-morrow with the same symptoms, with all the knowledge I have gained from this case, I should make the same diagnosis, that which proved to be wrong ; and the chances are fifty to one, or more, that it would be right. Why, in this instance, was it wrong ? Because all diagno- sis that rests on reasoning is a matter of probability ; only that which is simple observation is certain. The proba- bility may be great, or may be only that of a slight pre- ponderance of the balance of evidence, but wherever infer- ence comes in there is no certainty. Remember that infer- THE ROOM FOR ERROR. 23 ence plays a part which you can discern only by an effort, in a large part of diagnosis. You think you observe the presence of consolidation of the lung or of pleural effusion. You do nothing of the kind. You infer its existence from certain physical signs. I remember an incident which illus- trates the risk of error thus involved. A skilled physician gave orders to his house-physician that a surgeon should be called in to tap a chest, one side of which was full of fluid. When the surgeon came he took notice of the chest before plunging in his trocar, and, being a man of a thoroughly scientific mind and used to observation, he was struck by the fact that there was no enlargement of the chest. Although he was a surgeon, he percussed carefully, and found that every other sign of pleural effusion was present except enlargement of the side. He declined to tap. A little later the patient died, and the lung was found to be a solid mass of cancer. That was a case of incomplete observation, but it illustrates the fact that even a condition like pleural effusion, which you seem to observe, is really recognized by complex inference, and if either the process of reasoning or the observation is deficient, conclusion may be wrong. To return now to the case I mentioned. The patient is a man of forty-eight years. I saw him a few hours after he had been seized with hemiplegia, which had come on in a few minutes. Such a sudden onset means a vascular lesion, the rupture, or closure of an artery. The onset had occurred without any loss of consciousness and without prodromata. There was no sign of degeneration or of kidney disease. The ophthalmoscopic appearances were normal. The hemiplegia was complete; but the entire absence of any loss of consciousness made it distinctly unlikely to be hemorrhage ; and the facts that there was no tension in the pulse, and that there was a perfectly nor- mal heart free from hypertrophy, made rupture of a vessel 24 MISTAKEN DIAGNOSIS. still more improbable. The absence, so soon after the onset, of any indication of a source of embolism excludes embolism — as a practical fact ; and the absence of kidney disease made atheroma at that age most improbable. When no evidence of any other cause of arterial occlusion can be discovered, and there is no condition to cause the rare spontaneous thrombosis in a healthy vessel, syphilitic arterial disease is probable. It is the condition left by the exclusion of others, and it can very seldom be excluded. I came, therefore, to the conclusion that it was a case of syphilitic arterial disease leading to the sudden formation of clot in a branch coming off from the diseased part of the artery. But not only could syphilis not be excluded in this case ; there was a history of it. The patient was per- fectly conscious and able to talk, and gave me a history of syphilis more than twenty years ago; an intimate friend, moreover, told me of a further history of syphilis as recently as eight years ago. Under these circumstances the diag- nosis of syphilitic arterial disease and thrombosis was the proper diagnosis, and treatment was arranged accordingly. The patient was admitted here a short time afterwards, and, instead of improving as he should have done, he has got worse. By "improving" I mean in the general cere- bral symptoms, not necessarily improvement in the hemi- plegia. You cannot by treatment restore the lumen of the vessel or the tissue destroyed by necrotic softening. The treatment is capable only of influencing the condition of the wall, and not its consequence. Not only has the hemi- plegia continued unchanged, as would be compatible with the diagnosis, but the general cerebral symptoms have become gravely worse ; his mental state has become increasingly dull, weakness of the sixth nerves has devel- oped, and, although the optic discs were perfectly normal when I saw him, optic neuritis has developed with a rapidity and to a degree of intensity that I think I have INEVITABLE ERROR. 25 never seen before. It is certain that he is suffering from a tumour of the brain that is not syphilitic, and that in some way by pressing an artery the tumour led to the sudden formation of clot in it, and the sudden symptoms of the arrest of the blood supply in its region. Of course, I need hardly add that he was treated by iodide of potassium in full doses, and soon afterwards mercury was added and pushed just to the degree of affecting the gums ; yet not only without improvement, but with the contrary. I may incidentally add that the mercury was pushed to the degree of affecting the gums because, unless you see indications by the gums that there is enough mercury in the blood to act upon the tissues, the presumption is that the mercury is eliminated so fast that there is not enough in the blood to act upon the syphilitic tissue. Such a combination of symptoms is seldom met with ; but we do occasionally meet with a combination of symp- toms which illustrates the same rule — that the diagnosis which it is right to make is wrong in fact. I remember another instance of inevitable error in diag- nosis, which I will briefly mention. An old woman, aged sixty, was admitted in profound stupor and with indications of bilateral hemiplegia and irregular damage to the cranial nerves. This state had come on so suddenly we had no doubt that there was occlusion of the basilar artery. She was sixty, and presented signs of degeneration : therefore thrombosis in a branch of an atheromatous artery was probable. But she had also mitral constriction, the cardiac lesion which of all others is most prone to cause embolism. Embolism of the basilar artery, curiously enough, has been thought to be impossible, because the artery is larger than either of those by which the blood reaches it, but mani- festly a plug may pass through one of the vertebrals which cannot pass through either of the posterior cerebral arteries, 3 26 MISTAKEN DIAGNOSIS. which are smaller than the vertebrals. As a matter of fact, I have seen a distinct embolus in the anterior extremity of the basilar. Therefore embolism was also probable, for no age excludes it. The patient died, and was found to have syphilitic disease of the basilar, perfectly characteristic. This we had no reason to suspect. It is another instance of a case in which a correct diagnosis was impossible, and error was inevitable. From such cases only the general lesson can be learned, that accuracy is occasionally impos- sible ; we can only be right in nineteen cases by being wrong in the twentieth. It is well to realize this. But remember that in practice we have to treat that which is only probable as if it were certain. We could not treat two-thirds of our cases properly without doing this. But always discern the degree of probability, and if the proba- bility is low, reduce or modify your treatment so as to do as much good as you can, if your opinion is right, without certainly doing harm if your opinion is wrong. You will find that this principle is applicable to many and various contingencies. And now we pass to the case which I want especially to make the subject of your attention to-day; I should rather say "the object," because the attention that I want you to give to it is to a considerable extent by the use of your own eyes. The patient is a man who first consulted me on February 9th, complaining of numbness in his hands and feet, with pains, and twitching of the muscles, which were increased when he walked. These symptoms were of three years' duration and had gradually become worse. They began after overwork. He had sharp momentary pains in both feet, sometimes in the toes, sometimes in the soles, and sometimes in the balls of the feet — not much in the legs, and none in the arms. There was no trace of knee- jerk ; although there was a little complaint of unsteadiness, he could stand, with his eyes shut, fairly well. There was AVOIDABLE ERROR. 27 no incoordination in the hands, and no deficiency in their sensitiveness. The pupils acted to light. He had had slight brief double vision. There was diminution in sexual power but no difficulty in micturition. These symptoms, especially the loss of the knee-jerk and the sharp pains, afforded a strong presumption of tabes. Inquiry after the common antecedent of tabes gave no evidence of it, but did not enable it to be absolutely excluded. He came late one morning, after I had finished my work, and I had no time to examine the state of sensibility upon the legs ; I made a note that it was to be examined next time I saw him. I generally refuse to see a patient unless there is adequate time for investigation, but it is not easy to resist the urgent desire for an interview when a patient has come frpm a distance. I made the diagnosis of probable tabes, and ordered him a mixture containing belladonna, quinine, and arsenic. He came again, about five weeks afterwards, say- ing that he was about the same. The pains were a little less, but the other symptoms were still troublesome. I then proceeded to do what I had not been able to do on the first interview — to test his sensibility. Among the many aphorisms I heard from the lips of the greatest bedside teacher whom any living person remem- bers, was one that flashed across my mind when, to test sensation, the patient's skin was bared. I remember hear- ing Sir William Jenner once say : " Gentlemen, more mis- takes are made, many more, by not looking than by not knowing." To my astonishment, almost to my consterna- tion, I saw that the skin presented everywhere the charac- teristic pigmentation produced by arsenic. It was a case of arsenical poisoning simulating tabes. And I had pre- scribed arsenic for him ! Before his illness, and during its first year, he was by occupation an oil and colour mer- chant, handling papers of various tints, and all sorts of pigments, many no doubt containing arsenic. He had also 28 MISTAKEN DIAGNOSIS. during the first year he began to suffer taken a tonic mix- ture containing arsenic; but he did not take this long enough for it to do more than intensify the poisoning, which had no doubt been the result of his occupation. He had been exposed for many years ; and it is probable that during the two years which elapsed between the develop- ment of his symptoms and the time I saw him, that which I did others had done. Arsenic had been given him to cure his symptoms. I need hardly say that I changed the prescription. He has not, however, improved. I gave him iodide of potas- sium, which seems to have a definite action in eliminating arsenic from the system ; but the course of his symptoms during the last three months is curious ; they have rather increased than lessened, and especially the condition of the skin that I will show you in a moment, has become more intense. Unless there is some continued cause of arsenical poisoning, and we cannot discover it in any way, I think the iodide of potassium must have been elimin- ating the arsenic from the tissues to such an extent as to increase the amount in the blood to a degree that has fur- ther irritated the damaged structures. Before, however, we examine the skin, let me remind you that the nerve symptoms in arsenical poisoning are most important. In acute poisoning they are met with after the acute symptoms have subsided ; they come on gradually, and for a time increase. The arsenic taken in during the acute poisoning seems to enter into the nutrition of the nerve elements and gradually to derange their function and their structure. In chronic poisoning there is a gradual interference with function. This fact, that arsenic passing into the nerve structures, perhaps partly in place of phosphorus, should first gradu- ally modify their function, is not surprising. Function depends upon the release of force — nerve force — by chem- TOXIC INFLUENCE. 29 ical combinations ; and the amount of nerve energy latent in complex compounds, released when similar compounds are formed, depends upon the constituents ; so also does the readiness with which it is evolved and the kind of stimulus that excites the release. But the change in nutrition which at first disturbs function in greater degree, changes the structure, and leads even to disintegration of the substance of the nerve elements. As a fact, this is what arsenic and other metallic poisons do. Their action presents certain features which you should always bear in mind. They reach the structures by the blood. The struc- tures on the two sides have bilateral symmetry in intimate constitution as in outward form, and that involves a simi- lar susceptibility on the two sides. Hence it is that these poisons produce symptoms that are bilaterally symmetri- cal. Moreover, they produce symptoms which are limited according to function. There we have a mystery, and yet it is a mystery we must recognise as an indication of fact. Function must be related to the chemical and molecular constitution of the nerve elements. The difference in sus- ceptibility to different forms of force must depend on atomic constitution and molecular arrangement. We can- not conceive that the nerve endings in the skin which pro- duce a nerve impulse on the slightest touch are quite the same in constitution as those which produce a nerve impulse when warmth is applied. We cannot believe that the same kind of structure can subserve the susceptibility to the massive motion of a touch or to the wave motion, only a little less frequent than the waves of light, which consti- tutes heat. And that difference, we can understand, is equal or greater in the motor and sensory terminal struc- tures on which toxic influences for the most part primarily act. This enables us to understand why it is that poisons act specially upon structures with certain functions, and 3© MISTAKEN DIAGNOSIS. should even act first on one set of sensory nerves and not on another set. So we have, from different poisons, differ- ent effects upon the nerves, and we have also different effects from the same poison, probably dependent on the fact that it enters the blood in somewhat different form, or with greater or less rapidity, and thus is assimilated by and deranges certain structures rather than others. Thus we have, from arsenic, palsy of the extensors of the arms and feet, as we have from alcohol. We have sometimes a per- fect simulation of tabes, ataxy, muscular anaesthesia, and loss of knee jerk, in consequence of the preponderant affec- tion of the afferent muscle nerves, and sometimes we have chiefly sensory cutaneous symptoms, although scarcely ever without loss of the knee-jerk. Remember, that loss of the knee-jerk only indicates affection of the afferent muscle nerves, and that we know that the impairment of these nerves has many causes. To all toxic influences they seem specially susceptible. Hence, I beg you to remember, above all things, as a practical rule, that whenever you have bilateral disturbance of the peripheral nerves, symmetrical, with a distinct limit- ation to function, sensory or motor, it is almost certainly due to a blood state. There is reason to believe that the degeneration which may appear to constitute an exception is really due to a blood poison. Many are the cases in which the recognition of that fact will save from grave error. In the nerve symptoms of this patient there is little more for you to observe than I have told you, but if you ascer- tain for yourselves the absence of the knee-jerk, the fact may be more securely fixed in your mind, and it cannot be too surely held. But, as regards the cutaneous symptoms, there is a picture for you to study such as you will not see again. This arsenical pigmentation is so characteristic that the moment I saw it I was sure of its nature, but I never ARSENICAL PIGMENTATION. 31 like my own certainty to be without corroboration if it is outside my own special region. So I sent the case to Dr. Radcliffe Crocker, and he is also absolutely certain. He has seen, he said, one more intense case, and he has kindly allowed me to show you a plate from his Atlas, which illustrates it. I will also read to you his description of the pigmenta- tion. But I will add first one more remark. How I have become familiar with the pigmentation from arsenic is because I have often seen it in the cases of epilepsy. I have also seen it in well-marked form, in a lady whose love for working on muslins led her to their continuous hand- ling, and she preferred aesthetic tints. The result was that she began to suffer from pains that were ascribed to gout, until there came progressive palsy of the extensor muscles of arms and feet. This showed a toxic influence, by the symmetry and limitation I have mentioned. Then the pigmentation was discerned, although, I believe, chemical analysis had already made certain the cause. But the troublesome pustular eruption which bromide causes in most persons can only be prevented, or kept down to insignificant degree, by arsenic. No other drug, no modifi- cation of the way or form in which bromide is given, has any influence. Some patients are peculiarly prone to the rash, and have to take the drug for a long time ; they have to take arsenic also. A slight degree of pigmentation is often produced in such cases. It has, therefore, several times happened to me that I have had to put to the patient the two evils, and say, " Which will you have, the spots or the pigmentation ? " There is no other help for it ; they must have one or the other. They always choose the pigmentation. It does not, however, increase to a really serious degree, and I have not met with any other symp- toms of chronic arsenical poisoning. If you observe carefully the skin of this patient, you will 32 MISTAKEN DIAGNOSIS. see that the pigmentation begins as small dots, rounded, which coalesce, but leave small rounded areas of unpig- mented skin. These appear whiter than normal, perhaps only by contrast. The pigmentation is abundant every- where, but extreme on the neck, front and back of the trunk, arms and thighs. For the most part it appears as a pure process, but in many places you will observe small round spots of congestion, of the same size, which' suggest that, at least when the process is going on actively, con- gestion may be the first part of the process. Moreover, in the neck, where the skin has been subjected to habitual friction, the pigmentation is so intense as to be practically continuous, and with it so much congestion is combined as to give the skin a reddish or purplish-brown aspect. The congestion is also conspicuous on the hands, and there, on the palms especially, the spots are attended with distinct elevation. Indeed, where the epidermis is thick, there are minute raised elevations, at which no indication of conges- tion or pigmentation can be discerned. The change in the palms, as you have just learned from Dr. Radcliffe Crocker's description, may go on to a form of " palmar psoriasis." The great fact is the pigmentation, at first essentially punctate or in small spots. It is that that is so character- istic. You will at once understand how important it is to be able at once to recognise a sign so distinctive and so significant. You cannot note the state too carefully. Once firmly fixed in the visual memory, it will not be lost. It is seldom easy to retain an image securely from a single opportunity of inspection, but this condition is one, I think, which you are not likely to mistake when you see it again. You may have some other opportunity of observing a similar condition. You probably will have such if you carefully examine the skin of patients who have been, for long, taking arsenic with bromide. But the story of the BILATERAL SYMMETRY OF THE RASH. 33 origin of this patient's illness suggests that a careful inspec- tion of the skin of a number of men employed in shops where paints and papers are sold might afford an inquiring student opportunities of observing the pigmentation, which would be beneficial not to himself alone. It is certain that the change in the skin may long precede any other symptom. Having helped you so far, gentlemen, I must now ask you to help yourselves. Observe, also, what I have not mentioned, the bilateral symmetry of the rash, a feature that you will recognise, from what I have said, to be inevitable. Connect it with the bilateral pains and loss of knee-jerk, and fix these features of limitation and symmetry in your minds in association with the toxic cause. For a moment, indeed, forget what the cause here is, forget the precise character of the symptoms, in order that the general facts of combined functional limitation and bilateral sym- metry may stand out more clearly in your view as land- marks which, when and wherever they are seen, will always guide you surely. LECTURE III. ARGYRIA AND SYPHILIS. Gentlemen : — It is the duty of a clinical teacher to bring- out of his treasury things old and new. He is constantly under some temptation to present ideas that are new, because they possess more intrinsic interest, although it is generally more useful to give those that are old. Truth and novelty are by no means necessarily associated, although a familiar phrase which suggests mutual exclusiveness goes too far. Nevertheless, it is well for a teacher, if he can, to resist the attraction of the new, and it is always unwise for him to hesitate to inculcate that which is old merely because it is old. Years ago, when I was engaged in giv- ing the elements of clinical instruction to students who were beginning their practical studies, I used to paraphrase the saying of Demosthenes regarding oratory and maintain that the first thing in learning is repetition, the second repe- tition, and the third repetition. A teacher should remember that to neglect to repeat is an unpardonable sin. There is an unpardonable sin of the student ; in fact, it was the re- membrance of this that made me use the term. It is not my own idea. Years ago Sir William Jenner used to say to us (and whatever Sir William Jenner said may be a priori regared ascertain) that the unpardonable sin of the student was to say " Yes " when he ought to say " No " — to say that he heard a thing, that he felt a thing, that he under- stood a thing — when he did not. For you this lesson is probably useless, because you have British Medical Journal, December I, 1894. 34 AN AID TO PROGNOSIS. 35 all passed the stage at which it is needed. But it may not be too late to remember that it is a mistake to shrink from unattractive repetition. Consider — what are we all learn- ing, or should be learning, on this point, from our great Teacher? What does Disease impress upon us ? Disease is forever repeating to us the same things. All the more important laws and rules are ever being pressed upon us, in varied tone, by varied emphasis, or in varying language, but ever repeated. No repetition should be or will be use- less to us if we take the requisite pains, and are not deterred by weariness from striving to discern the lessons that seem the same, but present always an important dif- ference ; and no repetition of fact to us by disease is ever superfluous, unless we will to make it so. Yet the teacher is, as a rule, compelled to take as a subject for his instruction some fresh illustration of disease. It is seldom that he has an opportunity of taking that which is old. When he can it is well for him to do so, for more than one reason. All old cases afford an opportunity of observing the changes in disease, and the effects that time permits. The question which ever presses on us when we meet with disease in its active stage is this : What will be the future ; what will be the result ? We can only learn to look forward by taking every opportunity of looking back. We cannot combine in our observation the future and the present. Our own experience and that of others may enable us to guess something of the future from the present ; but every personal observation that increases our ability to forecast, everything that helps to make the guess more than a guess, is important. This help can only be obtained when the future has changed to the present. We can realise the past, and discern its relation to the future when that which was future has come. We cannot, with the same confidence, realise the unknown future and apply it to the present. Some of you may remember the lines 36 ARGYRIA AND SYPHILIS. which come to my mind — lines by a poet who is almost forgotten now, although his name should live in the mem- ory of medical readers. Sidney Dobell wrote in one of his sonnets, in which sadness touches softly — " And when the now is then, And when the then is now." If the " then " is in the past, we can, by fancy, make it " now ; " the " then " is clearly seen, for it is now a fact, and we can thus gain some secure experience in prognosis. I show you to-day a patient who presents an opportunity of illustrating the present by the past, and by a past in itself most instructive, such as seldom occurs to a teacher. I imagine it is almost a unique opportunity for a teacher to be able to take as a subject a patient who has been under his observation for a quarter of a century. That is the case here, and I am glad to have so unusual an opportunity in beginning again the series of our Wednesday lectures. It is nearly, if not quite, twenty-five years since this patient was under treatment in the wards of the hospital in the acute stage of his affection. Did you notice him as he came into the room ? If you did not, you should have done so. One of the habits to be acquired, and never omitted, is to observe a patient as he enters the room ; to note his aspect and his gait. If you did so, you would have seen that he seemed lame, and you may have been struck by that which must strike you now — an unusual tint of his face. Those two things are import- ant. They are, indeed, connected, but in a way that is rather curious than instructive. It is, indeed, so curious that I cannot resist the temptation of telling you the story it involves. The patient came here in 1870, with symptoms of a cere- bral tumour, of rather rapid onset for such a morbid process. The symptoms had reached to a considerable degree in A PATIENT OF LONG-STANDING. 37 about two months. The patient presented indications of a sub-chronic local cerebral lesion, with headache and optic neuritis. These two general cerebral symptoms with the onset indicate that the local process is a growth. More- over, there was a history of active syphilis ; and we know that whenever we have evidence of a local growth of rather rapid course in the subject of syphilis the probability is very great that the growth is syphilitic. They are much less if the growth is very chronic, and this point is important. The patient was treated according to the diagnosis. He was not under my care, although he was under my con- stant observation ; I was then Medical Registrar to the Hospital, and he was under the care of Dr. Hughlings Jackson, from whom it was my privilege to learn many lessons of ever-increasing value, and not the least important were connected with this case. After the patient's dis- charge from the hospital he was under my care as an out- patient, of late years only seen occasionally, chiefly for the benefit he is always ready to give to others as an illus- tration. When he first came to the hospital he was lame, as he is now, and he presented the complexion aspect you see, but in a greater degree. An inquiry into his history showed that two years previously he had been an in-patient at a general hospital, under the care of a physician then well known, who has now been dead many years. The symp- toms he then presented were those of a small syphilitic growth pressing on one side of the spinal cord, and causing effects that we now know to be very characteristic. For those symptoms he was treated with nitrate of silver, and his skin acquired the aspect which it has never lost. When he came here he had improved, and I think, for several reasons : it is exceedingly likely that after the nitrate of silver had been given for a considerable time, without other result, mercury was substituted. At any rate, the affection 3$ ARGYRIA AND SYPHILIS. of the leg ceased to increase, improved somewhat, and then became stationary, and when he came to this hospital its state was much as it is now. with due allowance for the effect of the fresh trouble which brought him to us. In connection with his case there is a little story which I cannot resist the temptation of telling you, especially since the patient here can correct me if I am wrong. Per- haps it is a little beyond the proper subject of a lecture, but I dare say. gentlemen, you will not be strict. The patient suffered from severe headache, optic neuritis, and signs of a local cerebral lesion in one cerebral hemisphere, of subacute onset. It was certainly a quickly growing tumor, and almost certainly syphilitic. Thus the case was most instructive. In 1S70. as you know, not quite so much was known of optic neuritis as is now known, and he was shown to a good many visitors. The interest, too, was not lessened bv the indications of ars^-ria — the staining of the face from nitrate of silver for a morbid process in the spinal cord similar to that which in the brain lessened with extreme rapidity under iodide of potassium. Great care and caution were taken in all that was said in his presence. I can even now remember the scrupulous circumlocutory care adopted to guard against any :ercep- tion, on his part, of what was thought about his previous treatment. But the man possesses a considerable amount of intelligence, and he picked up too much information, although he gave us no indication of the fact. The symp- toms rapidly subsided. On the morning after his discharge he paid a visit to the physician under whose care he had been, and from whom he had received the silver. He obtained an interview with the doctor. The result of that visit was, I am certain, to improve the therapeutical knowl- edge of the physician, and I have also no doubt that the result was very much to the advantage of any other patient who subsequently came under the care of that physician AN ILLUSTRATION OF ARGYRIA. 39 for a similar affection. But the immediate result was a considerable disturbance of equanimity. The patient was wise enough to content himself with thus conveying instruction. He might, I think, have gone further; but I doubt whether even a speculative lawyer would have induced him to do so, for he is, after all, a reasonable fellow. I think one cannot find very much fault with the lesson, or even, considering all things, with the way in which it was given. [Dr. Gowers here turned to the patient and asked if the account given was substantially correct. The reply was: " Yes, sir ; it's all right. I jacketed him."] I imagine that it is very likely that some of you have never seen the tint of argyria. It is less commonly met with now than formerly, because nitrate of silver is less frequently given, and when it is given, it is given with more care. It will therefore be wise for you to note very carefully the aspect of this patient. The tint is rather less than it was, but it persists, and it will persist as long as he lives. There is not now a black line at the edge of the crums ; I think it was there formerlv. We have been unable to find the old notes of the case, but all the essen- tial facts are adequately impressed upon my memory. I have only myself seen about four cases of staining from nitrate of silver. This is one. Two were in cases of epi- leptics, for which it was, as you know, once a reputed specific. It was held in very high esteem by some persons in what have been called the " pre-bromidic days." Both the patients I saw, who had been stained with nitrate of silver for epilepsy, were still patients here for the persisting disease, and therefore my own observation did not lead me to entertain a very high opinion of its value. The fourth case is instructive because it was due to the use of nitrate of silver for the good it can unquestionably do in gastric affections, especially when pain occurs before 40 ARGYRIA AND SYPHILIS. meals, that is, when it coincides with the absence of food. Although in cocaine we have an agent which seldom fails under these conditions, it is not unlikely that this use of silver will again increase. It may be then that the lesson this case gives may be again needed. A doctor, an esteemed practitioner in a suburb of London, gave his brother, who suffered much gastric pain, a "dinner pill," to be taken before food every day ; it contained some oxide of silver. A year or two after, when shooting together in Scotland, the doctor became uneasy at his brother's cya- notic aspect. He watched him closely, and at last asked him : " Do not you get short of breath as you go up hill ?" But there was no shortness of breath, and the doctor did not think anything more about it. Six months later, the tint had increased, and it suddenly flashed across the doc- tor's mind, " Why, it must be silver ; there is that pill ! " He turned to his brother and said, " Have you been taking those pills ever since I first gave them ?" " Yes," was the answer, " I have been taking them ever since, and am still." By this time his face had become deeply tinted. The line on the gums was most distinct. Although they were at once stopped, other remarkable troubles came on. There were no signs of lead poisoning, no colic, and no conceivable source of saturnism, but the patient developed wrist-drop, just like the wrist-drop of lead poisoning, and also developed the gout that is so often due to lead, and he developed the albuminuria associated with it. Silver does cause palsy in animals ; we know how many metals may cause the symmetrical extensor palsy, and I think there can be no doubt that in this case the palsy of the extensors of the arms was due to the silver, although the case, as far as I know, in that respect, is unique. The sequel of the symptom, I may add incidentally, was most illogical. The patient died two years later, but he died from cancer. With adequate mischief to terminate life by intelligible effects, he ORGANIC HEMIANESTHESIA. 41 died from something altogether different, which is an illus- tration of the limits there are to our power of inference and forecast. Let us turn to the patient before us. He left the hospital with some symptoms remaining from the disease in the brain ; these had become stationary, and they have persisted ever since. I said he had symptoms of a local cerebral lesion. These symptoms were slight left-sided weakness, left hemianaesthesia to all forms of sensation up to the mid- dle line, head, limbs, and trunk ; considerable diminution of the special senses on that side — taste and hearing, while vision was affected as hemianopia. Of smell I am not sure. If impaired, the defect did not persist, and his present re- collection is that it was normal. The hemianopia was at first complete. Leaving smell out of consideration for the moment, you know that left-side hemianaesthesia, involving the skin and the special sensories, is the characteristic of what is called hysterical hemianaesthesia, a functional condition of the existence and reality of which there can be no doubt. Nor can there be any doubt of its practical independence of the patient's will. There was the difference, however, in this condition from the hysterical form — that in the latter the affection of vision is a diminution in the whole field of vision. There is a considerable general diminution in the field of the opposite eye, and a slighter similar diminution in the field on the other side — that is, the side of the central hemisphere involved. In this case, however, there was hemianopia on the opposite side. That is the great differ- ence between this form of hemianaesthesia, that which is due to organic disease, and that due to a functional affection. Nevertheless, there is evidence to show that " crossed am- blyopia," dimness of vision, with general restriction of the field, greater on the opposite or " crossed " side to the affected hemisphere, may occur from organic disease as it 42 ARGYRIA AND SYPHILIS. does from hysteria. But the reason why an organic lesion generally causes hemianopia is this. The lesion is generally at one place, at the region which Charcot has called the " sensory crossway," at the posterior extremity of the inter- nal capsule between the optic thalamus and the end of the lenticular nucleus. The sensory fibres from the skin, from the head, and limbs run in the posterior third of the hinder limb of the capsule. The optic tract conveys impressions to that region, the fibres from the same-named half of each retina conduct- ing impressions from the opposite half of each field of vision pass to it. Thus the impressions passing to this hemisphere are those from the side on which the motor processes act on the limb. Each half of the brain receives impressions from the side on which it moves the limbs. The fibres from the optic tract that subserve vision pass into the white substance of the occipital lobe. They prob- ably have an intermediate station in the posterior extremity of the optic thalamus. But I need not now dwell on this. Whether they pass from the thalamus or directly from the tract they must pass close to the extremity of the capsule, close to the sensory fibres from the skin. Thus disease here causes hemianopia and cutaneous anaesthesia. But it also causes loss of taste and hearing. This is so well established that we are sure that the paths for these sensory impres- sions pass by this region. Moreover, there are cases on record of an affection of smell from disease of this region, but the point needs further evidence, and as this case has no definite bearing on it, I will pass it by. But I must emphasise the fact that cases of organic disease have been met with which cause symptoms resembling closely the hysterical form of hemianesthesia. The latter we must ascribe to an inhibition of the sensory structures in one hemisphere. In the cases of organic disease that cause similar symptoms, extensive disease has existed on the VISUAL DEFECTS. 43 convexity of the hemisphere, so extensive that we can- not infer more than that it is in the convexity of each hemisphere that the impressions are represented which come from all the special senses of the opposite side, in- cluding smell. In these cases there is " crossed amblyopia." But, beyond recognising the fact, you need not now con- sider the condition. Our subject is the sensory effect of disease that causes vision to suffer, as " hemianopia." Half- sight is lost with the other senses if the disease is in the tract, the sensory-crossway, or the half-vision centre in the occipital lobe. This was the combination the patient pre- sented. The association of hemianopia and impairment of the general and special senses on the same side proved that the disease was situated where all the paths are in contiguity — that is, at the place I have mentioned. Be- yond this the paths diverge, so that the combination can be produced only by disease invading the whole cortex of the convex surface, including the occipital lobe, and then there is the crossed amblyopia as well as the hemianopia. There is restriction of the field as well as half loss. We may now turn to the symptoms the patient still presents, the permanent residue of those caused by the active disease. Persisting as it has for the past quarter of a century, we may actually expect it to persist for the next quarter of a century, which is probably about as long as the patient may be expected to live. A slight defect of taste and of hear- ing is still distinct. In the limbs there is still imperfect sensation. The defect in vision that still persists is particularly in- structive. You can easily verify it for yourselves. In one pair of charts before you, you see first the fields of vision, as they were nine and a-half years ago — that is to say, about fifteen years after he came for treatment — and the others present the condition that exists at present. The two are practically identical. In each there is a loss of 44 ARGYRTA AND SYPHILIS. the left lower quadrant. At first the whole half field was lost, up to the middle line ; but the upper part recovered slowly, so as to leave the loss confined to the area you see. In it there are points that it will be instructive to you to note. The loss stops short of the fixing point. That, as you may know, is a characteristic of all forms of hemianopia, at any rate of lateral hemianopia. I show you another diagram, in which the loss is of the whole half field, in which the feature is well shown. The blind area reaches the middle line, above and also below ; but the dividing line between sight and blindness curves round the fixing point, so as to leave an area of vision. This is the rule — I believe invariable. Around the fixing point there is vision on the blind as well as the seeing side. I believe that apparent exceptions are due to imperfect observation — imperfect almost of necessity, because the area of sight around the fixing point varies, and special means are needed to ascertain it when it is small in ex- tent. The explanation of it is that from the region just around the macula lutea fibres pass by each optic tract to each hemisphere. Hence, disease of either optic tract, while causing hemianopia, does not cause loss in that small region from which the fibres go to each tract. It may have occurred to you that if fibres pass to each optic tract, disease of either should lessen the function of the whole of the region, although causing no absolute loss on either side. It is so. If you test minutely the central area of vision of a patient with hemianopia you will find that, although the region round about the fixing point is spared, the acuity of vision in it is definitely reduced. Note also another point of significance. There is but little restriction in the general field of vision. The very slight diminution in the general area of the peripheral parts of the fields is not greater than can be accounted for by a dark day, or even by an individual variation ; yet PERSISTENCE OF SYPHILITIC SYMPTOMS. 45 note how different is the extent of the remaining half fields in the other chart I show you. Notice the remark- able restriction of field, and that it is much greater in the eye on the side of the half loss — that is, the right half of each field being lost, the remaining part of the right field is much smaller than the left. I must not, however, allude to this to-day beyond asking you to remember it in connection with " crossed amblyopia." And now, gentlemen, in conclusion, I have to impress upon you one practical lesson which this patient gives us, and which he gives as an old patient. It is the persis- tence, in some degree, of the effects of syphilitic disease, and that in spite of the fact that the patient was treated thoroughly, a few weeks after the development of the symptoms — which is as promptly as most patients are treated. It was manifestly adequate, for the urgent gen- eral symptoms, headache and optic neuritis, began to lessen within a week, and in the course of a few weeks all the acute symptoms had passed away, and the local symptoms were rapidly improving, the loss of power had become trifling, the hemianesthesia gradually became partial in- stead of complete. Yet some of these symptoms have never quite passed away. The optic neuritis when he came in was considerable, but not extreme. The whole disc was obscured by a swelling of moderate prominence and considerable vascu- larity, yet the acuity of vision was perfect. The inflam- mation was not sufficient in itself to have involved the nerve fibres so as to impair sight by the process of inflam- mation in them, nor was it sufficient to produce new inflammatory products sufficient to damage the fibres by their cicatricial contraction. Those are the two ways in which sight suffers from the neuritis. In those days ophthalmic surgeons generally refused to believe that considerable neuritis could exist with perfect vision. The 46 ARGYRIA AND SYPHILIS. scepticism was chiefly dispelled by Dr. Hughlings Jackson, and I think that this patient was one of the cases by which the dissipation was effected. Fortunately the treatment was early enough to remove the neuritis before grave damage was done to the fibres. His sight has remained good so far as acuity of vision is concerned, and his optic discs now present such a perfectly normal aspect that you would not suspect they had ever been inflamed. I should advise you to examine them carefully, because it is very seldom that you have an opportunity of seeing optic discs which are known to have been the seat of inflammation, and which now present no indication of it, and especially discs that were inflamed long; ago. The practical lesson that I mentioned, and which I should like you to take away, is this : The idea is not yet extinct that all syphilitic disease will yield to treatment, and that if only the symptoms are certainly due to syphilis they can be cured. Probably you know well as regards the nervous system how erroneous that idea is. You know that a syphilitic ulcer of the skin will, and must, leave a scar which nothing can remove. It destroys the tissue of the skin, and that tissue of the skin being destroyed is never replaced by structure which is like the old skin in aspect and function. So it is with the brain. If there is absolute destruction of tissue by a syphilitic process, that tissue cannot be renewed, and is not renewed ; the symptoms dependent on its destruction will not pass away unless other parts of the brain can compensate for their loss. Remember that, as I have often said, the symptoms of a local lesion are never due directly to the syphilitic process. In true syphilitic affections, those which can be removed by iodide of potassium and by mercury, the syphilitic process is altogether outside the nerve elements themselves. These suffer secondarily, as they would from any other process of a similar general character. They suffer from compres- IMPORTANT LESSONS EMPHASISED. 47 sion by such a syphilitic gumma as this patient had, as they would from any other tumour. They suffer somewhat from the inflammation adjacent to any rapid growth, but this is usually simple inflammation with no necessary specific element. Through these processes they undergo damage and destruction, and no removal of the growth can do more than permit the recovery of those structures which are so little damaged that their recovery is possible. In the patient before you the damaged structures recovered. Destruction, of necessity, persisted, and the effects of the loss of tissue remain, where no compensation could be effected. That is the case with the half-vision centre and the fibres that carry impulses to it. Destruction of these causes destruction of the function and lasting loss. We have in the quadrantic loss of the fields of vision a proof of this fact. We have in it also evidence of the equally important fact that the effects of syphilis are often far more than that to which the term "syphilitic " can be applied. If you learn from this to look at a process with the imagination, that is as important in practice as in science ; if you learn to discern the elements on which symptoms depend, and to be cautious in your prognosis in such cases of syphilitic disease ; if you learn also the lessons the patient teaches regarding silver as well as syphilis, you will not have wasted your hour here to-day. LECTURE IV. SYPHILITIC HEMIPLEGIA. Gentlemen : — You have probably long since become con- versant with the truth that the most important part of learning is repetition. If this be true in the process of acquiring knowledge, it can hardly be otherwise in the process of imparting it. In conveying, as well as in receiv- ing, instruction it is essential that the knowledge which has to be securely retained should be impressed upon the learner as often as opportunities permit. It may not be the most attractive element in the process, but another lesson which you have doubtless by this time learnt only too well is that the attractive is seldom the most useful, or the useful the most attractive. Last week we examined an example of a rare disease and considered what is known of its nature, the process of its diagnosis, and the possibilities of its treatment. You had to hear of much that is conjectured but not proved, and of much that is still mysterious. To-day I propose to show you a patient who is suffering from a common malady whose symptoms can be traced to their source with reason- able certainty by means of the methods which must often have come under your notice, but which you will have to apply so frequently that they can scarcely be too often reconsidered. This is an advantage, not only for the sake of the repetition, but because no two cases of disease pre- sent precisely the same features in state and history ; and in every case you will meet with one or more elements International Med. Mag., October, 1893. 48 LOCATING THE HEMIPLEGIA. 49 with which in that combination you are not familiar, and only experience in applying the processes of reasoning to the facts observed will enable you to place them in their proper relation and make them yield you the guidance you require. The patient before you is suffering from hemiplegia, from paralysis of the right side. Common as hemiplegia is, one feature of this case should at once strike you as somewhat unusual. The forms of hemiplegia with which you are chiefly familiar have been in persons of advanced life, but this patient is only twenty-five years of age, and the affec- tion has only existed for two months. Probably you have seen some cases of the affection in young persons, but not many in whom it has come on at this period of life. If you note the power of movement which he possesses, you will see that he can move the upper part of the face as well on the right side as on the left, but he cannot move the lower part of the face so well. His tongue is not affected; it is protruded straight. His arm is feeble; he can move all parts of it, but cannot exert force with any part, and the movements of the hand are not quite steady. His leg is also weak, so that he can only just manage to walk alone. When you are accustomed to cases of the kind, your first thought will probably be, and, indeed, may reasonably be, What is the nature of the lesion ? and your second, What is its place? This is to put the chief elements in the diagnosis in the relative position which they occupy as regards the practical questions of prognosis and treatment as regards the interest of the patient. But the strictly logical order is to consider first where the lesion is, and, secondly, what is its nature. It is best to adopt this order in the case of dis- eases of the spinal cord, and we will adopt it now ; but in the case of diseases of the brain there is no real inconven- ience in considering first the nature of the affection. 5 50 SYPHILITIC HEMIPLEGIA. The first diagnostic indications should be familiar to you. Paralysis of the face, arm, and leg on the same side must be the result of a lesion above the middle of the pons. Here the motor path for the face leaves that for the limbs and crosses over the middle line, and in disease below this point the face escapes or is affected on the side opposite to the limbs. Disease in the upper half of the pons often in- volves the fifth nerve on the side of the lesion, — that is, the side opposite to the paralysis of the limbs. Very rarely it does not, and the palsy resembles that which is due to disease higher up the motor path ; but such exceptional cases, which are not met with in more than one case in a hundred, may be put out of consideration. Disease of the crus involves the third nerve on the side of the lesion, and opposite to the paralysis of the limbs. When these indi- cations of involvement of the cranial nerves are absent you may safely conclude that the lesion is within the cerebral hemisphere. But where? It must be in some place in which it can damage the fibres that conduct the motor im- pulses, or the centres from which the fibres proceed and in which the impulses are generated. Experience shows us that there are three regions in which lesions are prone to occur and in which they may cause this effect. One is the central ganglia, the second the white substance, the third the gray cortex. Lesions in the white substance of the hemisphere are not common. Moreover, if they are near the cortex the effects they produce are commonly very similar to those of disease which is in the cortex. If, on the other hand, they are near the central ganglia, the symptoms which they cause are like those that are produced by disease of the latter. Hence, for practical purposes, we may leave them out of consideration, and, indeed, we are compelled to do so, except in rare cases in which the diagnosis depends upon indications too complex to be included in a general LESION IN THE CENTRAL GANGLIA. 51 outline of the subject, such as I am now attempting to give you. Thus our problem is limited to the distinction of disease in the two remaining situations. I have spoken of the central ganglia, but a lesion there does not cause hemi- plegia by its effect on the ganglia themselves. Even ex- tensive destruction of their gray matter produces no paraly- sis. The hemiplegia depends upon the interference with the narrow tract of white fibres which passes between them, and which, bounding as it does the lenticular nucleus on the inner side, is spoken of as the internal capsule. When the middle of this and the part behind the middle are dis- eased, hemiplegia is produced, for the fibres here are those which conduct the voluntary impulses from the cortex from its " motor " or " central " region. This, as you well know, is the part whence the motor impulses leave the cortex, and disease in this region causes hemiplegia like that which is produced by disease of the internal capsule. How can we tell in which of these two situations the lesion is ? First, by the fact that cortical disease means damage to nerve-cells, and that damage to motor nerve- cells is apt to be attended by the spontaneous action which is manifested by convulsion, while disease of the central ganglia does not cause convulsion. Secondly, from the fact that the close approximation of the fibres for the face, arm, and leg in the internal capsule and the disassociation of the centres in the cortex make it common for all parts of one side to be affected in capsular, and for only some parts of the side to suffer in cortical, disease. But this is not an absolute rule. All parts of the motor region may be affected in the cortex, and only some parts of the motor part in the capsule. But it must be a very small lesion in the latter to produce only a partial effect, and a severe lesion in the former to produce a complete effect. Here we have the indication of a guiding distinction. You must consider the significance of the distribution of the palsy in 52 SYPHILITIC HEMIPLEGIA. connection with its extent. In this case all parts of the face, arm, and leg have suffered, but only in moderate degree. Hence we may assume that the lesion was not severe, and the distribution, therefore, is strongly in favor of the lesion being in the region of the central ganglia. The tongue, indeed, has escaped, but this is not surprising if, as we perceive is probable, the lesion is incomplete. The diagnosis is further supported by the fact that the patient did not suffer from convulsions at the onset, and has not suffered from them since. If such convulsions had occurred, or if there were absolute paralysis of the arm and no affection of the face, we should be justified in con- cluding that the disease was in the cortex. You may thus perceive the chief points that must be taken into consider- ation in this point of localisation, and you will find that they will afford trustworthy guidance in a majority of the cases which come under your notice. We may accordingly conclude with reasonable confi- dence that the symptoms are due to a lesion in the cen- tral nuclei, which has caused the hemiplegia by the damage, extensive but moderate in degree, to the fibres of the internal capsule. In scanning the brain of persons who have suffered from transient hemiplegia some time before, you may sometimes see nerve-fibres passing almost intact through an area of disease which involves the gray matter on each side of them, although the gray tint of the fibres on the surface of the layer shows that many of them have suffered structural damage, although there may not have been actual interruption. The next question is, What is the nature of the disease? This is, indeed, a double question, for we have to consider first what is the lesion which has caused the symptoms, and next on what pathological process does that lesion depend. Several features of the disease are commonly available for ascertaining the nature of the lesion, but of these one PROGNOSTIC VALUE OF MODE OF ONSET. 53 exceeds all the rest in its importance in the readiness with which it is ascertained and with which it is applied, and the security of the information which it affords. It is the mode of onset. Each morbid process develops in a certain time, which varies, it is true, but varies only between limits, and these limits constitute our guide. The onset of the symptoms is that which manifests the rate of development of the process. For practical purposes, within which the majority of the cases will be found to come, three different modes of onset are to be recognised, — the chronic, the acute, the sudden. I need not amplify the words. A chronic onset means a tumor or sclerosis, an acute onset inflammation, a sudden onset a vascular lesion. Fix, I beg you, the last axiom firmly in your minds. You may rely upon it absolutely, and you will find the value of its guid- ance, not seldom, when you are perplexed by the intricate maze of discordant indications. Of course, I am speaking only of organic disease, and of these it is always true that a sudden onset means a vascular lesion. It means rupture causing hemorrhage, or obstruction causing, first, anaemia, and, unless at once compensated by a collateral flow, necrotic softening. What was the mode of onset, what was the time occu- pied by the onset, of the symptoms from which this patient is suffering ? If you put the question to him he will tell you that they came on in the course of five or six days. Such a period is that of an " acute " onset; it is the period which suggests that the morbid process is inflammatory. But you must never be content with general statements. Always strive to obtain details, even when they seem superfluous. How important the details may be is usefully shown in this case : they change entirely the significance of the first statement. We learn, first, that three weeks before the symptoms came on he had a brief attack very much like that with which the paralysis began. He is 54 SYPHILITIC HEMIPLEGIA. fond, as many wise men have been fond, of finding recreation in the violin. One day he found, while playing, that he could not draw the bow across the strings. It seemed as if it were glued to the strings. Then he found the right arm and hand were altogether weak. But in an hour or two all this had passed away. Three weeks later the same disa- bility recurred under the same conditions, and it was accompanied by a peculiar feeling of numbness in the right arm and the right leg. When he walked he stag- gered, and so he continued until he went to bed. Next morning he found the arm and leg had become weaker. No change in them occurred during the following day, but in the next night complete loss of power came on, so that when he awoke he could not move either arm or leg. His speech was " thick," and, indeed, had been so upon the preceding day. Thus, when closely examined, we have a series of sudden attacks, the first in the day, after a sudden premonition such as had occurred three weeks before, the others in the night. The suddenness of onset during sleep we have to infer, but in most cases, when symptoms of considerable degree come on during sleep, their onset belongs to the class which we call sudden. We have fur- ther evidence of the truth of this in the fact that the first commencement was definitely, absolutely sudden, and succeeded a slight brief attack which was likewise sudden. So the details of the development of the symptoms show that the onset must be placed among those which indicate a vascular lesion. Which of the two processes had occurred? Cerebral hemorrhage is excluded by the age of the patient. It is not, indeed, excluded by the patient's age, in the abstract. Many of you may have seen cerebral hemorrhage in a person as young as he is. But you have seen it as the subject of a pathological, and not a clinical, demonstration. Note the difference implied in this dis- tinction. Hemorrhage occurs in the young, but it kills DISTINCTION BETWEEN EMBOLISM AND THROMBOSIS. 55 them. It is not survived, and the fact that this patient is very obviously living excludes hemorrhage. It is only the subject of clinical demonstration at a much later period of life, because the hemorrhage that is survived is the result of degenerative changes practically confined to the degen- erative period of life, and hemorrhage in the young is the result of disease of the larger arteries, such as permits their distention into an aneurism, the rupture of which is fatal. Vascular obstruction it must therefore be with which we have to deal in this case. Veins or arteries may be obstructed ; but the veins that are obstructed are on the surface of the brain, and we have found evidence that the lesion here is not at the surface, so we may put venous obstruction on one side. It must be arterial obstruction. This may be the result of two causes, which, widely differ- ing in origin, have the same ultimate effect. They are embolism, closure by a plug brought from a distance, and thrombosis, closure by a clot formed in situ. Do not for- get the distinction. The advice may seem superfluous, but it is curious how the idea of embolism dominates the minds of practitioners to the exclusion of thrombosis, and they call " embolism " whatever is not hemorrhage. But the distinction is quite as important as that between hem- orrhage and obstruction. An embolus must have a source. Almost invariably this source is a valve of the heart. To justify a diagnosis of embolism you must find a source, — that is, practically, you must find valvular disease of the heart, or, if the attack occurred some months ago, you must have a history of some malady, not long before the onset, known to cause endocarditis. Do not forget this also, because endocarditis, recently acute, may cause cerebral embolism, and yet may pass away so completely that at the end even of six months, not only may there be no murmur, but the walls of the heart may have so perfectly resumed their normal state 56 SYPHILITIC HEMIPLEGIA. that you cannot perceive any indication that there has been any cardiac disease. In this case no cardiac disease can be detected, and, further, no malady can be heard of that will cause endocarditis. Embolism may therefore be excluded, and we are left with thrombosis as the cause of the condition, thrombosis producing a focus of necrotic softening. But we have not yet done with the process of diagnostic analysis. Thrombosis has two causes. It is a clotting of the blood, and it may be due only to a strong tendency of the blood to clot. This, however, is very rare. It occurs in the old and gouty; it occurs in the subjects of cancer; it occurs in states of profound general weakness, and it occurs especially soon after childbirth, when the vessels of the uterus have to be closed by clot, and the tendency to coagulation may be regarded as physiological, — physiologi- cal, but, unhappily, often also pathological, as the disas- trous thrombosis in the pulmonary artery shows us too often. But here we have no evidence of such diathetic states, no history of general weakness, and, since the patient is a man, we may safely exclude the puerperal con- dition. The second cause of thrombosis to which we are thus reduced is disease of the artery at the spot, disease which induces the formation of the clot. Such disease thickens the wall, narrows the lumen of the vessel, often almost to the point of closure. It also changes the inner surface, so that this acts like a foreign body, and on it a coagulum readily forms. The effect is produced in greatest degree on the branches that come off where the wall of the main artery is diseased. Beyond the narrowed part the vessel is wider, but the narrowing permits little blood to flow through, and hence the circulation in the wider partis very feeble, and where the narrowing ends we have the change in the wall. Hence chronic disease of the walls of the DIAGNOSIS OF SPECIFIC DISEASE. 57 arteries leads to sudden occlusion by the formation of a clot, and gives rise to symptoms of sudden onset. There are only two forms of arterial disease which act upon the arteries going to the central ganglia of the brain. A third cause, traumatic arteritis, only acts upon the arte- ries of the convex surface, and it is, moreover, extremely rare. You know, probably, what the two conditions are, — atheroma and syphilitic disease. Each has the same effect; in each there is a thickening of the wall, a narrow- ing of the vessel, and especially a narrowing of the branches which the main trunk gives off at the spot, for it is generally a main trunk which is the seat of the change. But atheroma is a senile lesion. It is a disease of the old, even more emphatically than is hemorrhage, for in extreme old age it becomes the more common lesion of the two. But this disease also the patient's age excludes. By the mode of onset and the age of the patient, taken together, we thus arrive at the diagnosis that the cause of the symptoms must be an area of softening due to syphi- litic arterial disease. The time occupied by the onset, although the chief, is not the only guide. We may leave out of consideration those indications which apply only to patients in the later period of life, which we shall have another opportunity of considering. In the early period the distinction between the two common causes of hemiplegia, embolism, and thrombosis from syphilitic disease — for these two embrace certainly ninety-five per cent, of the cases of sudden onset in early adult life — is sometimes aided by the presence or absence of symptoms before the onset. Until the moment when an embolus is carried with the rushing blood into a vessel too small for it to traverse, the state of the intra- cranial structures is normal. In embolism, therefore, we have no premonitory symptoms. But it is not so in arte- rial disease. The narrowing of the arteries and the 6 58 SYPHILITIC HEMIPLEGIA. morbid process which is going on in their walls might be expected to cause symptoms constantly. It does so fre- quently, but not constantly, apparently because thrombosis occurs earlier, and with less extensive disease, in some cases than in others. One patient has headache, another has tingling or transient attacks of slight weakness in the limbs afterward paralyzed, the result of the interference with the blood-supply to the region from which it is after- wards altogether cut off. This patient had no premoni- tory symptoms in the limbs, but he had headache for a month, greater during the week immediately preceding the onset. So that this symptom yields us another indica- tion pointing to the same conclusion as that to which we have been conducted by the study of the onset and the process of successive exclusion. Loss of consciousness at the onset is chiefly important when the distinction has to be made between hemorrhage and softening. In the latter it is more often absent than present, and it was absent here. The only help it gives is that it is rather more often absent in thrombosis than in embolism, because the latter is more violently sudden, and it seems to give rise to a more definite shock to the brain. So this indication also affords confirmation of our opinion ; although it is slight it corresponds in direction with the rest, a correspondence you should never fail to notice. There are other occasional peculiarities of onset which afford a distinctly useful indication, and such is the onset of the symptoms in the patient before you. When the paralysis comes on in a series of distinct sudden attacks, each without loss of consciousness, each increasing the extent or degree of those which preceded it, we have a form of onset which is seldom, very seldom, met with ex- cept in thrombosis. The successive attacks mean the successive occlusion of branches of the diseased vessel. When hemorrhage has a deliberate onset it is gradually CONFIRMATORY EVIDENCE SOUGHT. 59 progressive from the slow forcing of blood into resistant tissues, or there are but two stages, of which the second is attended with profound coma and general paralysis, be- cause it is the expression of rupture into the ventricles of the brain. Thus, in this patient the features of the onset, and its immediate antecedents, entirely confirm the opinion to which we arrived from the comparative study of the pathological possibilities. Yet there remains another question of ultimate impor- tance which may destroy our inference, leave it unchanged, or strengthen it. It is the question whether any evidence of the assumed cause of the morbid process can be ascer- tained. That, in most cases, is the form in which the question must be put. But if no cause of any one of the possible lesions can be traced, it is necessary to consider whether the cause of that which is otherwise indicated can or cannot be excluded. If it cannot, the indication may be followed. Hence, in the case before us, having found that syphilitic disease is the probable cause of the cere- bral lesion, we ask, Has the patient had syphilis ? To this the answer is an uncompromising negative. We can ascertain no history of any primary lesion or of any second- ary manifestation of the disorder which we assume to be the cause of the morbid process which has brought him now under our observation. Some of you, and many of those who are not conversant with the progress of recent observation, would be inclined to accept the negation as a decisive disproof of our conclusion. They would be wrong. We have here a pertinent illustration of the im- portance of the absence of disproof where proof has been found wanting. Those who see many patients suffering from syphilitic affections of the nervous system, see among them some who, as this patient, can give no ac- count of any chancre or of any secondary symptoms, and among these there are some in whom a minute and care- 60 SYPHILITIC HEMIPLEGIA. ful examination reveals some conclusive indication of syphilitic disease, it may be in the tongue, in the throat, on the face, or in the eye, but it shows how small is the practical importance of the ignorance which might seem to some to be conclusive. There are others in whom no such indication can be found, but who at some future time come under our notice with distinct later symptoms un- mistakably syphilitic. Now, the existence of these two facts makes it impossible for us to accept the patient's belief that he has never had syphilis as destructive of our diagnosis. All that it does is to cause us to ask that second question, Can the patient have had syphilis ? The disease is, for practical purposes and outside the ranks of the medical profession, acquired in only one way. If there has been no exposure to the risk of contagion, we may exclude the disease and we must be wrong in our diagno- sis. If there has been such exposure to possible con- tagion we may be right, and a sufficiently extended observation compels us to regard the negative history as then without significance. It is so in this case. For five or six years the patient has been frequently exposed to the risk of contagion, and therefore we may regard our diagnosis as unaffected by the negation, and regard it as the expression of ignorance of the malady, not necessarily of its non-existence. One or two features in the symptoms and in the lesion should not be passed by. The general symptoms of hemi- plegia, as I said, we will defer, but almost every case pre- sents some special features which we may not soon meet with again. Notice that the hand, although weak, presents no complete paralysis of any movement, but notice also that the movements are a little irregular. When he tries to touch his nose with his first finger, and eyes shut, he does not carry the tip of the finger straight to the tip of the nose. Such defective co-ordination is common with mod- DIAGNOSTIC SYMPTOMS. 61 erate recovery of power, and has not, as far as we at present know, any definite significance. It is not here associated with the symptom which some patients present, an ignor- ance of the position of the hand on passive movement. Next observe that sensation is unimpaired throughout the side. From this we infer that the posterior third of the hinder limb of the internal capsule has escaped, and there- fore that the adjacent grey matter has escaped ; and we are, therefore, prepared for another negative fact which you should always ascertain and note, that there is no hemiano- pia. There is no affection of sight, and there are no mor- bid changes in the eye. Syphilitic disease of the arteries and its results never cause optic neuritis. If there is optic neuritis you may feel sure that, in addition to the arterial disease, there is a syphilitic growth in some position in which it has not caused symptoms. Lastly, look at the movements of the face, and remem- ber that in all cases of facial paralysis it is necessary to ob- serve the amount of interference with three different kinds of movement which take place in the face, especially in the lower part, which alone is involved in this case. The first is voluntary movement. When told to raise his upper lip and put his teeth together, you see that the movement is considerably less on the right side than on the left. The second is emotional movement, which is best manifested in the smile. But it does not do to tell the patient to smile. A patient who smiles to order produces only a voluntary and not a true emotional movement. To observe the latter you must produce the emotion. You may tickle a child, but you cannot well, without loss of dignity, adopt this method in the case of a grown-up person. But you may generally obtain what you need by asking the patient, as I now ask this patient, to " favor us with a graceful smile." You see that there is much less difference between the movement of the two sides than in the voluntary move- 62 SYPHILITIC HEMIPLEGIA. ment. Remember that this difference exists only when the disease is in the cerebral hemisphere, or at least above the middle of the pons. It is not seen in any case in which there is disease of the facial nerve, or its root fibres, or its nucleus. Lastly, the associated movement of the face, which occurs when the patient exerts force with some other part of the body, when he grasps with energy, for instance, may present a difference from the voluntary movement similar to that which is observed in the case of the smile. It may be far more equal on the two sides, and often no difference can be detected. We do not yet know the signifi- cance of these differences when there is disease in the hemisphere, but they show that the cause of the facial weakness is in this position, and they will probably ulti- mately prove to have a special meaning. The last point is the state of reflex action. The patient presents an in- creased knee-jerk and a foot-clonus on the right side ; these, as you know, show that there is secondary degener- ation in the pyramidal fibres. But the plantar reflex is almost absent on the right side while active on the left, and there is a distinct difference in the abdominal reflex, al- though it is less conspicuous, because, as is often the case in adults, this reflex action is not readily obtained. Such a contrast between the two forms of reflex action is com- mon in hemiplegia, and is very remarkable. It is also mysterious, because we do not yet know to what the dimi- nution of reflex action from the skin is due. It is present immediately after the occurrence of the cerebral lesion, while permanent increase in the myotatic irritability does not manifest itself until toward the end of the first week. We urgently need a series of careful observations on the position of the cerebral lesions in cases in which this dimi- nution in the superficial action is present, and their posi- tion in the cases in which it is unchanged. It is not a recondite subject, and I would commend it to your notice. ILLUSTRATION OF DISEASED ARTERIES. 63 Regarding the lesion, several points demand our notice. What is the change in the walls of the vessels which we have assumed to exist in this case ? I show you under the microscope some sections of such diseased arteries, which have been prepared by Dr. Taylor. They show very beautifully the main features of the morbid change. You can see that there is a very extensive growth of cells and fibres in both the inner and outer coat of the vessel, the separation between the two being indicated by the wavy line of the elastic lamina. You may also see that the growth has almost closed the chief vessel, and that some branches which are divided in the section are completely occluded. This disease of the arteries causes nodular swellings on the external surface, often considerable in their promi- nence. They are less translucent than the normal wall, but less opaque than the enlargement which occurs in atheroma. That is because, prone as syphilitic tumors are to undergo caseation, the tendency is slight in these arte- rial growths, and hence we have not the dense opacity which is due to fatty degeneration in the senile lesion. The change is especially common on the basilar artery and on the middle cerebrals, and in this case we can have little doubt that such disease in the left middle cerebral has oc- cluded those branches which go to the central ganglia through the anterior perforated spot. It probably has not arrested the circulation through the main trunk, and so the cortex is intact and the patient has escaped the impair- ment of speech, which is the common result of the soften- ing of the cortex, which occurs when the main trunk of this vessel is occluded. But I would call your attention to one point, which is not, I think, remarked in any text-book. If the patient has been subjected to treatment, the disease in the wall of the artery is considerably changed. The swelling dimin- 64 SYPHILITIC HEMIPLEGIA. ishes and may be scarcely recognizable, but the wall is still a little thicker and a little more opaque than normal. The aspect resembles that of slight atheroma far more closely than does the syphilitic disease in its recent and unaltered state. You must therefore be prepared for this appearance in the case of many patients who have been under the care of those who recognized the cause of the symptoms. Often you find such old disease, altered in the manner I have described, in one artery, and in another artery recent change of characteristic aspect. Not long since I met with a very instructive example of this, most instructive, also, in its clinical manifestations. The patient was a doctor, thirty- five years of age, of reticent habits, who had suffered from a sudden brief attack of loss of speech. Six months before, he was attacked with sudden symptoms of an extremely grave character. Rising in the morning well, although there was some reason to believe that he had suffered from headache for a few weeks, he was suddenly attacked with difficulty of articulation and with mental stupor, deepening in a tew hours almost to coma, and attended by paralysis of the right arm, right leg, and lower part of the face, com- plete paralysis of the muscles of the left eyeball, and almost complete of those of the right, with immobility of the pupils, the right being small and the left dilated. So he continued for two or three days, and then, with rising tem- perature, he died. The symptoms pointed conclusively to a lesion of the oculo-motor nuclei beneath the corpora quadrigemina, greater on the left side and extending through to the motor tract, a lesion of sudden onset, and therefore vascular, and, from the reasons we have consid- ered in our review of the case before us, certainly due to vascular occlusion, to arterial disease, and almost certainly to syphilitic disease of the basilar artery and posterior cere- brals. The attack six months before might reasonably be ascribed to similar disease in the middle cerebral of the POST-MORTEM CONFIRMATION. 65 left side, disease which, on the assumption that he was conscious of previous syphilis, would naturally lead him to take iodide of potassium, the influence of which would be to prevent, for the time, further effects of the disease. He could easily have obtained it, and would have been un- likely to communicate his apprehensions or self-treatment to any person. The omission of the drug, or possibly, even, as we shall presently see, its too long continuance, would have permitted a fresh development of arterial dis- ease in the basilar and posterior cerebrals, with resulting and fatal thrombosis within them. Such was the opinion expressed on the first examination of the patient, the day after the onset, and the subsequent condition affoided me no grounds for changing the diagnosis. The post-mortem examination showed its correctness in a degree which would be remarkable, were it not that the elements of the diagnosis were simple, and only required trust in the pro- cess of reasoning to induce confidence in the conclusion. There was thrombosis of the front of the basilar, extending into the posterior cerebrals, further into the left, and on this there was a characteristic nodule of syphilitic disease, while slighter thickening, narrowing the cavity of the ves- sels, was seen on the other posterior cerebral and on the front of the basilar. In the middle cerebral, within the fissure of Sylvius, was an area of altered wall of greater opacity and thickness, but without much prominence, pre- cisely the aspect I have mentioned to you as left by syphi- litic disease which has been removed by treatment, and justifying the opinion that I have mentioned. What is the relation of the vascular lesion to the consti- tutional affection ? Although not yet demonstrated, it is impossible to doubt that syphilis, like every disease that is transmitted by inoculation, that has a period of incubation and is first manifested by exanthematous and allied symp- toms, depends on a specific organism which multiplies in 66 SYPHILITIC HEMIPLEGIA. the system and produces germs. As far as we can read the meaning of the course of this malady, it would seem that the various morbid processes which it causes are the result of the presence of developed organisms, abundant in the blood in the exanthematous stage, and leaving germs which rest in some tissue until a time comes when they develop and give rise to a process of tissue-growth such as that which we have been considering in the walls of the arteries. Accumulated and accumulating experience leads us to the conclusion that the remedies employed for this disease — mercury in the first rank and iodide of potassium in the second — destroy the developed and developing organisms, and remove, or promote the removal, of the tissue-growth which the organisms have caused. Hence the manifesta- tions of the disease in the early florid stage, or the later occasional effects, may be with certainty swept away. But the agents seem to have no influence on the germs which are deposited in the system and are not yet in the process of development. You are doubtless familiar with the fact that the germs of bacterial organisms have far greater power of resistance than the organisms themselves ; that they resist, for instance, temperatures which are fatal to the latter. Hence the removal of the symptoms at any stage of the disorder does not free the patient from the liability to the occurrence of future symptoms, the result of the development of the germs that have been untouched by the therapeutical agent that has been employed. The most thorough treatment at any stage of the disease does not prevent its future manifestations, and the freedom from these is as frequent in those who are not so treated as in those who are. It depends on the disease, on the amount of organized material in the system, and on no other ele- ment. It is in this sense that I have expressed my con- viction that syphilis as a disease is not curable. I have UNFAVORABLE PROGNOSIS. 67 been taken' severely to task for the assertion, but it has received no dissent from those who know most of the malady, and whose opinions deserve chief attention. I believe that the statement is the assertion of an incontro- vertible truth. All that we have hitherto considered is but the path to the problems that are of practical moment, — those that concern the interest of the patient, — and, therefore, to us as practitioners are of paramount importance. These are the prognosis and the treatment, the attempt to forecast the future, and the attempt to remove or lessen the symptoms and to prevent their return. Strive, gentlemen, when you consider the prognosis of such affections, to form a mental picture of the morbid process. Note its elements, their mutual relation, and their relation to the symptoms. You may perhaps be surprised to perceive, when you do this, that in all cases of true syphilitic lesions of the nerve-centres the symptoms do not directly depend, in every case, upon the syphilitic process itself. This is conspicuous in the case of the disease with which we have to deal to-day. The symptoms depend on the necrotic softening of the brain, which is precisely the same as that which would be caused by any other form of arterial closure, by embolism, or by thrombosis due to any other mechanism. It is the result of the arrest of the blood-supply, and the clot which finally stops this is also common and not specific. It is only when we reach the cause of the narrowing of the artery that we reach the actual syphilitic element in the process. But you cannot fail to perceive, further, from these facts, that the course of the symptoms must be altogether independent of the course of the arterial disease which led to the obstruction. Remove the disease in the artery ; it still remains imper- vious, and ultimately becomes a mere fibrous thread ; nothing can restore its cavity and the destroyed brain-tissue 68 SYPHILITIC HEMIPLEGIA. also nothing can renew. Such a patient will improve, partly by the recovery of the least damaged structures, partly and chiefly by the compensation which happily the other hemisphere is able to effect in the case of all move- ments which are related to both sides of the brain, those in the upper part of the arm and face, and the chief move- ments of the leg. But the future course of the paralysis, in every case of the kind, will be the same as if the obstruction were the result of embolism, and the fact must be frankly recognized. Do not, as is so often done, assure a patient with hemiplegia of a year's duration that he will recover, because his malady was the result of a process which you know can be removed, and which most likely has already been removed. Is treatment therefore useless ? By no means. Its importance is great, and the greater the sooner after the onset you can apply it, and it is greatest of all before the onset, when the shadow of the coming disaster is thrown by premonitory symptoms, and when the substance may very often be averted by the prompt recognition and energetic treatment of the morbid process. I am certain that in cases which have come under my notice this result has been achieved, and palsy, perhaps life-long, and some- times even death, have been averted. In this patient we have an illustration of the opportunity which many others present: when he first felt the disability in the arm, the merest suspicion of its possible cause would have justified treatment which would certainly have averted the sequel. After the closure has done its work, and the softening has produced its symptoms, treatment can do nothing for that which is, but it may still do much for that which may be. We cannot tell how many other arteries are diseased, how many other branches are in imminent danger of being closed, and it is essential, therefore, at once to give the patient iodide of potassium, seven or ten grains three times TREATMENT. 69 a day, and even fifteen grains for the first few days, with, if you like, mercurial inunctions. I believe that this dose of iodide is adequate, and will do all that can be done in the course of four or six weeks. It is better not to give larger doses, because coagulation of the blood plays a part in the mechanism of destruction, and large doses of iodide, as the treatment of aneurism has taught us, increase the tendency of the blood to clot. Of even more importance is it not to continue the iodide beyond six weeks or, at most, two months. In that time, as visible processes show us, all the specific element in the process is removed. If recovery is not complete, it is because the simple elements may take yet a longer time to pass away, or may be of such deeper character that they must persist. If you go on with iodide for four or six months, you may find the process that was at first arrested regains activity, and the very same syphilitic process may actually kill the patient, in spite of the continuance of the treatment which at first arrested it. Apparently the organisms become acclimatized to the presence of the agent, are able to resist it, and thrive in spite of it. Analogous phenomena are known, as I have hinted in the case of similar organisms and the effect of high temperatures. But while you should not continue the treatment, you should always resume it. After three or four months' cessation it is as effective as at first. I consider that every patient who has had syphilis should for at least eight years from the primary disease, or five from the last manifestation of it, take a course of iodide for three weeks twice a year. In this way developing organ- isms which have not yet caused tissue-changes sufficient to produce symptoms may be destroyed, and in many cases it is certain that this measure would save the patient from grave disease. We cannot see the result, but in this it is but on the level of all the other forms of preventive medi- 70 SYPHILITIC HEMIPLEGIA. cine, the branch of our professional work to which we may look up with the highest pride, and go on with, confident in its unseen achievements, despite the fact that the recipi- ents of its greatest blessings are all-unconscious of them, as they are always of the health they enjoy until they lose it, and of the air they breathe until the last breath may bring- to their mind the flash of revelation, all too late. LECTURE V. BULBAR PARALYSIS. Gentlemen : — I wish to-day to take the opportunity that this case affords me of showing you the symptoms of the malady generally known as " bulbar paralysis." It was formerly called by a descriptive designation, given to it by Duchenne, " labio-glosso-laryngeal paralysis." The symp- toms in this case are so complete in degree that they may impress upon your memory the chief features of this malady. You are not likely to forget them after you have observed a well-marked case. Let us first briefly note what they are. (i) You will observe, when I tell the patient to move his lips and mouth, that he has but little power of moving the lower part of the face; although he can put in perfect action the muscles of the forehead and eyelids. He can only slightly raise the upper lip ; he cannot narrow the orifice of the mouth ; although he can just approximate the lips, it is imperfectly; a slight chink remains between them, and they cannot be kept together. (2) His tongue is almost completely para- lysed ; when he tries to put it out, you see that he cannot get the tip beyond his teeth. Moreover, if you look at the surface of the tongue as it lies in his mouth, you will see that it is uneven ; irregular depressions upon it indi- cate the atrophy of the muscular tissue of which it chiefly consists. (3) His palate can be raised, but if he tries to swallow, liquids come back through his nose, evidence that the palate does not shut off, as it should, the cavity of Clinical Journal, May 2, 1894. 71 72 BULBAR PARALYSIS. the nose, on account of feebleness of the palatine muscles. But semi-solid food is swallowed well ; it does not regur- gitate into the nose, and when it has once reached the pha- rynx it is passed down into the oesophagus without diffi- culty. This is proof that the pharyngeal muscles are not paralyzed. (4) It is very different with the muscles of the larynx. When he tries to phonate he only succeeds in making one uniform vowel sound, not altered by any changes of intonation. When he tries to cough, you can readily perceive, by the sound, that he does not close his glottis ; there is a rush of air through the larynx, but there is no true cough, for which, you know, the glottis needs to be first closed and then suddenly opened. The paralysis which is thus indicated can be seen by the laryngoscope. Dr. Semon has been good enough to examine his larynx, and finds that the left vocal cord is near the middle line, and cannot be moved away from it, but the right is habit- ually some distance from the middle line, so as to leave a space of about 4 mm. between the two. When he tries to bring the cords together, the right moves toward the left, but does not reach it. In the movement, both cords pre- sent a peculiar tremor. The left cord thus presents abduc- tor paralysis, while in the right adduction is most affected. It is fortunate for him that there is not loss of abduction on both sides. If there were, he could probably cough and utter vocal sounds much better, but his life would be in constant danger, because, as you doubtless know, when the cords cannot be separated, the slightest catarrhal swell- ing suffices to close up the narrow space that remains, and, unless an opening is made into the larynx below the glottis, breathing becomes impossible, and the inevitable result ensues. So this man presents paralysis of his larynx, palate, tongue, and lips, together with the lower part of the face. The malady has come on gradually, in the course of the RELATION OF SYMPTOMS TO FUNCTION. 73 last two years, and the first part to suffer was that which is now most affected, the larynx. He first found that he could not sing ; then that he could not speak well, from weakening of the tongue ; gradually the other symptoms were added. Thus, it is an example of slow gradual bulbar paralysis, a malady which we know depends upon a slow degeneration of the structures of the medulla oblongata concerned in the movements for articulation. Let us glance briefly at the parts that are affected. In doing so I would ask you to notice how remarkably the affection corresponds to their functional use. Those are paralysed which are con- cerned in articulate speech. Some of these are also con- cerned in deglutition, and hence swallowing is to some extent affected, but the fact that the pharnyx is not para- lysed is proof of the manner in which the functional limita- tion of the affection has spared the special act of swallow- ing. We shall presently see the reason for paying such close attention to the relation of the symptoms to function. It is a point of great importance because it is of great signifi- cance, and you should always ascertain how far it is to be traced. It is one of the chief features to be attended to in each case that comes under your notice. This bulbar paralysis is generally a disease of the nuclei of the nerves which convey motor power to the muscles affected, and when such, it is a disease of the medulla oblongata analogous to the degeneration of the motor nerve cells of the spinal cord which gives rise to the symp- toms of "progressive muscular atrophy." In the medulla, you will remember, the regular arrangement of nerve cells, root-fibres, and nerve-roots, which obtains in the spin il cord, gives place to a very irregular arrangement, in which the fibres which subserve a certain function are, for the most part, gathered together in a single nerve, and rise from a defined group of cells, the nucleus of the nerve. The irregularity is indeed far greater in the case of the 7 74 BULBAR PARALYSIS. sensory than in that of the motor fibres, for almost all the sensory fibres leave the medulla as a single nerve, the fifth, which represents the posterior roots of all the motor cranial nerves. But the irregularity is chiefly in the way it leaves the brain, for within the pons the fibres of the nerve at once separate ; some ascend and some descend, and all seem to end in sensory nerve cells at about the level of the motor cells to which they are related, as do the sensory fibres of the spinal nerves. But the malady we are now considering is one of motion, not of sensation, and so we may dismiss the sensory fibres of the fifth nerve from further consideration, and attend only to the arrangement of the motor nerves and nuclei, whose function is so con- spicuously impaired. As we pass up the cervical region of the cord, before we reach the medulla, we have, as it were, an intimation of the coming irregularity. The nerve-fibres for the muscles of the neck in part arise, and leave the cord, as do the motor fibres lower down. But some of them, arising from the same anterior grey matter, pass outwards instead of forwards, and leave the side of the cord, instead of the front, to form by their junction a nerve which ascends to the medulla and passes within the foramen magnum as if it were trying to attain the dignity of a cranial nerve. It does actually join one of the bulbar nerves for a short dis- tance, and so this is called the " spinal accessory," but its fibres have to separate from the bulbar nerve, and descend to the neck, where, as you will remember, it supplies the sternomastoid and the trapezius, muscles that serve to rotate the head. I mention this nerve thus specially, because the bulbar nerve, which its fibres join for a short distance, particularly concerns us. The grey matter from which it comes would be continuous with that from the neck muscles, had not its relation to the other parts been deranged by the fibres of the anterior pyramids of the DERANGEMENT OF MOTOR NERVES. 75 medulla, which here cross the middle line and cut up the anterior cornua. The cells for the spinal accessory in the medulla are the lowest of the bulbar groups. Just as the spinal part of this nerve goes to the external muscles of the neck, the bulbar part goes chiefly to the internal mus- cles of the neck, the muscles of the larynx. But you per- ceive that there is a great difference between the two parts in source, in destination, and in function, and this difference is reproduced in disease. In the patient before you the outer spinal part of the nerve is not paralysed ; the inner, bulbar part, as we have seen, is almost completely para- lysed ; the cells of its nucleus must be extensively diseased. If we pass upwards still higher, in observing the parts affected, we come to the palate ; this is supplied by the same nerve, the spinal accessory, but probably by fibres that arise from its highest cells. The supply of the palate by this nerve has only lately been proved, but it has been proved beyond the possibility of doubt, and the fact is of much interest in connection with such a case as this. Next, after the affection of the palate, we pass to that of the tongue, which is so severe. The hypoglossal nucleus lies parallel to that of the inner part of the spinal accessory, at first below it, afterwards to the inner side of it. Not only are the nerve cells for these three parts near together, but so also are their nerve fibres, as they leave the medulla, and the tongue, palate, and vocal cord are often paralysed together by a morbid process at the surface, damaging the nerve roots. How closely structural relations reproduce those of func- tion, how, indeed, they determine the latter, is again illus- trated by the last part of the combined paralysis which we have to consider, that of the lower part of the face, and es- pecially the lips. The chief nucleus of the facial nerve is only a little above the upper extremity of the hypoglossal. The nerve leaves the pons above the level of its chief nu- 76 BULBAR PARALYSIS. cleus, to which its fibres pass downwards by a somewhat circuitous course. But the relations of the fibres of the nerve which subserve its particular functional relations are still, in the main, unknown to us. The muscles supplied by it have a twofold function (i) in connection with articula- tion, in which the lower functional muscles are chiefly con- cerned, and, in particular, those of the lips. These are the muscles which are so much affected in the case before you. (2) The muscle which closes the eyelids is concerned, in its function, with the muscles of the eyes, and this is also true to some extent of the muscles of the forehead, for when we look up the frontales raise the eyebrows, and when the eye has to be protected by a strong contraction of the orbicu- laries, the corrugator assists. (3) All the muscles of the face are concerned in the expression of emotion, the upper as well as the lower muscles of the face. It is only the first of these three functional relations which concerns us now. It is certain that there must be a close connection be- tween the fibres or cells which have to do with the muscles about the mouth, and with those for the tongue, but the anatomical connection has not yet been traced. That there is such an anatomical connection, we cannot doubt, nor can we doubt that it is especially diseased in such a case as this. Apart from the evidence of the relation afforded by disease, I may call your attention to a curious proof that is to be obtained, even under normal conditions, of the closeness of the relation between the tongue and the lips, and also be- tween the fibres or cells of the facial and hypoglossal nerves which innervate the parts. Try to narrow your tongue ; you will find you cannot do so without, at the same time, narrowing the opening of the mouth. You cannot contract the transverse fibres of the tongue without also contracting the transverse fibres which run in the lips. If you try to do either of these you will find that you do the other also. The experiment succeeds best, or at any rate we are most CORRESPONDENCE OF PARALYSIS TO FUNCTION. 77 conscious of it, if the tip of the tongue is first placed be- tween the lips. Try it now, and your sensations will speed- ily convince you of the truth of the statement. Thus the symptoms are dependent on a paralysis of the muscles supplied by part of the facial, part of the spinal accessory, and by the hypoglossal nerves. The paralysis corresponds to function in its distribution, and has been gradual in development. These two features are always evidence of the degenerative nature of the process on which the paralysis depends. But it is important to note that there are two forms of this chronic bulbar paralysis. I have told you that these nerves and their nuclei, although situated in the medulla, correspond to the motor nerves and motor grey matter of the spinal cord. The same correspondence is to be traced in the chronic palsies which affect the two parts. In the limbs we may have the same two forms of paralysis. We may have the slow paralysis confined to the wasting muscles and dependent on the degeneration of the grey matter of the spinal cord from which the motor fibres proceed. We may also have a slow paralysis without wasting, but with excess of the muscle-reflex action, and this, as you prob- ably know, depends on a slow degeneration of the fibres which conduct the voluntary impulse from the brain to the spinal grey matter. In some cases of chronic bulbar par- alysis we have conspicuous wasting of the parts paralysed, and then we always find degeneration of the cells of the nuclei from which the fibres proceed, especially conspicuous in the cells of the hypoglossal nuclei. This form is often associated with the corresponding muscular atrophy in the parts supplied by the spinal nerves, especially in the arms. In such a case, however, we have no wasting, but a slow paralysis similar to that which I have described as met with in the limbs. Although the precise lesion in this case has not yet been made out there can be little doubt that it de- 78 BULBAR PARALYSIS. pends on a degeneration of the fibres which convey the voluntary impressions from the brain to the nuclei. These pass with the fibres for the limbs until a short distance above the nuclei, and then they pass across the middle line to reach the nerve cells for which they are destined. The fibres for the limbs cross the middle line chiefly at the "decussation of the pyramids." The only explanation of such cases that we can give is that, in them, these upper fibres suffer just as do those in the spinal cord in primary lateral sclerosis. In the spinal cord the degeneration probably begins at the lower ex- tremity of the fibres and extends upward, precisely as it does in the peripheral nerves in some forms of multiple neuritis, and we may reasonably assume that the affection of the similar fibres for the nerve-nuclei of the medulla has a corresponding course. If so, we are able to understand the strict correspondence with function which may be often discerned in such cases, as well as in those in which the nuclei themselves suffer and the muscles waste. Even in the case of the degeneration of the peripheral nerves, the correspondence with function is often remarkably close. But in both forms of bulbar paralysis the degree of corre- spondence with function varies. In the patient before you it is strict; in the case from which I am about to show you a section, not only the muscles in articulation, but also those in deglutition, were affected ; the pharynx was para- lysed as well as the larynx. We are not able, at present, to attach any special significance to this difference ; it is possible that some day we may find in it an important indi- cation of the cause of the disease. These then are the two forms of chronic bulbar paralysis. Because the nuclei that are affected are those for the lower cranial nerves, it is sometimes called " inferior nuclear paralysis;" the group of nuclei in the upper part of the mid-brain, those which innervate the eye muscles, are some- VARIOUS FORMS OF THE DISEASE. 79 times affected in a similar manner in what is called " supe- rior nuclear paralysis." It is indeed "superior" only as regards the chief nucleus affected ; in this we have an inter- esting instance of the way in which function, rather than locality, determines the grouping of degeneration. The nucleus of the sixth nerve, you will remember, is quite low in the pons ; so low, indeed, that the facial nerve, which is affected in the inferior palsy, actually winds round it, and yet this nucleus is involved with that of the third nerve in the superior nuclear palsy giving rise to the affection which has also been termed " ophthalmoplegia externa." In other rare cases both sets of nuclei suffer, generally, however, in- completely. Some of you may remember a woman who has attended here, who has a considerable degree of superior nuclear paralysis and a slight degree of inferior paralysis. I shall have to allude to her case again, because it is remark- able in being a sequel to diphtheria. The two forms of chronic disease which I have described do not include all the cases of bulbar paralysis. There is also an acute form, or rather, I should say, a sudden form, which depends upon the closure of some of the arteries which supply this part with blood. The closure is the result of disease of the main trunk from which the small arteries spring ; thickening of the wall necessarily entails the narrowing of the orifice of the branches that arise there. Most of the blood comes from branches which pass near the middle line, either from the vertebral or from the basilar artery. The resulting paralysis is generally much more irregular than in the degenerative chronic variety, as we might expect it to be, since the effect of vascular disease is always to cause a more or. less random lesion. When one vertebral is diseased, and the branch which is closed comes from it, the affection is chiefly one-sided. Now and then, however, even when thus produced, it is bilateral, and the explanation of this, at first puzzling, fact is to be found in 80 BULBAR PARALYSIS. the frequent disparity in size of the two vertebrals. This disparity may be such that the nerve nuclei near the middle line, which are those especially diseased, are supplied with blood on both sides by the branches which come from the one large vertebral artery. Then, of necessity, disease of that artery and closure of the branch cause symptoms which are bilateral and symmetrical. I will show you in a moment an example of this instance of bulbar paralysis. There is yet a fourth variety to be considered, which is, however, so rare that it may be dismissed with a mere men- tion. Very, very rarely these nuclei seem to be the seat of a true inflammation, and we then have a bulbar paralysis which is not sudden but acute in onset, as the effects of all inflammations are. Such a lesion is extremely rare, but it is a little less so in the case of the superior group of nuclei. It is important, however, to mention it, because it has been known to supervene on a chronic degeneration, and to be an immediate cause of grave danger. Indeed, if the list is to be made complete it is necessary to mention that myste- rious cases have been met with in which the characteristic symptoms of bulbar paralysis have existed and have caused death, although no morbid state could be discovered either in the medulla or in the nerves. We can only explain such a fact by supposing that there are grave alterations in the nutrition of the nerve elements, sufficient to abolish their function, which are at present not within the range of our means of observation. We know that this is the case in chorea and also in paralysis agitans ; although in these affections the changes do not abolish function, it is quite conceivable that, in the almost inconceivable complexity of the molecular nutrition of the nerve cells, alterations may occur which do arrest function, and which yet entirely elude our present means of detecting them. But it is also possible that in some of these cases there may be a degen- eration of the fibres which conduct the motor impulse from A LAW OF PATHOLOGICAL SIGNIFICANCE. 81 the brain to the nuclei, and are, in a sense, homologous with the fibres that pass to the grey matter of the spinal cord in the pyramidal tracts. We have reason to believe that, in the latter, the process of degeneration only begins at their lower extremities, as it does at the lower extremities of the peripheral nerves in most forms of multiple neuritis. In such cases degeneration, chronic or acute, may give rise to symptoms of considerable severity, although confined to the terminations of the fibres, and thus the lesion may escape detection. Let us now consider in greater detail the facts of the malady which is before us, the chronic degenerative bulbar paralysis. We know very little of its cause. It is most frequent in the old, but it occasionally attacks the middle- aged, as in this instance, and now and then, for some reason or other, it occurs in those who are still children. .Some day we may know the reason of this. At present all we can do is to recognise what may be termed the outlines of the general causation of such affections. These diseases are degenerative. A short time ago this statement would have taught us nothing. But we have reason, of late, to recognise in most degenerations the evidence of the pre- vious action of a toxic influence. Note this important law, that the affection of the nerve elements related to function, when acute, indicates the action of a toxic influence ; when chronic, a degenerative process. It is a law of important practical application, but it has also an important pathologi- cal significance. The two things are not entirely separated. The toxic influence which has no action on the nerve ele- ments may yet leave behind it some effect, possibly some chemical product, which fixes itself on the nerve elements, in consequence of which these nerve elements, at some future time, undergo degeneration. This has not been proved to be true in the case of bulbar paralysis, but it is very probable that it will be proved, at any rate in the case 8 82 BULBAR PARALYSIS. of the forms which occur comparatively early in life, and cannot be attributed to merely senile changes. We see this connection between an early toxic blood state, and a late degenerative process, conspicuously in the case of syphilis and tabes. The same relation is to be traced in the case of the superior nuclear paralysis, the degeneration of the nuclei for the motor nerves for the eyeball of which I spoke just now. This may be asso- ciated with tabes, and it may also occur as a late sequel of syphilis apart from tabes, a fact of considerable signifi- cance. But we cannot trace its relation in the case of the inferior nuclear palsy, bulbar paralysis, which especially occupies our attention to-day. But some of you may have seen a patient who attends here occasionally, and who suf- fers from the double affection, from considerable paralysis of the ocular muscles, and from slight bulbar paralysis, in whom these symptoms have slowly developed as sequel to diphtheria. We must, however, separate the rare cases that are met with before or soon after the completion of development. Allied facts suggest* that the cause of these is different. They may be analogous to the mysterious " hereditary ataxy," " Friedreich's disease," and depend on some con- genital defect of vital endurance in the structures concerned. But to pursue this would take us too far to-day. The pos- sible relation to toxic influences is a subject of great importance, because at present we have little power of checking the developed disease, but if we can discover such a cause, the discovery may bring with it some means of prevention or arrest, of which at present we have no indi- cation. One other fact deserves mention as bearing upon this matter, and that is, that I have once known charac- teristic bulbar paralysis to follow ordinary lead poisoning. The patient was in the late period of life, and so may have been disposed by mere senility to a degenerative process, GENERAL LAWS FOR DIAGNOSIS. 83 but this chemical poison is known to be capable of giving rise to a similar process in the motor grey matter of the spinal cord, and it must be regarded as a very probable cause of the bulbar paralysis in the case in which the sequence was observed. The general diagnosis of bulbar paralysis resolves itself into certain problems, or rather certain general laws. First, it is practically only the sudden forms in which a distinction has to be discerned from " pseudo-bulbar paralysis" — a condition which I have not been able to describe to-day. It is the effect of a lesion in each cere- bral hemisphere, so situated that the second prevents the compensation for the first which is commonly effected. The only chronic process which may simulate the bulbar degeneration is the slow compression of the medulla, or its slow damage by growths inside or outside of it, and the symptoms then produced are always irregular, and com- bined with other, generally more obtrusive, disturbances, which sufficiently indicate the actual nature of the case. But the diagnostic difficulties presented by the sudden form are numerous. The double cerebral lesion generally declares itself by the closeness of two pronounced attacks of hemiplegia, first on one side and then on the other, the first of which has the common features of cerebral hemi- plegia in the affection of the tongue and face on the same side as the limbs. It is not uncommon for sudden bulbar paralysis to come on in two or three stages by the suc- cessive closure of different arterial branches, but in these the paralysis of the tongue and lower part of the face is much more irregular than it is in the case of the cerebral lesions. The distribution of the symptoms in these cases is indeed so irregular as to make it almost impossible to give any general description of them, and this very irregu- larity constitutes their most important distinguishing feat- ure. In the cases in which the symptoms are not irregular, 84 BULBAR PARALYSIS. their symmetry is generally conspicuous from the first, for the reason that I have explained, — because the nuclei on both sides receive their blood from the branches of a vessel on one side. But diagnosis of the precise nature of the sudden vascular lesion which causes the symptoms is often a difficult matter. When there is syphilitic disease of the whole of the vessel from which the nutritive branches pro- ceed, the recognition of the fact is indeed seldom difficult. It is determined by the same evidence which enables us to recognise such disease in other arteries, and therefore need not detain us now. Embolism we may also dismiss as rare in this region, and associated with a distinct source of the obstructing plug. But the point of chief difficult}', and also of chief importance, is the determination of the question whether a senile lesion is hemorrhage, or softening from atheromatous thrombosis. The latter process is the com- mon one here, and it may be safely assumed when the symptoms come on in several distinct attacks, as was the case in the patient before us. But the difficulty is rendered greater by the fact that a good many cases which have been described as hemorrhage have really been cases of arterial occlusion with secondary extravasations in the affected region. The distinction of such a lesion, in its early stage, from a primary hemorrhage, is not easy when the affected areas are small. A hemorrhage which is suf- ficiently extensive to cause considerable paralysis would certainly give rise to severe initial symptoms, and if these are absent it is justifiable to assume atheromatous throm- bosis rather than the rupture of an artery. The prognosis and course of the sudden and the chronic forms of necessity present a considerable difference. The chronic form, at whatever age it occurs, has a progressive tendency, and as long as the symptoms continue to increase serious anxiety cannot but be felt regarding the issue. Many cases of this description end fatally in one or two ELEMENTS OF PROGNOSIS. 85 years from the onset. But in this disease, as in analogous progressive muscular atrophy, there is a marked tendency for the progress of the lesion to undergo arrest after it has reached a considerable degree of intensity, and has produced so much damage that the state in which the patient is left is pitiable. It seems to be the case in the patient before you, in whom the malady seems no longer to be increasing. The prognosis is certainly worse in a case in which the dis- ease of the medulla supervenes on degeneration in the spinal cord, and bulbar paralysis is thus added to pre-existing muscular atrophy. It is worse, moreover, in the cases in which the symptoms are extensive and involve the parts concerned in swallowing as well as in articulation, the pharynx as well as the tongue and lips, than in the cases in which, as in this patient, only the structures for articulation are each affected. But of necessity the prognosis may have to be modified by the state of the larynx, and by the ten- dency there may be for particles of food to pass into the glottis and trachea. The chronic inflammation of the lungs, which is thus produced, is one of the most common causes of death, and when this source of danger exists the pros- pect of life being long preserved is always considerably les- sened. From the larynx the chief risk is that which arises from the occasional paralysis of the abductors of the vocal cords on each side. In this patient, fortunately, such par- alysis exists on one side only, and so all its characteristic indications are absent, with them the danger which bilat- eral paralysis always involves. When both cords are close together — so close together that the inrush of air inbreath- ing gives rise to inspiratory stridor — the danger of suffoca- tion from slight catarrhal swelling of the cords is, as I have said, extreme, and adds much to the gravity of the case. In the acute form, if the onset is survived, some degree of improvement maybe confidently anticipated, and it may be expected to go on for a considerable time. The chief 86 BULBAR PARALYSIS. danger arises from the existence of conditions favoring other attacks ; it varies according to the probable amount of disease in this and other cerebral arteries, and is also in- fluenced by the extent to which other influences have co- operated in producing the obstruction. It is governed by the same rules as similar softening in other parts of the brain, and I need not advert to it here. The danger of later extension is never entirely absent in atheroma, and even in the case of syphilitic disease it may continue for some time. The cicatricial process in the wall of the diseased artery, by which the disease is removed under the influence of treat- ment, may involve the formation of fibrous tissue in the wall, which goes on contracting, and this may entail the closure of some other small branch, months after the occur- rence of the primary lesion, and after the disease of the wall has yielded to the influence of the drugs which are administered. Thus in all cases of sudden bulbar paralysis, the prognosis should be tinged with caution, and sometimes, even after the onset is survived, with distinct concern. The rare cases of acute inflammation, in which the symp- toms develop in the course of a few weeks, must at present be regarded as very serious. Ignorant of its causes, we are without the means of arresting the morbid process, which, in most of the few cases hitherto observed, has gone on to a fatal termination. But the probability that the process is due to a toxic influence is very great, and here also we may hope that the future will soon bring that which the present does not afford, and that we shall not have long to wait before we obtain the means of arresting the disease, and of doing so before the change in the nerve elements has pro- ceeded to a degree that precludes their recovery. You will have gathered from what I have said that the treatment of bulbar paralysis is one of the gloomy regions of the therapeutics of the nervous system. It is a subject on which there is, unfortunately, only too little to say. TREATMENT. 87 That of the sudden form does not differ in any material point from the treatment of similar vascular lesions in other parts of the brain. When the sudden paralysis is followed by loss of faradic irritability, with preservation of voltaic irritability, that is, the reaction of degeneration, it is desir- able for a time to keep up the irritability of the muscular fibres by the application of voltaism in any part to which electricity can be applied, in the hope that the nuclei from which the degenerated nerves proceed may be only dam- aged and not destroyed, and that recovery, with regenera- tion of the nerves, may ultimately occur. In such a case we follow, therefore, the same rule which would guide us in the treatment of an acute lesion of the motor grey matter of the spinal cord, and endeavour to keep the muscles in as good a condition as possible, in case the motor impulses should ultimately be again able to reach them and to act upon them. In the chronic form, which occurs in the old, treatment directed to the morbid process seems to be entirely useless. The disease is a local senile change, and as such is beyond the reach of any influence we can bring to bear upon it or which the future is likely to afford us. The facts at present available regarding the chronic forms, which occur at earlier periods of life, suggest the same correspondence with analogous affections of the spinal cord and a similar practical application of the correspondence. In the form in which the muscles waste, the hypodermic injection of strychnine, of unquestionable service in the spinal affection, should always have a thorough trial. There is no reason why it should not be effective in the atrophic bulbar paral- ysis as well as in " progressive " muscular atrophy. Its use is not contra-indicated by previous administration by the mouth, and should not be delayed, because, if arrest can be obtained before the malady is advanced, even if im- provement is not secured, the condition of the patient is far 88 BULBAR PARALYSIS. better than if the disease had obliterated most of his power of articulation, and some of his ability to swallow. The danger of pulmonary consequences, which result from the latter, must also be kept in mind in the treatment of every form of bulbar palsy, and every expedient, such as slow, deliberate eating, and food in a semi-solid form, which may lessen the chance of the entrance of particles into the trachea, and thence into the bronchi, should be insisted on. By these means, death may be averted or postponed more fre- quently than we can perceive. I have unfortunately not found anything effective in diminishing the flow of saliva, which is so great a trouble to many patients. In general management, do not forget that the sufferers are often much distressed by the inability to make them- selves understood ; endeavor to lessen this as far as you can, by persuading them to express themselves in writing rather than by efforts to speak, which only end in failure, and involve a strain on the affected structures which cannot be other than injurious. LECTURE VI. FACIAL PARALYSIS. Gentlemen : — Complete paralysis of the right side of the face, in a child of seven, who presents no other symptoms, — that is the problem before us. You have heard the questions asked, and the answers given by the child's mother, and that we found no evidence of a cause ; you heard the mother give a negative answer to every inquiry, — there had been no blow, no exposure to cold, no dis- charge from the ear. Those are the three chief causal facts to be sought in every case. Even pain in the region of the nerve was denied. But one other question was asked that should never be omitted in such a case : Has the child suf- fered from earache ? The mother then remembered the fact that had been forgotten, remembered it with surprise, so little had she thought it connected with the affection of the face. You see the symptoms which the child now presents, and you have heard the story of the onset.* You see that the face is symmetrical at rest, but in movement all sym- metry is lost in the distortion produced by the limitation to one side of all the familiar movements that the will can cause, or by which emotion is expressed. There is no movement, voluntary or emotional, in any part of the face. Paralysis of all the muscles supplied by the facial nerve, on one side only, and without other symptoms, always means disease of the nerve trunk. Practically, moreover, Clinical Journal, February 14, 1 894. * From a report by Mr. H. Caiger. 89 90 FACIAL PARALYSIS. if it occurs without obvious disease or injury near the nerve after it emerges, it means disease of the nerve during its passage through the bone. These two facts should be fixed in your mind, and the reasons for the conclusion should also be clearly understood. Never learn a diag- nostic rule, indeed, never accept any general assertion, without also endeavouring to ascertain on what the rule depends, or the assertion rests. Unreasoned conclusions are the bane of students. Without the facts to make them cohere with our previous knowledge, such conclusions will soon slip from the memory, or if they do happen to be retained, it is only by dint of pressure which alters their form, or drives them into some place to which they do not properly belong. They are remembered wrongly, and do you more harm than good. But fix the assertions by their evidence, the rules by their reasons, and not only do they remain, but they take root, and they become part of your real knowledge and a source of increasing power. You are no doubt familiar with the distinction between the two chief forms of paralysis of facial muscles, — the general palsy of all parts which is often, as here, absolute, and the form in which only the lower part of the face is involved, which is never absolute. The first, as you know, means disease of the nerve, or of the nucleus from which the nerve comes, while the other (so common as part of hemiplegia) generally means disease of the cerebral hemis- phere, and always means disease of the path between the cortex and the nucleus, or disease of the cortex itself. Hence the latter is sometimes called cerebral facial palsy, but this is not an accurate term, because this form may result from disease in the crus, or in the upper part of the pons. A more accurate name, therefore, and one by no means inconvenient, is supra nuclear palsy. On the other hand, the general palsy of the face may be either nuclear or infra-nuclear. It is often called " peripheral," but here STEPS OF ADVANCING SCIENCE. 91 again we have a word which it is wise to avoid. If it does not mislead, it is liable to distort a student's first concep- tions. Such distortion involves a waste of mental effort, since it has to be rectified. The "periphery" properly means the distal extremity ; " peripheral palsy " should be due to disease of nerve-endings — such paralysis as many poisons produce, not rarely, in other parts. But the word is not used in that sense in reference to the face. It is used to mean disease of the fibres anywhere in the course of the facial nerve, and even palsy that is produced by a lesion of the nucleus within the pons ; that is of the central origin of the nerve, is sometimes spoken of as "peripheral." This point, — the use of the word " peripheral " in connec- tion with palsy of the face, may excuse a digression for a moment, to note an illustration of that oscillation of opinion that always attends advancing knowledge. The truth of yesterday may be untrue to-day and true again to-morrow. The grounds of an induction may become insecure when we know the facts more thoroughly, and yet, again, new facts may re-establish that which seemed exploded. When many of those who now teach, first studied medicine, they were taught that the nerve endings are the seat of the lesion in the commonest forms of facial palsy — that which follows exposure of the side of the head to cold. The cold, acting on the surface, was assumed to act on the nerve endings in the muscles of the face and thus to cause the paralysis. The same explanation was given of other palsies following an exposure to cold, such as the local atrophic palsy of children, "infantile paralysis." This pathology was purely hypothetical ; no evidence of it had been ascertained, and there was no demonstrable analogy that could be produced to support it. Indeed, it now seems to us strange that the opinion should have been accepted so generally, since a little consideration will show how great a difficulty there must always have been in adjusting it to the facts. Why 92 FACIAL PARALYSIS. should cold act upon the extremities of one nerve, upon all of them up to the middle line, and no further, not a single muscle or fibre on the other side? Why should the effect of cold acting on the surface be absolutely limited to the endings of one nerve? This difficulty does not seem ever to have been recognised. When increasing precision of observation, and a wider comparison of facts, suggested irresistibly that such a paralysis must be due to a process acting on the fibres where all are near together, or on the contiguous structures from which they come, and when it was discovered that the latter was the true pathology of infantile palsy, the idea that disease of nerve endings was a cause of paralysis disappeared. All such palsies were ascribed to disease of the centre or nerve trunk. Again, new facts came, and have compelled a partial reversion to the old opinion, that a primary affection of the terminal parts of nerves is a frequent cause of palsy. But it is chiefly in the limbs that we find evidence of this. We have, indeed, returned to this opinion in some cases of paralysis of the face, but we have not returned to it as an explanation of the palsy that is limited to one side of the face, and for which it was once considered the adequate explanation. It is very unlikely, however the pathological pendulum may swing under the impulses of new discov- eries, that we shall ever find reason to think that such palsy as you see before you in this child, is ever the result of disease of the nerve endings. We shall presently note the significance of this. Let us return to the two main forms and their distinc- tions — the infra-nuclear palsy which is often complete, and the supra-nuclear palsy that is never complete. You per- haps know the hypothesis by which the difference is ex- plained ; it is indeed somewhat more than a mere hypothe- sis. In proportion as muscles act together on two sides, the muscle on each side is represented in both hemispheres DIAGNOSIS OF COMPLETE PALSY. 93 of the brain ; that is, either hemisphere can act on the mus- cles of both sides. Hence, in disease of one hemisphere of the cortical centre, or the path from that centre, the muscles escape in proportion to their bilateral association. When I say they escape, I ought to say they either escape, or they suffer only at first. They may be affected for a few days, but the initial weakness soon passes away, at least in great measure. Apparently the structural arrangements by which the hemisphere acts on those muscles on its own side are only slightly in habitual use, but their functional capacity quickly increases. This point, however, is one that we must consider another day; it would lead us too far from our present subject. It is the complete palsy — complete in range — of which an instance is before us, with which we are now concerned. We can carry the diagnosis beyond that of the exclusion of supra-nuclear disease. Paralysis of all parts of the face, if it exists alone and is on one side only, means disease of the nerve after it has left the cranial cavity. Theoretically it does not ; the whole nerve might be affected alone within the skull. But, as a matter of experience, it is not. The statement I have made is true in fact. Further, if there is no indication of a cause acting on the nerve after it has emerged from the canal in which it passes through the bone, such a paralysis means disease of the nerve within this canal. This conclusion is, I need not say, of the utmost practi- cal importance. In a large number of cases of facial palsy — in the majority indeed — it takes us at once to the seat of the disease. But to be able to use such a diagnostic rule you must know its reasons. Without a knowledge of the reasons few such rules are useful, because you cannot feel so sure of their validity as you must feel if you are to employ them with the confidence that is essential. You cannot feel sure that there are no exceptions, unless you 94 FACIAL PARALYSIS. realise why there cannot be exceptions. The facts on which this rule is based are these : — First, there are two portions of the nerve which cannot suffer alone on one side only : these parts are the beginning of the fibres and their terminations — the cells from which they spring, and the structure in which they end within the muscle. These parts are so situated that all of those of one nerve cannot be affected on one side and alone. The nerve cells occupy a considerable area of the pons, and those for the muscles of the eyelids ?,:t some distance from those for the lips. The cells are adjacent to other important structures. Hence, the whole of the nucleus cannot be damaged by one lesion, so as to affect all parts ::' the face without other structures suffering, and suffering in such a manner as to give rise to obtrusive symptoms. Moreover, this is true not only of the nucleus, but also of the fibres proceeding from i: di their passage through the pons. All the feres : f the nerve are never damaged within the pons without adjacent struc- tures being alsc damaged soas to cause conspicuous symp- toms. Thus the isolation of the paralysis of the face, the affection of the whole of one nerve and of that only, ex- cludes an organic lesion within the pons. The nerve-cells may suffer, and may have their function abolished apart from what is commonly termed an organic lesion. They may cease to act in consequence of a degen- erative process, or in consequence of the influence of a toxic agent. This is true, also, as I have already intimated t: you, of the other extremity of nerve fibres — their ter- mination within the muscles. These are now known to be occasionally the seat : ■:' degenerative processes, and they are also known to be the parts influenced by certain poisons. But such agents do not act only on the nerve endings of one side. The reason for this is important, and, as I just now said, must be specially noted. It is a general law of extreme importance. The same structures on the AN IMPORTANT GENERAL LAW. 95 two sides possess the same pathological susceptibilities. They suffer together in degenerative processes, and they suffer together under toxic influences, whether these be what we commonly term poisons or whether they are only seen in what we call diseases. Thus both sides of the face may be paralysed throughout as part of diphtheritic par- alysis, and as part of multiple neuritis, due to some toxic agency, — but one side of the face never suffers alone. This is the point to which I adverted in speaking of the theory that cold acts on the extremity of the fibres. Our present knowledge shows that it could not affect the terminations of all the nerves on one side and none of those on the other. Such a cause must, if it acts on the extremities of the nerves, cause bilateral paralysis. As a fact, it does so. Remember, then, in all parts of the nervous system, and in relation to all cases of the character I have mentioned — those due to degenerative processes and to toxic influ- ences — the effect is bilateral ; and a complete palsy of any part on one side practically excludes such general causes. The law which underlies these facts, therefore, may be expressed thus : A palsy which is directly due to a general cause is bilateral ; conversely, a bilateral palsy indicates a general constitutional influence. A unilateral palsy is due to a local cause ; it is not the direct result of a general process. I use the word "direct" result in order to be strictly accurate. The meaning of the restriction is this. The direct effects of general causes are due to their action on the nerve elements themselves ; but general states sometimes cause local effects by acting on the vascular system, or they may predispose all parts to suffer from an influence which acts only on one. Moreover, a unilateral palsy means a process which begins outside the nerve elements and affects them secondarily. That is a most important general law, applicable to all parts of the 96 FACIAL PARALYSIS. nervous system — most important to remember in all cir- cumstances and in all localities. Next, the limitation of the palsy excludes disease of the nerve within the skull. After it has left the pons, a process outside the nerve fibres cannot reach a considerable decree and affect the facial nerve alone. The auditory nerve is contiguous to it from the surface of the pons to the bottom of the internal auditory meatus, and a process external to the facial nerve, sufficiently severe to paralyse it com- pletely and in all its parts, must affect the auditory nerve as well ; often it affects other nerves that are in the neighborhood. On the other hand, a cause acting on the facial nerve after it has passed through the skull will be conspicuous ; it is always an injury or a considerable local inflammation, and, in either case, is obvious. Hence, therefore, you can now understand the foundation of the important diagnostic rule which I laid down before : that a complete unilateral palsy of the face, without other symptoms, must mean disease of the nerve as it passes through the bone. The diagnostic problem is thus narrowed in regard to the seat of the lesion, and this facilitates the process of diagnosis even more than may at first sight appear. We cannot, indeed, carry further the process of exact localisa- tion, except that, if the short length of nerve which inter- venes between the origin from it of the vidian above and the chorda tympani below, is affected, taste is lost on the side of the tongue in front. But this is of little real value, because a morbid process may begin at one spot and may spread through a considerable extent of the nerve. The state of the palate gives us no localising information. Paralysis of the palate is never produced by disease of the facial nerve. The belief that it is so is one of those curious pathological myths which have arisen from the misinterpre- tation of what may be termed normal abnormalities, if you NATURE OF THE LESION. 97 will forgive the expression — deviations from perfect sym- metry, which have no significance for us except in so far as they may lead us into error. But the special aid in diagnosis which we derive from this restriction of a morbid process to the part of the nerve which lies in the narrow winding canal in which it passes through the temporal bone, is due to the aid it gives in the next step in diagnosis — the step which is of chief practical importance, the nature of the lesion of the nerve. In this part of its course the nerve suffers from only three morbid processes, and of these three one can be excluded without difficulty. The processes are: — (1) primary inflammation of the sheath and interstitial tissue, which is the cause by which the most common form is produced, that which is due to cold ; (2) the spread to the nerve of inflammation due to ear-disease; (3) its compression or destruction in consequence of a growth in the bone. It is the last of these which may safely be put on one side unless there are other obtrusive indications of a growth, because it is a very rare cause, and not to be thought of unless such indications exist. The diagnosis between the two forms of inflammation, that which is communicated from the ear, and that which is primary (the so-called " rheumatic " form), is seldom difficult. They differ in the period of life at which they chiefly occur; although each may be met with at any age, ear-disease is a rare cause except in childhood, while primary neuritis is seldom met with till childhood is over. Moreover, the mischief in the ear which spreads to the nerve is generally considerable in degree and in duration. In most cases there is actual disease of the bone, and it is the progress of the caries that brings the associated inflam- mation into such proximity to the nerve that this becomes affected. In such disease there is almost invariably per- foration of the tympanum and a constant discharge from 9 98 FACIAL PARALYSIS. the ear. If there is no history of discharge, disease of the ear is not likely to be the cause. But this rule, true of facial paralysis in general, is not invariably true of it. Exceptions are met with, and this case is an illustration of the fact. We often hear vaguely of exceptions which " prove " or " test " the rule. The rules to which this say- ing is applicable are in most cases general rules ; they are laws that are true of the majority of instances which come under them ; if these are separated and scrutinised, it will be found that the exceptions occur under special condi- tions, and that these special conditions have been ignored in formulating the general statement. We have to be chiefly guided by the majority of cases, but we should recognise the existence of exceptions, and know when they may occur, that we may search for their indications if we have any reason to suspect an exception to our rule. If we attempt always to give weight to them, such weight as we give to the majority, we shall be in constant uncertainty. Indeed, it is not unlikely that we may come to the con- clusion, as a distinguished scientific man remarked to me of the impression left on him by one of the most famous teachers of his early days, that " no sane man could make a diagnosis." The exception to the rule that obtrusive signs of caries long precede the facial paralysis which results from otitis, depends upon the anatomical conditions of the ear. This has not indeed been actually proved, but we know that exceptional conditions often exist ; they explain that which would be otherwise inexplicable, and which nothing else explains. In some cases, fortunately not common, the facial nerve is separated from the tympanic cavity by a layer of bone so thin that inflammation can readily pass from the cavity to the nerve sheath. In such, bone disease is not necessary for the extension of inflammation to the nerve. A SIMPLE SOURCE OF ERROR. 99 You can now understand why I asked so carefully about the earache. It is not enough to ask if there has been discharge. Discharge generally means disease of bone : disease of bone is the common cause of secondary facial paralysis, but the nerve is sometimes affected by extension when there is no bone disease. This case seems to be of that character, and hence, I am anxious to impress its facts upon you. It is an illustration of one of the two chief simple sources of error in diagnosis, not seeing the com- mon, not knowing the rare. I say simple causes as opposed to the more complex sources of error in rea- soning. This is an instance of the rare; it is an example of the occurrence of facial neuritis by extension from the middle ear, without bone disease. It is exceptional, because it must depend on exceptional conditions. A " passage " of inflammation implies a way for it to pass out. Normally, there is no way free enough for the passage of simple in- flammation. But simple inflammation may spread from the lining membrane of the tympanum to the nerve, when the layer of bone which separates the tympanic cavity and the nerve is thin. Indeed not only may it be thin, it may be even actually deficient. Inflammation not only may, but, if considerable, must then spread to the nerve. Ves- sels also pass through the bone, and by these an intense inflammation may pass ; but it is doubtful whether a simple catarrhal inflammation, such as this child seems to have had, would do so, if the conditions were normal. One other point, and that of great importance, the case also illustrates. The earache of children is almost always due to inflammation of the middle ear. Most of you know its character from the recollections of childhood, for few children escape some attack, and the peculiarity of the pain impresses itself on the memory. It is one of the maladies that may be called " domestic diseases " ; few ioo FACIAL PARALYSIS. mothers dream of sending for medical assistance for what they are pleased to call " simple earache." But between the earache which lasts a few hours and then passes off, and the earache which is the prelude to a suppurative in- flamation, there is every gradation. There seems to be no difference in the character of the pain. It is important to remember this. It is one of the facts that should be part of the education of the mother — one of the many facts that might, with great advantage, replace the rubbish that has drifted into the maternal mind: by no means harmless in the minds of the many persons who "get on much better without a doctor." I lately saw a child with infantile palsy of one leg. This leg had suddenly become powerless. The mother — with no excuse of poverty or station — consulted her female friends, and, resting content with their assurance that it was " only the teething," and that "the leg would soon get all right again," allowed five whole months of utter immobility to pass before she thought it necessary to ask her doctor to look at the child's leg. And many cases of suppuration in the ear and of disease of the bone are due to neglect of the warn- ing of simple earache. In the case of children who are liable to earache, great care should be taken to guard against exposure of this part of the head to cold, and especially to cold east winds ; and still greater care should be taken to get rid of an attack as soon as possible. How seldom are the lessons learned that are, nevertheless, familiar in proverbial form, such as that embodied in the adage about the " stitch in time." We could wish the statement were literally true, and that the absence of early treatment, which so often would cut short a grave disease, could be made up for by the amount represented by the multiplier of the proverb. If you take all the nerves of the body, and consider the frequency in which they are so diseased as to cause symp- OBSCURE CAUSES. 101 toms, I think the fifth, the sciatic, the ocular, and the facial nerve would be their order. The reason for the frequency of facial paralysis is not yet entirely understood. We can understand, however, why it is so obtrusive when it does occur. It is manifested with peculiar readiness, because outward swelling is prevented by the rigid walls of the canal. Hence the inflammatory effusion compresses the nerve fibres, and at once interrupts the conduction of the motor impulses, quickly causing inability to move the facial muscles. But this, although it explains the fact that even a slight inflammation causes considerable palsy, leaves its actual frequency still mysterious. Paralysis of this nerve from cold is certainly more common than we should anticipate, considering that other nerves are not less exposed. Perhaps the cause is to be found in some conditions of the circulation within the canal, in consequence of which a congestion that would otherwise be transient and harmless, leads to undue stasis, and becomes an actual inflammation with all its grave results. The more remote causes of the primary neuritis are obscure. It is occasionally associated with the diathesis that causes fibrous and muscular rheumatism, and which is probably not far removed from that which causes or results from gout. I have twice known a patient to have facial neuritis at one time, and at another an analogous rheu- matic inflammation of all the nerves at the back of the orbit. Remember, too, that what we call fibrous rheuma- tism is probably also not very different from an inflamma- tion. Certain it is that this still mysterious muscular rheumatism may become inflammation. Many cases of sciatic neuritis, certain and severe, arise by the traceable extension, along the fasciae to the sciatic notch, of a pri- mary lumbago. We will not attempt to ascertain, in this child, the electri- cal irritability of the muscles. We know what condition 102 FACIAL PARALYSIS. we should find. No complete paralysis of a motor nerve continues for a month without the nerve fibres degenerat- ing below the lesion, and there is always loss of all irrita- bility of the nerve trunk, loss of the faradic irritability of the muscles, and increased irritability of the muscular fibres to voltasim. We should learn nothing by the exam- ination, and you have frequent opportunities for observing the facts. Without doing good to ourselves, we may do harm to the patient. The paralysed muscles will need electrical stimulation, and in the case of children it is neces- sary to be extremely careful in the application of electricity. If you so use it as to cause pain the child will be fright- ened, and will dread each therapeutical application. Once thoroughly frightened, a child seldom loses the dread, and no child can endure a frequent distressing emotion without harm. If care is taken, all that is needed, or almost all, can be achieved without the production of any of this injurious alarm. But to secure this result the first applica- tion must be so feeble that no new sensation is produced. Indeed, it is well, the first time, not to allow the current to be strong enough to be felt. Then, if the current is grad- ually increased in strength in successive applications, after a few days a strength may be used that will make the mus- cles contract visibly without eliciting a tear. It is surpris- ing how strong a current children will bear if this plan is adopted. For the same reason you should never use the faradic "current" in the case of young children. Move the hammer with your fingers instead of permitting it to oscillate automatically. The less frequent momentary cur- rents will cause a muscle to contract with merely a pecu- liar shock-like sensation, devoid of pain. If you use the repeatedly recurring shocks, produced by the automatic interruption, a painful stimulation of the sensory fibres is produced, even with a lower current than will cause the muscles to contract. The difference, in the case of child- AIDS TO PROGNOSIS. 103 ren, is of great importance, for the reason I have men- tioned. For the same reason, also, if you find it impossible to cause visible contraction of the muscles without produc- ing distress, be content with a little weaker current. I cannot understand the prevalence of the notion that in any therapeutic procedure, with drugs, electricity, or any other agency, no good is done unless a certain physiological effect is produced. Surely the production of a manifest effect is only a question of degree. If a certain strength of current cause a visible contraction of a muscle, it does so by making a large number of the fibres contract with sufficient energy for the effect to be seen. Are we to assume that a current just below this strength causes no contraction because we cannot see the whole muscle shorten ? There must be contraction of the fibres before we reach that degree which causes visible movement, and it is surely impossible that this stimulation can be without the influence on the nutrition of the fibres, and on the maintenance of their irritability, which the stronger current has. On the contrary, the consideration of the fact sug- gests distinctly that such a weaker current only needs to be continued for a longer time to do all that the stronger current can do. Remember, all that electricity can do in such cases is that which I have mentioned. We have no reason to believe that it has any influence on the nerve fibres ; it does seem to keep the muscles in a better con- dition for the nerve impulses to act upon them when con- duction is restored, and to enable them to respond better to the motor influences when these can again reach them. Hence, in all cases in which the paralysis is complete, or considerable, for more than two or three weeks, it is desir- able to employ electricity. Moreover, we do not here need an electrical examina- tion to assist in forming a prognosis. This is, it is true, one of the most important services electricity renders in 104 FACIAL PARALYSIS. such cases. It is surprising to those who cannot read its language, that of two cases of facial palsy alike complete and of the same duration, say two weeks, an electrical examination should enable a positive statement to be made that one case will be well in a fortnight, and that another will endure for months and will never completely pass away. And yet it is so. But we have other aids in prognosis. We can often draw conclusions of great value from the simple consideration of the conditions under which a malady developed, and, when a process in one place is due to extension from the same process elsewhere, the character of the primary symptoms is sometimes of significance. Here we have the fact that the paralysis re- sulted from an inflammation which was transient. The pain in the ear lasted only a few hours. Doubtless the process may have lasted longer, since when effusion and swelling occur, the pain of inflammation often lessens, but the duration of pain which is due to inflammation in the ear is of some value as a guide to the duration, and, there- fore, to the intensity, of the process. Thus judged, the process here was brief, and, being brief, was also slight ; for duration and degree are proportioned in acute inflam- mation. From this we may conclude that the inflam- mation of the nerve was probably only such a brief moderate process as would be limited to the sheath to which it first extended. The effects have already lasted a month, but before long some signs of improvement may be expected to occur. Over and above this, we have also the general law that cases of facial paralysis from ear dis- ease, on the whole, run a more favorable course than cases of facial paralysis from the primary neuritis. The reason for the average slighter severity of the affection is, perhaps, to be found in a more limited extent of the in- flammation. It is reasonable to assume that an inflamma- tion which is communicated will not affect so considerable TREATMENT. 105 an extent of the nerve as one which is due to a primary process. In all such cases of facial paralysis the treatment must be twofold. Besides that suited to any recognisable con- stitutional state (which I assume you do not need to be told) we have to treat the primary disease of the ear, and we have to treat its consequence. In the case of ear dis- ease, this must first engage our attention, but the measures needful do not come within my own province to describe in detail. The most important, however, is unquestionably to afford a free exit to any pus which may have accumu- lated in the cavity of the tympanum, and to guard against any obstruction of such exit, including that which is pro- duced by non-absorbent cotton wool. Absorbent cotton wool changed two, three, or six times a day, according to the amount of discharge, is safe. The importance of cleansing by antiseptic washes you will have already learned. For the secondary inflammation of the nerve sheath we can do nothing directly, except by the applica- tion of counter-irritation. This is, indeed, the chief local measure in all cases, whether produced by cold or produced by extension. A blister should be applied over the mastoid process, and should be repeated as soon as the skin will permit. Never put a blister in front of the ear, over the place of exit of the nerve. You cannot blister without producing a little subcutaneous cellulitis, and I have known a trifling cellulitis in this situation to be a cause of facial paralysis — the inflammation reaching the nerve sheath, and spreading along it into the Fallopian canal. In all recent cases, moreover, hot fomentations applied over the ear and its neighborhood, constitute a very useful measure. They should be used for one quarter of every hour during the first day. We do not precisely know how hot fomenta- tions act, but it is certain that no measure has so potent an 106 FACIAL PARALYSIS. influence on inflammation in its earliest stage. Probably the stimulation of the sensory nerves and the heat com- bined cause alternate contraction and dilatation of the vessels, which lessens the blood stasis. Perhaps, more- over, the nerve stimulation has some even more direct influence on the process of inflammation. But whatever the explanation, the fact is certain. The last element in the treatment is the application of electricity. This is only needed when the nerve fibres undergo degeneration. As long as the nerve retains any degree of excitability, and as long as the muscles contract when the induced or faradic current is applied to them, electricity is not needed. Its use is to keep up the nutri- tion of the muscles and to keep up their excitability. We have conclusive proof that it does the latter ; we have no proof that it does the former, because we do not find that the muscles to which it is applied waste less or waste more slowly than those to which it is not applied. But it is cer- tain that functional excitability cannot be maintained with- out nutrition being also influenced. You know the kind of electricity that should be applied. This is one of the elementary facts which every student learns or should learn as his first acquisition in the therapeutics of the nervous system. You must apply the kind of electricity to which the muscles respond, and this, as we have seen, is the voltaic current. You apply it in such a manner that it is interrupted, not frequently as the induced current is, but slowly, as by stroking down the muscle with one electrode, the positive or the negative, to whichever the muscle is most sensitive. It is not always the same, even when the nerves are degenerated. We interrupt slowly because, as far as we can judge, the element in the form of electricity which causes the difference in the reaction is one of time. Muscular tissue is much lower in the scale of excitable THE VOLTAIC CURRENT. 107 tissues than is nerve tissue, and seems to be unable to respond to an electric current unless this has a duration considerably exceeding the very small fraction of a second which elapses before the automatic interruption stops the current of the induction coil. LECTURE VII. FACIAL CONTRACTION AFTER PALSY. Gentlemen : — When I asked your attention to the chief symptoms of facial paralysis, I had to leave for subsequent consideration the troublesome condition which accompanies return of power — the secondary state of contraction. It is not a trifling matter in many cases. It may be almost as great an evil as the original paralysis. It is, moreover, a subject of considerable interest, and its study throws light on other conditions. Observe carefully the face of the patient before you. The left side presents a slightly deeper naso-labial furrow than there is upon the right side. It would be natural for an observer to assume that the side on which the furrow is least marked is the weaker side, is that which was para- lysed. As a matter of fact, such an assumption is commonly made by students in such cases. The furrows of the face are determined by muscular action, and where furrows are less there is less muscular action, and therefore muscular weakness seems to exist. The impression is strengthened by the aspect of a gentle smile ; it is not easy to obtain it — the attempt often results in no movement, or in one that is much in excess of " gentle." But the request for one, cautiously put, often obtains what is desired. You see it does so here. You see that the gentle movement is dis- tinctly greater on the side of the deeper furrow, and the smile thus seems to confirm the opinion that there is more power of movement on this side. Clinical Journal, September 19, 1894. 108 ASSOCIATED MOVEMENTS. 109 But our ability to excite emotion may be sufficient to make a patient laugh ; or we may obtain an energetic movement in elevation of the upper lip. In this you can see how much slighter the strong movement is on the side on which the slight movement was greatest. On the side of the deep furrow a slight movement is greater, but a strong movement is much less. Note, also, that when the eyes are tightly closed there is the same difference. An excessive degree of slight movement is associated with a diminished degree of strong movement. But you may have already observed for yourselves that there is more than this difference in degree. Even in the gentle smile you could not fail to see that the eyelids were more approximated on the side on which the slight move- ment was most considerable, and that this was conspicuous also with the stronger movement. It is very marked in the laugh. You know that emotional movement normally involves all parts of the face. The ocular fissure is nar- rowed in a laugh, the eyes may even be almost closed, while the angles of the mouth are drawn outward and up- ward by the zygomatic muscles in the ungraceful distortion of the face which custom makes us appreciate so keenly. This associated action is increased in the condition we are considering. With the increased readiness of action there is an increase in the associated action of the different parts. Moreover, with these there is associated the feature to which I first directed your attention — the muscular con- traction which deepens the furrows. This muscular con- traction is a persistent state ; it is accompanied by, and, indeed, I might say, illustrated and emphasised by occa- sional slight clonic contraction in the muscles, which you will have no difficulty in perceiving in this patient, if you presently carefully observe her. The furrows which come with years, sooner or later, are the result of the shortening of the muscles which have no FACIAL CONTRACTION AFTER PALSY. been most frequently employed in their varied uses, and especially in the expression of emotion, a shortening in- creasing with years and being increased in effect by the loss of elasticity of the skin. The emotions that are chiefly dominant impress their transient influence on the muscles by long repetition as a permanent effect. Everyone knows the varying expression which the furrows of age produce in the aspect of the face, and those who know the story of the life can often read it in the lines. Yet most persons remember instances which cannot altogether be thus ex- plained. We sometimes meet with a face which bears the aspect of sorrow as the sequel to a tranquil life, or see the furrows of a constant frown in one who is seldom angry, or the transverse frontal lines when surprise or concern has seldom been dominant. Of course, many of the lines which Time's fingers trace are due to that which is common to all persons, habitual movements which have only a little to do with emotion. The zygomatic muscles, for instance, are constantly employed otherwise than in the smile or in ex- pressing pain. Their contraction is the result of their habitual action, defectively opposed. But when all allow- ance is made for such influences, there remains much con- traction which we cannot explain except as the result of hereditary influence — the effect of emotion in past gener- ations, persisting to the present, and telling a story which is untrue of the present, but, like so much else that is myste- rious, is true of a perhaps distant past. It seems to be an indication of the continuity which makes our race extend unbroken, in all its variations, through time which we can- not estimate. The fact that the contracture which follows facial paral- ysis resembles in its effects the condition which age induces, renders its effect most distressing to the young. The con- tracture follows inevitably in every case of facial paralysis, with the exception of those which last but a few days. If DISTRESSING DISTORTION. in the paralysis is moderate, the contraction is moderate ; if severe, so is this sequel. If the time of life has come when lines are proper, then the contraction reproduces the due symmetry of the two sides of the face ; and if it does a little more, the excess is almost imperceptible. Although the associated overaction may cause a difference in the expression of emotion, unless there is energetic movement, the difference is slight. But that which is natural in age is unnatural in youth. The contraction produces its furrows in all periods of life, and the result in ybuth is that a face may one side correspond to the proper aspect of " sweet seventeen," and on the other to forty-five. The contrast is an evil as great as is the facial paralysis itself, and even greater. In the young, during the stage of palsy when the features are at rest, no difference is perceptible between the two sides, and it is only on movement that distortion occurs. But this late contraction causes a conspicuous contrast between the two sides, which is constant at rest. It is the more distressing because no hope can be held out that it will pass away. So keenly is it felt, that one patient of mine, a girl of eighteen, formerly epileptic and not of very stable mind, was so distressed at the daily reflection in the looking-glass, which she could not avoid, that she one day wrote a letter to me. It was to thank me for my efforts to do her good, but to say that she could bear it no longer, and that when the letter reached me she would be dead. The letter was to reach me on my visit to the hospital, and when it reached me it was true. In the old, however, the contraction is cosmetic in its influence. The early palsy obliterates the bilateral sym- metry, because it substitutes the smoothness of youth for the wrinkles of age, but the symmetry is restored by the late contraction. It is curious, by-the-by, to note how difficult it is for many persons who have long ceased to be young, to believe that the normal side of the face is not 112 FACIAL CONTRACTION AFTER PALSY. that which is affected. In the early stage of the paralysis the smoothness of the affected side causes the furrows on the other, familiar as they are, to seem by contrast entirely strange. That which is so unseemly cannot be natural. They assume, therefore, that the side toward which the mouth is " drawn," is that which is affected. The convic- tion is not, as you might think, confined to one sex ; I will not say that it is not a little more common in one, but it is frequent enough in the other. A mirror has an interest for men as well as women. This late contraction, as I have told you, always follows facial paralysis except when very slight. It accompanies return of power, but only return of power that is incom- plete. This statement may surprise you. It implies that recovery of power is incomplete except in the slightest cases. You may have seen many cases of paralysis of the face which seem to have perfectly recovered. But you will find, on careful examination of the movements, that recov- ery is hardly ever perfect. You will find, also, that when you find any indication of contraction and associated over- action, you can always detect imperfect, power, however long it may be since the acute affection. Indeed, the over- action may be more conspicuous than the persistent defect ; but the defect is always there, although it may be incon- spicuous. In every organ and structure of the body the condition after inflammation is never as it was before. Could we perceive the minuter structure, we should be astonished at the degree of persistent change. In most parts the effect on function is not perceptible, because the function of the organ is manifested by the combined action of all its parts. Where each part has a separate function, the effect of persistent change is evident. This is so with the face. The evidence of imperfection is the greater the more pronounced the function, and the more delicate its adjustment. The face is the great vehicle of emotional A CASE OF BILATERAL PARALYSIS. 113 expression. No pain or sense of pleasure comes and goes without a change in the state of contraction of the facial muscles, too slight perhaps to be analysed, but too definite to be unperceived. It is not surprising that an organ of expression so delicate should manifest the slightest inter- ference with the process of conduction of the impulses from the brain. This peculiar condition does not cause suffering. It may be attended with a sense of tightness in the face at the onset, but afterwards the patient is scarcely conscious of it, except when movement brings its presence to the mind. This, however, is not more than the patient quickly gets used to. The muscles in the lower part of the face on one side are, in general, adequately opposed by the action of those on the other side. If the second side should sub- sequently become paralysed, the contracture instantly becomes far more obtrusive both to the patient and to others. It is indeed very rare for the second side to suffer at a subsequent period, and very few instances are on record. A case lately came under my notice : A lady had been under my care five years ago for severe paralysis of the right side of the face, which had left considerable con- tracture. One day she was startled by finding a great increase in the contraction, and that there was deviation of the mouth considerable even on rest, greatly increased by movement. On looking in the glass, her previous expe- rience made it at once obvious to her that the left side had become paralysed. She came to London in a day or two, and the condition was very striking ; the vast increase in the effect of the old contraction, consequent on the loss of power on the other side, caused extreme distortion. For- tunately the attack proved brief, and in a fortnight the fresh palsy had passed away, and her state was as before. How far the quick recovery was due to the treatment adopted (which I have described to you before) or to the slightness U4 FACIAL CONTRACTION AFTER PALSY. of the inflammation, I cannot say. Remember that this is one of the maladies in which nine-tenths of that which can be done by treatment, can be done only during the first few days. It is one of the many diseases in which therapeutic knowledge on the part of the general practitioner, per- fectly ready for use, is of the utmost importance. Physi- cians are impressed perhaps unduly with this fact, because most of the patients who have reason to seek their subse- quent aid, are cases in which the needed treatment has not been forthcoming. It is impossible to overlook the room there is for more knowledge of the proper treatment of many diseases, knowledge which is such as to enable the necessary measures to be at once and confidently adopted. Practitioners, perhaps, do not adequately realize how large is the proportion of all possible good which they alone can do. I have dwelt at length on the features of this late con- traction because, trifling as it may seem to be, its practical importance is very great. As I said, although the annoy- ance occasioned by the paralysis may be greater in degree, it is almost always limited in its duration. That which is caused by the contracture is generally considerable in the first half of life, and it is not limited in its duration. It only ceases to annoy when the effect of years neutralizes, by balancing, its effect. Why it should be thus persistent will be seen if we endeavour to perceive something of its cause. Its cause is certainly the functional state of the nucleus of the nerve produced during the period of palsy. During the state of immobility a condition of the motor centre of the nerve has been induced by which a slight degree of excitation causes undue action, although the hindrance to conduction along the nerve is such as to lessen the total amount of energy which can be transmitted. It is not difficult to conceive that the narrowed fibres should have ITS CAUSE. 115 their capacity for conduction restored so that they conduct perfectly a slight amount of nerve energy, and yet the amount of force they can conduct is permanently reduced. There is an explanation of this in the mode of conduction ; but I cannot now make it clear, because I should have to begin far, far back, in the elements of molecular physics. This state which follows paralysis is evidently due to a state of the motor elements of the nucleus, that is, the structures from which nerve energy is evolved. These act too readily, they act spontaneously so as to cause the twitch- ing, and they act on each other too readily, so as to pro- duce the associated overaction. The structures from which proceed the fibres for the different parts of the face, are connected in the nucleus, and these connections which subserve the normal associated action present in health a certain resistance, which limits this action. It is on this resistance, developed by normal function, that the due rela- tion of the associated action depends. That resistance becomes lessened during the palsy. It is lessened in con- sequence of overaction of the centre. Wherever we have overaction of nerve structures extending from one to another, the resistance between them is lessened, and the activity tends to increase the condition. It is as if the grooves in which wheels run are deepened into ruts, and the continued motion of the wheels precludes any diminu- tion in the depth of the ruts. It is as if a channel for water had become widened by the flow, and yet is inaccessible to any direct process of repair ; the constant flow of water through it precludes any natural reproduction of the natural resistance. So this overaction of the nerve centre, once set up, goes on. It is not difficult to perceive how it is probably set up. At least it is easy to see a mechanism that must inevitably have the result. Normally, the action of the central struc- tures of a motor nerve is regulated and restrained by u6 FACIAL CONTRACTION AFTER PALSY. impulses which reach the centre from the contracting fibres of the muscle by the afferent nerves, stimulated by the contraction of these muscular fibres. The voluntary and other impulses continually act on and excite the nucleus when the nerve is interrupted. In the absence of the con- trol which should be exerted on them by the afferent nerves from the muscles, this must lead to overaction of the nerve centre. Moreover, the obstruction in the nerve fibres can- not be without its influence in disordering the centre. Nerve fibres conduct by a process similar to, but less in degree than, that by which the nerve force is generated, and although we do not understand the process, we must conceive that the arrest of the transmission onward of the energy has an effect on the functional state of the centre. If the molecular changes of conduction cannot pass down the nerve, this passage into it must, after a time, be hindered, although the production force is not lessened. Hence also the abnormal action within the centre is increased. The occurrence of spontaneous twitching in the face is explained by the tendency to overaction on the part of the motor structures in the nucleus, and the fact that slight impulses of nerve force can pass readily through the nerve. I should like you to note the difference which there is between this form of contraction after paralysis and that which you meet with in the limbs. In these, the loss of power with changed reaction to electricity is often followed by muscular contraction, but the contraction is in the oppo- nents of the muscles that are paralysed, and not in these muscles themselves. It is due to the fact that the dimin- ished extension permits shortening. But the peculiarity of the contraction after facial paralysis is that it occurs in the muscles that are paralysed. The difference must be referred to the peculiar conditions which cause the loss of power. Nowhere else do we meet with so absolute an arrest of TREATMENT FUTILE. 117 conduction in consequence of a lesion of the nerve which afterward lessens or almost recovers, as we do in the case of the facial nerve. It is, of course, due to the course of the nerve within the bony canal. Moreover, nowhere else have we the constant energetic influences from the brain exciting the motor centre, as in the case of the centre for the face. I have already dwelt on the peculiar relations of the face to emotion, which involve the constant excitation of the centre. The perfect bilateral association of the action of the two sides is another condition which precludes any cessation of the action of the brain upon it. It is very different with the spinal centres for the muscles which may be paralysed by disease of the nerves. Thus we can under- stand that the unique conditions of function and disease, in the case of facial paralysis, give rise to a condition here which we do not meet with elsewhere. You will perceive from what I have said that the condi- tion is beyond the influence of treatment. I have never been able to form an opinion on the question whether it lessens with time or not. If it does, the diminution is very slight. But there are three practical points connected with it. The first is that its advent is certain in all cases in which faradic irritability is lost or considerably reduced. Except in the very rare cases of absolute lasting atonic palsy, some power is always regained, but with it comes this sequel. Since it is annoying, it is always well to warn your patient that when the face recovers power, it will tend to overact, and that this is inevitable. Otherwise the patient is apt to think a fresh morbid state is coming on. When it begins, I think it well to stop electrical treatment. Its commencement co- incides with, or soon follows, the return of voluntary power. There is then enough power to influence the nutrition of the muscular fibres, and to render electrical excitation un- necessary. On the other hand, the stimulation of the sen- sory nerves by electricity tends, I think, by reflex action, 1 1 S FACIAL CONTRACTION AFTER PALSY. to increase the contraction. The third point is the only measure which has, or can have, any influence. A trifling effect on it is probably exerted by what may be called "downward massage" of the face. The fingers should be drawn from the zygoma to the angle of the mouth repeat- edly, with gentle pressure, for a minute or two, several times daily. It is not pleasant to be able to advise so little. But, Gentlemen, remember this — it is true of ourselves and it is true of our patients — next to knowing what can be done and how to do it, the most important thing is to know what cannot be done. Sad, indeed, is the waste of time and money caused by efforts to get that which cannot be ; sad- der still is the waste of hope — hope created by baseless expectation — and the destruction of it that we call disap- pointment, disappointment that would not be were it not for the anticipations that have no justification. LECTURE VIII. ACLTE ASCENDING MYELITIS. G •:'■::'. ■:■■■:■:■: : — Be:": re '■■■t rr::ee: :: the spe:iai subject which ~e are :: ::ns:ier to-day, I cry::se :: describe :: you a case I saw last evening. There are many diseases which, because they are rare, or for other reasons, seldom come under the student's notice, and although description can never take the place of personal observation, it may be useful if the latter is impossible, and it may be the more useful if a case has been recently seen, and its facts are fresh and vivid in the narrator's mind. The patient whom I saw (with Dr. Parnell and Dr. Gray- ling, of Forest Hill) was a lad of 19, paralysed almost completely from head to foot, pale, with the distressing struggle for breath that comes when breathing power is getting less and less. The story of the case is as follows : — A month ago he had his first lapse from virtue, and the consequence was — not a rare one — an attack of gonorrhoea ; it was treated, and he recovered in a fortnight. He had not previously been in good health, having been over- worked at night A little more than three weeks after the onset of the gonorrhoea — about a week after its termination cays be:*: re I saw him — his le~s. :r.e fay, became weal-:, he scumbled : :: the stairs arte in a :e h: crs the weak- ness became considerable. The next day his abdominal muscles were feeble, and the legs oowerless, but with ex:es- sive knee-jerks (which disappeared two days later). On ::•:*::..-: j':-wkz: I::-::: ::. :':-.}. 120 ACUTE ASCENDING MYELITIS. weakness increased and extended during the next three days, and he became febrile. When I saw him last night his state was this : He was sitting, half upright, in a chair, breathing, as I have said, with difficulty, by means of his diaphragm and lower intercostal muscles, at the rate of 56 respirations per minute, the deficiency in quantity of air inhaled having to be made up by increased frequency of respiration. His temperature was raised to 103 . His legs were absolutely powerless, without a trace of knee- jerk — flaccid palsy. From the right sole there was no reflex, but a touch on the left caused a considerable move- ment of the foot ; there was no abdominal reflex, but a stroke on the skin left the bright red line, lasting a long time, which shows acute disturbance of vascular innerva- tion. He could just flex his fingers, and that was all the movement left in his arms. The abdominal and trunk muscles were paralysed, except the lower third of the inter- costals and the diaphragm. The upper part of his thorax was motionless, and the sterno-mastoids were also power- less. He could swallow, and he could just speak, although with difficulty. The left side of his face moved less than the right; the left masseter contracted a little less than the right ; and movement of the eyes was accompanied with brief but wide nystagmus, greater on movement to the left side than to the right. Without any discoverable cause — without any hot application or the like — there had developed, within a few hours, a very large bulla over the inner side of the right ankle. It was thus a case of acute ascending paralysis, using the words in a symptomatic sense, and it had naturally been regarded as a case of that mysterious disease to which the term is specially appropriated, and which is also called "Landry's paralysis." I think that this reasonable conclu- sion is very near the truth ; but there was evidence of more than we have in that disease. Indeed, the case is particu- AN INSTRUCTIVE CASE. 121 larly instructive from the point of diagnosis. " Acute ascending paralysis " depends apparently on a peculiar arrest of central, and perhaps peripheral function, by a peculiar blood state. In it there is no pyrexia ; the loss of function presents a perfect bilateral symmetry, never trans- gressed by such a deviation as the preservation of the plantar reflex on one side when it is lost on the other. The course of the symptoms is uniform, the knee-jerk is lost from the first, and never excessive at the beginning, as in this case. Although it is said that some parts may be passed over in the upward march of the palsy, as, in this patient, were the lower intercostal muscles, yet I think it is doubtful whether this is really true of Landry's paralysis, and whether the assertion has not been due to cases — such as this — which may so readily be mistaken for that affec- tion. The symptoms point to a derangement of the functions of the spinal cord, slightly but distinctly irregu- lar, both in time and place. There is no doubt that Landry's paralysis, as I have just said, is the effect of a toxaemic state on the nerve functions ; but all toxaemic conditions, in their isolated action, cause symmetrical arrest of func- tion; and the difference we had here in different parts and at different times, showed that there was an irregularity in the cause of the symptoms, — something more irregular than the simple inhibition of a blood state. The early ex- cess of the knee-jerk, together with the difference between the two plantar reflexes, and the nystagmus, pointed to a central affection ; the first of these suggested a process in the spinal cord beginning, and at first most intense, above the lumbar enlargement ; while the pyrexia, developing late in the course of the affection, was evidence of inflam- mation rather than of a primary blood state. Had the fever been due simply to the cause of the symptom, it would have occurred early. Its indication — that there was inflammation — agreed with the spinal symptoms. How- 122 ACUTE ASCENDING MYELITIS. ever set up, acute inflammation in the nerve centres always tends to more or less irregular and random influence. This is no doubt largely determined by the participation of the vessels in the process, and by other causes, such as a slightly greater intensity in one tract of grey matter in which it may spread, or the local influence of a small inflam- matory extravasation. Hence inflammation, when set up, extends in a more or less random manner, and its symptoms are often irregular, at least in some degree. The significance of these indications is that, in this patient, we had to deal with acute ascending myelitis. This is the condition between which and the special " acute ascending paralysis " there is generally most difficulty in diagnosis. Although the initial excess of the knee-jerk suggests that the inflammation began above the lumbar enlargement, and that the centres in the latter were merely irritated at that stage, yet the complete palsy of the legs shows, however, that even then the damage to the cord above the lumbar enlargement must have been considerable. As the inflam- mation spread, upwards and downwards, the damage to the centres in the grey matter of the lumbar enlargement caused absolute loss of the knee-jerk, but it would seem that, even to the last, the lowest part of the cord on the left side was not completely involved, so that a plantar reflex could still be obtained. Trophic disturbance, such as the large bulla on the side of the heel, does not occur in Landry's paralysis : it is an indication of an intensely irritative state of the nerves, transmitted to them from the spinal cord through the posterior roots. Although the posterior nerve roots conduct impressions upwards they transmit nutritional, in- fluences downwards. This transmission is, indeed, probably by their own nutrition ; the disorder in their molecules passes to the tissues in which they end. This bulla, in connection with the other symptom, showed that there was an intense irritative inflammation in the spinal cord. It POSSIBLY LANDRY'S PARALYSIS. 123 was on the side on which the plantar reflex was abolished — that is, on the side on which the inflammation of the grey matter, in the lowest part of the cord, was most intense. The lad's condition did not allow of our ascertaining exactly the state of sensation. But, in the lessened movement of the face, of the mas- seter, and of the eyeballs, trifling as all were, there was evidence that the inflammation was spreading upwards into the pons. It seemed to have passed the medulla, ex- cept that the heart-sounds were unnaturally short, with a peculiar character, seldom met with except in acute dis- turbance of innervation, as if the cardiac centre were becoming affected. I have said that the opinion that he was suffering from Landry's paralysis was probably not far from the truth. There is no doubt that Landry's paralysis is due to an acute blood poison acting on the nerve centres, from below upward. All present knowledge leads us to ascribe an acute inflammation of similar course to a like cause. Both maladies occur under analogous conditions. It is probably only a question of a slight degree of difference in the nature of the blood poison, whether it simply arrests function and causes slight nutritional changes, or whether it sets up actual inflammation. Measles may be followed by typical Landry's paralysis, without evidence of inflam- mation ; and measles may also be followed by an intense spreading myelitis. We are learning to see more and more distinctly the profound causal influence of toxic blood states in producing inflammation of the spinal cord as well as of the peripheral nerves, and it is most important to recognize the fact that allied blood states seem to cause one, or the other, or both. At present, multiple neuritis has the hold on professional thought which novelty always entails, and central diseases are sometimes overlooked. I have lately read some descriptions of acute fatal " multiple I2 4 ACUTE ASCENDING MYELITIS. neuritis " (without autopsy), due to toxaemia, which I am sure were cases of ascending myelitis. Among these blood states which act on the nervous sys- tem some of the most important are those connected with acute specific diseases — the diseases that are due to an organismal virus. The question was put to me in this case, Was the lad's malady connected with the gonorrhoea? I do not know that anything of the kind has been hitherto observed as a result of gonorrhoea, but a causal relation between the two is highly probable. We know how such states are produced by various acute specific diseases ; and we know that the poison, the organized poison, of gonor- rhoea may induce a subsequent toxemic state which is manifested by the arthritis of " gonorrhceal rheumatism." Let us consider, for a moment, what this means. The organisms of such diseases seem to cause these con- sequences, in most instances, by producing a chemical organic poison in the blood, and analogy makes it highly probable that such an agent is the cause of the arthritis. But we know that such products can act on the nervous system, peripheral and central, inducing grave disturbance of function, and definite, often severe, inflammation ; it is therefore not surprising to find that in some individuals, especially predisposed, the gonorrhceal virus should cause an after-poison capable of producing the terrible effect which was seen in this case. Let me remind you of a fact, which I may have mentioned before and may have to men- tion again, because it is one of the cardinal facts of medical knowledge, throwing light on many problems, and, among others, on the influence of predispositions. It is the fact ascertained by the admirable researches of Dr. Sidney Martin on the process by which diphtheritic paralysis is produced. He seems to have conclusively proved this, — that this paralysis is due to a poison of organic, chemical nature; that the poison is not produced directly by the INDICATIONS OF SPECIFIC RELATION. 125 diphtheritic organisms, but that these generate a particular chemical substance of the nature of a ferment, which acts upon albuminous materials (" albuminoses ") in the body, especially in the spleen, and converts them into a poison which has this intense action on certain nerve structures. It is probable that the mechanism is similar by which many other specific organisms give rise to toxic agents. You can readily understand that if a person has previously been in bad health, as this lad had been, there should be some slight defect in the chemical constitution of such substances as the albuminoses of the spleen, etc., and a very slight defect may render the product of such a ferment different from that which would result in perfect previous health. A very slight difference in constitution, a little more or less of one element, may change atoxic substance into a harmless one, and vice versa. Thus we can under- stand how it is that the same causal influence shall have but little effect in one individual, and shall have a profound effect in another, especially in another who has been in conspicuously bad health. We can understand, for instance, that the organisms of gonorrhoea shall, in one person, give rise to a toxic chemical material which acts on the joints, and causes the arthritis of " gonorrhceal rheumatism," while, in another, the poison produced shall act on the spinal cord and set up acute myelitis, as in this case. The lad was in a state in which death seemed inevitable. In " acute ascending paralysis " (Landry's paralysis), as I have said, we have only an arrest of function by the blood state ; in inflammation we have also a process of organic change, in the vessels and the tissues, which tends to pro- gress even apart from its cause. The mere arrest of func- tion, once stopped, may not increase, and may be followed by steady recovery; but the effects of inflammatory damage must persist, and the process itself may continue for a time after the blood state which has caused it ceases to act. 126 ACUTE ASCENDING MYELITIS. Hence, I fear that the chances for the poor fellow's life are very small. I doubt, indeed, whether he is alive now. Although the expectation of saving life may be absent, and even the hope of doing so can find no room in reason, we are always under the compulsion of striving to achieve that which we may not be able to think possible. Our knowledge is never certain — even our possibilities may be incorrect, and so our hopelessness may be wrong. We must act in spite of our anticipations. Therefore, it will be right for you to ask, what did I suggest should be done for him ? As far as I have been able to see, there are only two agents which have any considerable influence upon toxic blood states of this kind. One is mercury. I believe the influence of mercury in syphilis is only one instance of its power over organisms. It is the most striking and perhaps the only one surely known to us, but it is probably not the only one, and is not likely to be the only one. Some years ago, a discovery was made in some researches at the Brown Institution which has been almost unnoticed since, — that the fatal effects of a certain inoculation (I think with the poison of anthrax) could be prevented by mercury. Its influence in other diseases of like nature is supported by many facts. That on inflammation, which few can doubt may be connected with its effect on blood states. So I advised that mercury should be rubbed in — a drachm of mercurial ointment every four hours. In such a case, however, you must "hit right and left;" there is not an hour to lose ; you must do anything and everything you can, always provided you do no harm. In septicaemia, such septicaemia as occurs after childbirth, for instance, the agent which has seemed to me effective, and which, as far as I could judge, has certainly saved life, is perchloride of iron in full doses. So I advised that every three hours 20 minims of the tincture of perchloride of iron should be given. The only other opportunity for treatment, beyond FATAL TERMINATION. 127 ordinary stimulation, was this. He might die from failure of the functions of the medulla; they might fail from the influence of disturbance of other related parts, or from an influence which was not an actual inflammation ; it was just possible that if they could be kept active he might tide over a period of danger. The measure that has seemed to me to have most influence in stimulating and keeping up the functions of the medulla, is the hypodermic injection of a small quantity of strychnia, one-eightieth of a grain, together with a very minute, stimulant dose of morphia, about one-thirty-sixth of a grain, repeated every two hours if necessary. Of course, if the failure of function were due to invasion by inflammation, this could have no effect. But there was no reasonable ground for hope that the treatment would be effective. The cause is more likely to be a chemical substance than organisms, and we have even less power over the former than we have over the living agents of disease. Had the symptoms been due merely to the restraining influence on function exerted by the blood state, we could have felt a slender hope that its maximum might have been reached. But inflammation, to whatever due, is a process with its own effects — proportioned to its initial energy, but locally independent of its origin — certain, at such a stage, to cause more destruction. Already at the limit of the functions essential for life, it could scarcely stay there ; the passage of that limit seemed inevitable, and with it the thread of life would break. Next week I shall doubt- less be able to tell you. [In his lecture the following week Dr. Gowers mentioned the sequel. The boy lived only six hours. A post-mortem examination, made by Dr. Arkle, revealed characteristic signs of myelitis, most intense at the place at which the inflammation had been supposed to begin — the lower part of the dorsal region, just above the lumbar enlargement. There, indeed, it had caused the front of the cord to be 128 ACUTE ASCENDING MYELITIS. bulged forward in an unusual manner by the swelling and softening of its substance. There was also some softening in the mid-dorsal and lower cervical region, where the grey- substance was hemorrhagic, almost diffluent. Culture- investigation for organisms has been commenced, and the pathological results of these and of the microscopical exam- ination of the cord, will be published by Dr. Arkle.] LECTURE IX. LOCOMOTOR ATAXY. I. Gentlemen : — I desire to-day to direct your attention to the disease known as locomotor ataxy. It is a subject with which you are all probably more or less familiar, but per- haps not so familiar as to make it superfluous for you to consider afresh some of its features and some of the lessons they convey. The subject is so large that it will be im- practicable to bring into the compass of a single lecture all the points to which I desire to direct your attention, and, therefore, I think it will be wiser for us to devote to it two of our weekly meetings. The subject is instructive on many accounts. The study of this disease supplies a use- ful example of pathological reasoning, and of the mode by which we discern the mechanism by which symptoms are produced. It is instructive also on account of the wide re- lation of its facts : those which you learn in considering it have an application to many other definite diseases, and to many other general morbid processes. It is also instruc- tive because it is a typical example of the class to which it belongs, — typical not only in its nature, but typical also in its treatment. If you learn these thoroughly you will have learned also much relating to other diseases. I propose to consider to-day the leading symptom of the disease, the ataxy, in reference to its mechanism — to con- sider it, not in confusing detail, but with perhaps superflu- ous deliberation. It is an advantage in clinical teaching, Delivered June 14, 1893. Clinical Journal, September 27, 1893. 12 129 130 LOCOMOTOR ATAXY. that information can be given in less compressed form than is possible in systematic lectures. In the latter, so many facts have to be described and discussed in such rapid suc- cession, that, in the mind of the student, they are apt to jostle one another out of place, so that, at the last, perhaps only a very few facts remain prostrate in his mind. I pro- pose to avail myself of this freedom, and to strive to fix in your mind only the leading facts relating to the subject. There are few diseases of which the name has given more trouble than that of locomotor ataxy. This name was assigned to the malady by Duchenne, with the additional epithet " progressive," an epithet which soon came to be dropped on account of its clumsiness, and the dropping has since been confirmed by the discovery that the adjective was inaccurate. The disease in a large number of cases is not progressive. The ability to recognise the early stage of the disease has revealed to us the fact that at least half the cases, I think more than half, are not, or need not be, progressive. But the term "locomotor ataxy " is itself some- what redundant., because the disorder is one of movement in general, and not merely of that which causes change of place. So " motor ataxy " would be as accurate as " loco- motor ataxy." But the term " ataxy," as a simple term, is only applied to disorders of motion, so even the epithet " motor " has been largely discarded, and the term " ataxy " has been used alone. And yet it has been found that even this is inaccurate, because the disease may exist in its slighter form without even a trace of inco-ordination. Hence the example has been widely followed, which was set in Germany, of using a still more general name, a term which had been before applied to various affections of the spinal cord, the term "wasting of the back," "tabes dorsalis." But, again, with the strange swing of the pen- dulum which seems inevitable in the progressive altera- tion of idea inseparable from the advance of scientific QUESTIONABLE NOMENCLATURE. .131 knowledge, we have found that the disease may exist when there is no affection of the spinal cord, and so it has been necessary to supplement the term " tabes dorsalis " with the term " peripheral neuro-tabes." I shall explain the necessity later on. It is rather surprising that in this difficulty more writers have not been disposed to fall back upon the name which has sometimes been applied to this disease, that of DncJicnne 's disease. The labours of that most distinguished and most energetic clinical investigator of diseases of the nervous system have been recognised by the ascription to him of the ownership of three maladies, — two diseases and one paralysis, — pseudo-hypertrophic paralysis, and locomotor ataxy are Duchenne's " diseases," and labio-glosso-laryngeal palsy is Duchenne's "paralysis." But this method of nomenclature has its disadvantages. It involves a grave burden upon the memory of the student, especially when it is applied to diseases and symptoms. It has not only inconveniences ; occasionally, in the admiration of one man, an injustice is done to others. I yield to no one in my admiration for Duchenne. Duch- enne's description of the disease first called general atten- tion to it, and the name he gave it was adopted. But what constitutes the discovery of a disease ? If to dis- tinguish the symptoms from those of other similar maladies, to affirm from physiological considerations that a definite structure in the nervous system is diseased in such cases, and that this disease is the cause of the symp- toms, and then, in the body of a patient, to find and demonstrate that very structure to be diseased, — if this is not the discovery of the disease, I would ask you, gentlemen, what is. And that was done for locomotor ataxy by an English physician at King's College Hospital, Dr. Todd, and done nearly 15 years before Duchenne's description of the disease was published, and probably 132 LOCOMOTOR ATAXY. many years before he had begun to suspect its special character. If locomotor ataxy is to be called by the name of any physician it ought to be called " Todd's disease." Let me now direct your attention to the general feat- ures of the cases that I have to show you, and then proceed to consider the leading symptom of the disease, its origin, and nature ; next week we will consider in de- tail some of the other symptoms, and some of the general practical questions connected with the affection. The first patient is a man aged 41. He can tell us of no apparent cause for the disease. But most of you know that there is one antecedent that must always be searched for — syphilis. His preceding history, however, affords no evidence of this disease. But those of you who were here a few weeks ago will remember that I in- sisted on the fact that the absence of a history of syphilis does not enable us to exclude the disease unless we can exclude the conditions of its causation. I need not re- peat the considerations which compel this conclusion. We are not able to exclude it here. Three years ago he began to suffer from tightness and numbness around the chest across the front, with some pain there. These symp- toms in the chest were followed by numbness in the feet, a year and a half ago, and then by shooting pains in the thighs, extending to the calves, heels, and arms, and accompanied by some unsteadiness in walking in the dark. He says also that some weakness of the legs came on ; but patients with this disease very often describe weakness of the legs when there is none. They mistake the diffi- culty of co-ordination for deficiency of power. Tardiness of micturition also developed, a symptom which is sometimes the earliest to attract the attention of the patient. Remember that many sufferers from diseases of the spinal cord first seek medical advice on account of difficulty of micturition ; unless we are aware of the fact, ILLUSTRATIVE CASES. 133 they may be treated for a long time, and treated unsuc- cessfully, because the real cause is not suspected. There was also transient double vision, — another early symptom of the disease, the relations of which we shall have to consider next week in greater detail. He has now, as you see, slight inco-ordination of the left hand and of the legs, greater in the left leg than in the right. When told to touch his nose with his fore- finger, he succeeds fairly well. With the left hand he can unbutton and button his waistcoat. If made to hold both his hands out, when his eyes are shut, there is very little spontaneous movement to be detected. When made to walk without the aid of his stick his gait is a little uncertain. When he puts his feet close together and shuts his eyes, you observe that he has difficulty in maintaining his balance. When asked to put his feet close together with his eyes open, he is just able to stand upright, not so steadily as is natural, and if then he raises his head up the tendency to sway is distinctly greater. In this position the base of support of the body is reduced in area, the muscu- lar contractions have to be more accurately adjusted to maintain equilibrium, and the defect in the power of adjust- ment reveals itself. When, after placing my hand in a certain position, I ask him, with shut eyes, to touch it with his left foot, very little inco-ordination is revealed. His defect shows itself chiefly in walking and in standing. There is no knee-jerk in either leg. A little impairment of sensibility to touch can be found, chiefly on the outer surface of the legs. Reflex action from the skin is defec- tive, and the reaction of the pupils to light is slight. In the second patient, the gait is much more defective than in the first. His toes are raised too high as he walks. When asked to stand still with his feet w T ide apart he can do so, but with the feet close together he has great diffi- culty in standing. He separates his feet to increase the 154 LOCOMOTOR ATAXY. base of support ; but even then the muscles are in constant action, so that the toes are every moment raised, because the over-action of one set of muscles is so great that it has to be compensated for by the contraction of another set, and this is also too great. Withdraw his visual guidance and he is quite unable to stand. After observing the posi- tion of my hand he cannot touch it with his left foot while his eyes are shut. His foot goes too much to the left. With the right foot he does better, but this also goes too high. He succeeds in touching his nose with the tip of his forefinger while his eyes are shut. When made to hold out both hands with his fingers extended, involuntary move- ments are seen, the action of the muscles which should keep the limbs at rest being disordered, like that of the muscles during movement. In this patient also there is no knee-jerk in either leg. The symptoms in this case are of six years' duration ; and the unsteadiness has been considerable, and accom- panied by shooting pains for five years. For three years he has had symptoms of unusual character in certain cra- nial nerves which also we shall consider next week. The muscles of mastication are affected. He has difficulty in approximating his lips and likewise some paralysis of the ocular muscles. In addition, he has deficient reflex action in the pupils, impairment of sensation in the limbs, and also in the face. In the other patient we could only recognise syphilis as a possible antecedent; in this, it is certain. We have dis- tinct evidence of its occurrence. Twelve years ago he had the primary lesion, followed by sore throat. These, then, gentlemen, are the chief features of the disease. We will now consider especially the characters of the leading symptom, that round which the others are grouped, as it were about a centre, and which, although not invariable, is vet the dominant element in the devel- ixco-ordixat: ::t: : -^ ; t In ab :ase= ever. vbier :b:= symarmm zzts net exis: mr.LLim ran be :ra:ea vLaiLv are :ar_abie abb- mately, of producing it. 7be ."-'- :■: .: ::..:::: ataxy v = vvy:::m which, as its name expresses, is in aspect a motor symp- tom. But it is not a motor symptom in its nature. It is r :: zee :: a zrimary ae ran verier.: : :' r::::: rr: :essea are motor structures. In a typical case, if you test the strength of the limb in which there is the greatest disorder of move- ment, you find no deficiency of power, and if you examine :: y: j vva r. : :ra:e ::' v.vvv v:.::: M: reiver : y : = :r_: are wbv vie rvvmsmre vie rr. :::■: =:r em.arei :be i : rrex n'rbe brair. Lie bares via: v ;; v i :he ivir.ia :be :eb = if :be aaieriir mrrvzz :f Lie :ir: Lie vires vv-: raas vrriayb vie ar.rer::: rer eriirs:: bze nemes ar. a ever, r. .:: r v iys in be ma = Le = y: fr.a r. : :ra:e ::' - 7 be viiii r sir z::z re= : v i y vv b it very different when you examine the sensory structures. Li ib : a ~ e 5 vi " b vi vie srbrai ::ra b zb"e::ea — v:i b bvey :::i5Vii:e :be yrea: maLrivy — -.ve vv: Lbease vv Lie :-. ;- ■swrp part of the cord. Wherever there is inco-ordination there is sclerosis in the posterior columns at the place at which the posterior nerve roots from the seat of the inco- irviLiv zzv- in:: vie ::::". zrz zi=: zeyererzLir :: :be ivvr.v nerve r: ::= Tie ex = :er:e b:.v zbezse vbzere Lee r irieriir vev.e r: i:s enter Lie ::rz ana ::" zeveverzbir in :ne :::: ; viewer .es i ryev Lva: bze esservbai ieLir ar.ae.Lyiry vie rvvvv if an in:errvr:i:n in. :be : : v: i ::- iry :"::.::. : v ::' :be yisreriir nerve n :r= Ave :ba: :::.- elusion is remarkably confirmed by some rare cases in which the posterior nerve roots have been damaged 05^ other causes. In one case especially (recorded by Dr. H j~be= Lev rev in vv:b vie r. aven: baa v: vv ve ram : zrs ariv many ::' ve river:: :::v Lvere v.-ere Lie vyvirni symptoms of locomotor ataxy. This case ye a: 136 LOCOMOTOR ATAXY. importance ; it establishes the fact that the symptoms of the malady we are considering may depend on a simple lesion. It affords us one of the firm facts of pathology by which we may ascertain the meaning of other complex phenomena. But the disease of the root-fibres is not the only lesion in tabes. All the way up the cord, above the place at which the root-fibres entering the cord are conspicuously diseased, we find degeneration of the posterior median columns — of that part of these columns that is adjacent to the posterior median septum, — the so-called " column of Goll." At the lumbar region, where, as a rule, the root- fibres are affected, this degeneration of the posterior median columns widens out, so as to blend with that at the en- trance of the nerve roots. That is because the fibres which ascend in the median columns reach these from the roots by passing through the external parts of the posterior columns. It is because the median parts are composed of root-fibres that pass to them without interruption, and ascend in them without decussation. Hence disease of the lumbar roots necessarily entails the degeneration of the median columns, and its extension outwards to the root region of the affected fibres. The fibres degenerate up- wards, and do so in the same way if the lumbar roots are destroyed by any other morbid process. It is thus certain some root-fibres ascend the cord without interruption, and that these fibres are diseased in tabes. That is the first great pathological fact for you to remember. Now, leaving this point for the present, what evidence can we discern as to the way in which disease of the nerve roots causes the symptoms of this malady ? Certain clini- cal facts, when taken together, have a very clear pathologi- cal significance. First, the leading symptom, ataxy, exists, in its typical form, and even in high degree, in many cases in which there is no impairment of any form of cutaneous AFFERENT IMPULSES BY COMPRESSION. 137 sensation. Further, in agreement with this, we have the fact that when there is impairment of cutaneous sensation there is no proportion between it and the amount of inco- ordination. It is, therefore, certain that the disease of the posterior nerve roots does not cause ataxy by interrupting the impressions conveyed from the skin. Yet it is also certain that the symptom is due to the interruption of some impulses which these roots convey. What other afferent impulses pass to the cord by these roots ? They are the impulses from the deeper structures, from the muscles especially, also from the fibrous struc- tures, from the fasciae, perhaps from the tendons, and cer- tainly from the joints. Remember that we must not con- found afferent impulses and sensory impressions. It is probable that the afferent nerves are continually being tra- versed by impulses which do not act directly on conscious- ness, and do not, when in normal degree, produce a true sensation. We are unconscious of any sensations in our muscles at the present moment. But a vigorous compres- sion of any muscle will at once convince you that afferent impulses can be readily generated in it ; so also will an attack of nocturnal cramp, and so also will the process of going upstairs on the following morning. In cramp of the calf, the afferent nerve-fibres within the muscles seem to be stimulated by their compression in consequence of the nar- rowing and widening of the strongly contracted muscular fibres, and when the nerves are rendered hyperaesthetic by that disturbance, they are stimulated in a very marked de- gree by the tension to which they are exposed when the muscle is extended in mounting stairs. ... So that we have evidence that afferent impulses are generated in these fibres by compression during contraction, and by tension during the passive elongation which the muscles endure on every movement. We can understand these facts by the mode in which these afferent fibres arise in the 138 LOCOMOTOR ATAXY. muscles. According to the investigations of Tschiriew, they commence in the interstitial tissue between the fibres, and must thus be equally exposed to the mechanical effect of compression or of tension. Thus, by the exclusion of cutaneous impulses we are driven to refer the origin of ataxy to the interruption of these deeper impulses. We can perceive more of the features of the impulses from the muscles, than of those from other deep structures, and the large extent of the muscular tissue makes its probable that they constitute the chief part of the deeper impulses. Moreover, the phenomenon of muscular co-ordination sug- gests that they, rather than the impulses from the joints, etc., take the chief parts in its regulation. This is con- firmed by the constancy with which the knee-jerk is lost in locomotor ataxy. Whatever is the precise mechanism of the knee-jerk, it certainly depends upon a reflex process in which the afferent impressions proceed either from fibrous tissues or from muscles, and probably chiefly or exclusively from the muscles. But the constancy of that loss confirms the conclusion that it is by the interruption of these afferent muscular impulses that the ataxy is produced. What is the significance of the degeneration in the pos- terior median column ? We find it whenever the posterior nerve roots are destroyed. Moreover, if the nerve roots are destroyed on one side only, the degeneration of the posterior median column is on that side ; this and the facts already mentioned show, as I have stated, that nerve-fibres pass from the posterior root, through the posterior column, into the posterior median column on their own side, and do not cross. And as this degeneration of the posterior median column is constant in locomotor ataxy, we cannot doubt that the fibres which constitute it are some of those the disease of which gives rise to the symptom in question. And that is confirmed by two other very interesting facts. If that posterior median column is divided on one side in a MECHANISM OF CO-ORDINATION. 139 unilateral lesion of the cord, the effect is to cause inco- ordination in the muscles of the same side; whereas it is found that an extensive unilateral lesion causes loss of pain and sensibility on the opposite side. Further, in advanced tabes we commonly have a very marked loss of the muscular sensibility : the muscles may even be insensible to pressure or extension, thus confirming the conclusion that there is disease of their nerves, and supporting the opinion that this is the lesion which causes the characteristic ataxy of developed tabes. Where do these posterior median fibres go ? You must often have been struck by the similarity between the mode of standing and walking in some cases of ataxy, in which the power of maintaining equilibrium is especially de- ranged, and that of a patient with cerebellar disease. These posterior median fibres go up to the posterior median nucleus of the medulla; and it has been proved that there are many other fibres proceeding from the cells of that nucleus to the middle lobe of the cerebellum, the part of the cerebellum that is alone concerned in co- ordination. The probability is thus very great that these fibres conduct impulses which reach the middle lobe of the cerebellum and guide it in its co-ordinating function. So that we seem to perceive two processes in the mechan- ism of co-ordination : (1) Impulses from the muscles go to the cord; some of these must enter the grey matter to subserve the reflex action on which the knee-jerk depends, and they probably regulate a series of subsidiary co-ordinat- ing mechanisms. When you stand, you stand by a volun- tary impulse, but it is probable that the voluntary impulse acts through a mechanism in the cord which determines the arrangement of the movements for standing. The action of this mechanism is conspicuously seen in cases in which disease at the middle of the cord cuts off the lower part. Then, as you well know, spastic paraplegia is pro- Ho LOCOMOTOR ATAXY. duced, and the extensor spasm perfectly reproduces the condition of the limb in standing. That extensor spasm, which is attended by a gradually increased knee-jerk, is apparently dependent upon arrangements which are acted on by these impulses from the muscles. (2) Impulses pass up the cord, and probably reach the cerebellum, and through the cerebellum they may act on the motor cerebral cortex, and influence the relative energy of its cells. Thus we have this double reflex process of co-ordination from the muscular impulses, one by the cord, the other through the cerebellum. When a sensation of pain is produced in a muscle by a strong impression on its nerves, it is probably through those that terminate in the cord ; but we have no direct proof of this, and it has no special concern for us in our present problem. We perceive, then, that muscular inco-ordination is due to loss of impulses from the deeper structures, chiefly from the muscles. If there is a total loss of all the impulses, the muscle reflex is abolished, the knee-jerk is lost, and we have extreme ataxy. If there is only a loss of the impulses that pass up to the cerebellum, when there is disease of the posterior column above the lumbar enlargement (as in ataxic paraplegia), the knee- jerk is not lost, and we have only a moderate degree of inco-ordination, manifested especially in the defective main- tenance of equilibrium, and almost identical with that of cerebellar disease. I mentioned, just now, that the spinal cord is not invari- ably affected in tabes. Exceptions to the common rule are extremely rare, but their occurrence is well established. When the cord is free from disease the peripheral nerves have been found to be degenerated. They suffer, indeed, in a large proportion of the cases in which there is degen- eration of the root-fibres ; the peripheral endings are altered in chief degree. In the cases in which the cord is unchanged this lesion of the nerves, usually concurrent, is THEORETICAL PROBLEMS. 141 isolated, and the sole cause of the symptoms. These do not differ from the symptoms of cases with the usual lesion. Hence, we are justified in concluding that the nerve-fibres, the disease of which gives rise to the characteristic symp- tom, ataxy, are the same in both forms. The nerves which are affected first and most in these cases, as in the others, are the nerves of the muscles. So that, whether we have disease of these muscle nerve-fibres in the posterior nerve roots and posterior columns of the cord, or disease in their peripheral extremities, we have precisely the same condi- tion ; if the mechanism is the interruption of the conduct- ing fibres, the effect must be the same, wherever in their course the lesion is situated. Hence, in " peripheral neuro-tabes," as the rare form is termed, — locomotor ataxy without disease of the cord, — we have a condition precisely similar to that which depends on the spinal lesion. Hence, also, there is ataxy similar in its features in the other forms of degeneration of the peripheral nerves, such as are pro- duced by Alcohol and by Arsenic. It may even be im- possible, from the symptoms themselves, as far as the ataxy is concerned, to distinguish a case of true locomotor ataxy from one of what is called " alcoholic pseudo-tabes." I hoped to have gone over to-day other points regarding the nature of the lesion, but it will be better to postpone these until next week. I must be content to-day if I have succeeded in giving you a clear conception of the me- chanism of the special symptom of the disease, or at least of the more salient elements of the evidence regarding its nature. I would only point out to you, in conclusion, how interesting is the enlargement of our ideas regarding this disease which has come from modern research. There is much that we cannot understand concerning the precise mechanism of co-ordination and of its defect; nevertheless, we have a glimpse into the process, though we cannot surely perceive all its outlines. It is certain that the 142 LOCOMOTOR ATAXY. afferent nerves of muscle are stimulated by the changes which they must endure in the two states of the muscle ; the muscular contraction when the muscle is active, and the tension it undergoes when its opponent is acting, alike generate afferent impulses. These we do not ordinarily perceive, except in the vaguest manner. They are doubt- less concerned in the production of our conceptions of posture and movement, but we only perceive them as sen- sations, when from some cause they become excessive. But the}- must be constant, in unperceived degree, influ- encing and guiding the central mechanism. Consider the subject over at your leisure, and I think it will become clear to you, and that you will find assistance from it in understanding many processes and many prob- lems to which I have not ventured to refer. LECTURE X. LOCOMOTOR ATAXY. II. You may remember, gentlemen, that last week we studied at considerable length the chief symptom of loco- motor ataxy and its mechanism. To-day we will briefly review the other symptoms of the disease. They are, how- ever, so numerous that I shall have to content myself with a mere glance at those that are of chief importance and at their relations, considering in more detail some that are illustrated by the cases we examined, or that are of most significance, or with which you are not likely to have become familiar. The morbid process which, as you saw in the microscopi- cal sections, is indicated by an overgrowth of the connective tissue, is essentially a degeneration of the nerve elements, beginning in them. Such degeneration has many causes. It seems sometimes to be the result of a mere inherent de- fect of vitality ; sometimes it is a local manifestation of senility; sometimes it is due to the influence of general conditions, depressing emotion, and like influences ; but in the vast majority of cases it is the result of the action on the nerve elements of a toxic agent. We see it as the effect of chemical agents of inorganic nature, and also as the effect of certain organic chemical agents which are produced by minute organisms — such, for instance, as alcohol, which is due to the growth of minute organisms outside the body, and materials of unknown composition produced by the Delivered June 21, 1S93. Clinical Journal, October, 1S93. 143 144 LOCOMOTOR ATAXY. minute organisms which constitute the virus of some acute diseases. Such organisms cause their own direct effects, but they also leave behind them in many cases such a sub- stance as I have referred to, which acts upon nerve elements as inorganic poisons may do, and causes degeneration. And this, which has been proved to be true of the virus of some acute specific diseases, seems also to be true of the chronic specific disease which is so frequent an antecedent of ataxy, syphilis. This was, you will remember, certainly traceable in one, and possibly in the other, of the patients whom we saw last week, and whom we will presently again examine. As a fact, this antecedent is to be ascertained in more than two-thirds of the cases ; and it has probably preceded tabes in at least four-fifths. Such a relation means a causal influence in the majority, for mere coinci- dence will explain only a small proportion. At first the statement of these facts was received with general scepti- cism. But we have to be careful, in science, how we deny. Some denied the possibility of the causal relation, but they have had to reconsider their denial ; fresh observation of facts yielded evidence of overwhelming cogency. One rea- son why the relation was doubted was because it was found that anti-syphilitic remedies did no good to the disease; and the fact is unquestionable. But the mystery is ex- plained when we perceive that only on the lesions which are the direct result of the syphilitic organisms does anti- syphilitic treatment exert its influence ; on those lesions which are the result of the toxic material produced by the organisms, the anti-syphilitic treatment has no influence. This material has not been isolated, but the analogous material has been isolated in some acute diseases dependent upon organisms ; and so close an analogy can be traced between the whole series of such diseases which have after effects, that we may feel reasonably sure that it is in this way that syphilis produces the disorder we are now con- OPERATIVE EXPERIMENTS. 145 sidering. It is not possible for me now to describe to you the evidence, but it is, in my opinion, convincing. In these facts we have the secret of the inutility of anti-syphilitic treatment in cases of locomotor ataxy, in which the disease is, nevertheless, the sequel of syphilis. You will remember that we traced the chief symptom of tabes, the ataxy, simply to the interruption of the muscle nerves and the arrest of the afferent impulses which guide alike the spinal and the cerebral motor centres. You will also remember that I pointed out to you that, in this mechan- ism, it makes no difference in what part of the nerve fibres the interruption takes place, and that when the ends of the nerves alone are diseased, the effect is the same as when the same fibres are diseased in the nerve roots or within the cord. A curious historical fact may be mentioned in this con- nection. In what may be termed " medical surgery," from time to time, some operative procedure becomes fashionable ; it is adopted on all sides, partly on the strength of an in- itial laudation, and partly because it is something to " do " where little had been effected. After a time, if it possesses no real value, it sinks into desuetude. Of late we have had an epidemic of " suspension " for this malady. A little before, it was nerve stretching which was said to do remark- able good in cases of tabes. The curious fact to which I allude is this, that the first patient who seemed to derive benefit from nerve stretching was about to undergo the pro- cedure a second time, but died under the anaesthetic before the operation could be performed, and so afforded patholo- gists an opportunity of correlating the operation and the morbid process. The operation, unfortunately, was only partially utilised. It was found that there was no discerni- ble disease of the spinal cord or of the nerve roots. The peripheral nerves were not, in that case, examined ; they have been examined in others and found diseased in so 146 LOCOMOTOR ATAXY. many, that there can be little doubt that the case I have just referred to was one in which the degeneration was at the periphery of the nerves. I mention this fact to impress on your minds the occurrence of this peripheral change. It has been found to be exceeding common, if not constant, in tabes, although it is very rarely confined to this position. If it is the only lesion, there is no disease in the posterior roots. The disease above the ganglia seems to be an inde- pendent, coincident, degeneration in the same fibres. The disease is not limited, however, to the afferent mus- cle nerves (to which I especially directed your attention last week), although their affection seems to be the primary and essential element. The nerves from the skin suffer in a large proportion of the cases, and, in consequence, loss of sensation to touch and to pain and to heat (usually to some, occasionally to all) is very common, and is distribu- ted according to the position of the nerve lesions. The anaesthesia, like the ataxy, may be due to disease either of the nerve roots in the cord or in the periphery. It is found sometimes in the arms, and there is, occasionally, a similar loss in the cutaneous sensibility of the head. We may have anaesthesia in the region of the fifth nerve ; and de- generation of the optic nerve, as you know, is not rare. The latter is a typical nerve degeneration, involving the peripheral part of the nerve, and is a visible type, in a nerve of special sense, of that which occurs in the sensory nerves of the limbs. The process of degeneration is one of molecular change. But the function of the sensory nerves is carried on by molecular change. It is thus that they perform their mar- vellous function of transforming, for instance, the mechani- cal force of motion, in a touch, into nerve energy, or the thermic force, which is another form of motion, into nerve energy. And the molecular changes of degeneration seem to give rise to nerve impulses similar to those which are A -i: : V MALADY. 147 produced under normal circumstances, — similar, I mean, as regards their general character, not as regards their special occurrence. Hence we have the symptom, so famil- iar to you, of the pains of tabes. Most common are the peculiar, sudden, sharp, " lightning " pains ; but duller pains are very common ; and they are often curiously " rheumatic" in character, so that they are almost always at first mistaken for rheumatism. So, indeed, with less excuse, are the sharp momentary pains. This mistake is facilitated by the fact that change in weather, especially change to cold, has a remarkable effect in increasing these pains, just as it does the pains of true rheumatism, and many patients have been treated for several years as the subjects of rheumatism or of gout, whose pains were the result of this malady. These are felt most often in the legs, but also sometimes in the trunk and sometimes in the arms, and occasionally with great severity even in the head. In the head and trunk they are apt to be mistaken for sim- ple neuralgia, and even thought to be the indication of encephalic or abdominal disea e Thus, these being the essential symptoms of the dise- ise it is by its nature a sensory malady; it is an affection of the sensory structures. Its dominant feature, the inco- ordination, is, as I told you, motor only in aspect. The cardinal symptoms of the disease are the result of changes in afferent paths. They are, however, often associated with subordinate symptoms which are actually motor. All toxic agents which may act, primarily and chiefly, on sen- sory structures, can also act, to some extent, on motor structures, and some of the agents often act on these in a preponderant degree. Among the motor structures which are most often impaired in tabes are the nervous structures for the bladder, sometimes those of the sphincter, but more com- monly those of the wall of the bladder. The wall of the 148 LOCOMOTOR ATAXY. bowel may suffer also and give rise to constipation ; this is inconvenient, but the weakening of the wall of the bladder involves definite danger. In consequence, the bladder is not emptied ; there is residual urine, and that which descends from the kidneys overfills the bladder ; the con- stant overfilling necessarily weakens still more the wall, because when overdistended its muscular power is rela- tively more incompetent, and it is less and less able effici- ently to contract. Chemical changes occur in the residual urine, and cystitis is excited, with a secretion that increases the alteration. The increased pressure in the bladder, thus gradually augmented, acts back on the kidneys, interferes with their function, and ultimately induces disease in them of the nature familiar to you as surgical kidney. The blood, from the first, is not properly relieved of effete material, which seems to become, at last, actually toxic, and a state of imminent peril to life is often thus induced. Of all this chain of effects the patient may be unaware. The overfilled bladder indeed often overflows, and incon- tinence reveals the condition. But overflow does not always occur, or occurs so sel- dom as to attract little notice, and lowered sensitiveness may aid in preventing any consciousness of what is wrong. Remember that you must never take a patient's statement as true that he empties the bladder. Patients are con- stantly under the impression that the bladder is emptied, when you may easily ascertain by percussion over the pubes that a considerable amount of urine remains in the bladder after micturition. This evil may reach a high degree in a patient who is in the early stage of the disease. It is the chief way in which locomotor ataxy causes death; and death in this way can always be prevented. When- ever you suspect residual urine, ascertain whether your suspicion is correct by passing a catheter. Remember that it is absolutely essential for the bladder to be perfectly EFFECTS ON THE URINARY SYSTEM. 149 emptied at least once a day; and if there is any decompo- sition of the urine or distinct muco-pus in it, the bladder must be daily washed out with a disinfectant. The effects of the retention on the kidneys and their function when the patient has been brought to the edge of disaster, are sometimes precipitated by the reflex effect of the passage of a catheter; then the patient is said to have "catheter fever," but the condition would have developed spontane- ously in a few days, and is due essentially to want of the previous use of the instrument. Only two days ago I saw a gentleman whom I saw also a year ago in the early stage of this disorder. At that time he was having mysterious attacks every few weeks, in which considerable pyrexia lasted for several days. The doctor who was attending him had been greatly puzzled by these, and suggested various theories to explain them ; but I found the patient was not emptying his bladder. Such attacks of pyrexia in that association are generally due to the interference with the action of the kidneys and the effect on the blood. The patient commenced the use of the catheter, had one more slight attack of pyrexia, had since used the catheter every day for a whole year, and had not had another attack. I ought to add, to avoid misconception, that incon- tinence is not always due to overdistension, any more than overdistension always causes incontinence. In many diseases of the spinal cord you have paralysis of the sphincter, urine is always escaping, and the bladder is always almost empty. In some others, the bladder is peri- odically emptied perfectly, by a pure, involuntary, reflex, or automatic process. But in locomotor ataxy incon- tinence is generally due to overfilling. Of other motor symptoms, the next in importance are those with which we meet in the eye, of which the most common is the loss of the reflex action of the iris to light. ISO LOCOMOTOR ATAXY. This seems to be due to changes in the centre for this movement in the nucleus of the third nerve. Instead of this loss, or with it, there may be loss of accommodation, or loss of the action of the iris associated with accommo- dation. Frequently there is weakness of external ocular muscles, sometimes amounting to complete paralysis. Such weakness is usually at first transient, but after several successive attacks, which last longer and longer, the loss of power finally becomes persistent, and it may be present in muscles of both sides. All these ocular symptoms are probably due to central degeneration in the related nuclei. There is one important fact connected with this loss of action of the pupil to light. It is exceedingly common in tabes ; tabes generally has syphilis for its antecedent, and this is also true of the isolated loss of reflex action to light, apart from the other symptoms of tabes, or even from any symptoms of disease of the nervous system. Whenever you meet with it as a persistent, isolated symp- tom, you will always be justified in suspecting that the patient has had syphilis, and in most all cases you will find the suspicion verified. This indication is often of very great importance as putting you on the track of syphilis when you would not otherwise suspect it. Other cranial nerves rarely suffer in tabes ; but in one of the patients I show you, by the courtesy of my friend, Dr. J. Taylor, under whose care the man is, there is a remarkable affection of the motor cranial nerves ; not only the eyeball muscles, but the muscles of mastication are weakened, while the face, palate, and larynx are also all more or less paralysed, one vocal cord being motionless. Remember this case as an example of the wide extent of cranial nerve-palsy we sometimes meet with. There may be some actual loss of power in the limbs, EFFECTS ON NUTRITION OF TISSUES. 151 but only in late cases of true tabes. It is due to the fact that, ultimately, even the spinal motor structures may occasionally suffer ; but such weakness is not part of the malady in its common form. Allied to the pains (which, I should have added, are often curiously paroxysmal) we have other paroxysmal symptoms which I need only mention — those which are called " visceral crises " — gastric pain and vomiting, laryn- geal spasm, nephritic pains, rectal pains. These I can only mention to-day. But the affection of the sensory nerves has often another and very important consequence. The sensory fibres of the posterior roots are those which govern nutrition in all the structures except the muscles. We do not know how they govern nutrition ; indeed we can only say that their own nutritional state seems to determine that of the tissues, so that the latter is dis- ordered when the nerves are actively diseased. If they are the seat of acute inflammation, such as may be pro- duced by passing a thread with some irritating material through the nerve, we have acute changes in the skin, but when there is merely chronic disease, slow, and without irritation, we have only a gradual thinning of the skin and wasting of the bones. For instance, such disease of the nerves of the arm gives rise, in the finger-tips, to thinning of the skin, and distinct narrowing of the ends of the phalangeal bones, and often also changes in the joints, by which adhesions form. When you have a peculiar irrita- tion of the nerves, such as gives rise to the severe pains of tabes, changes may occur in the nutrition of the skin. The irritation of the nerves is shown by the intense hyper- esthesia that developes at any spot in the skin to which the pains are referred. The influence on nutrition was curiously shown by one patient who had such localized tabetic pains in various spots on the hairy scalp. After the pains had lasted for a (qw days every hair broke off 152 LOCOMOTOR ATAXY. near its root, and when the pains ceased, the growth of the hairs became natural. Among the most common trophic changes, however, are those of the bones and joints, when, instead of wasting, as they sometimes do, they enlarge, and new bone forms in an irregular manner. This may be due to an irritation, rather than to a mere defect of nerve influence. Its occurrence is an interesting illustration of the degree and extent in which the deeper nerves suffer in this disease. One other trophic disturbance is both common and of great practical importance. It is the tendency to deep "perforating ulceration" on the toes, consequent on slight injury. It may reach the bone and lead to necrosis. These ulcers are often set up by the deep cutting of a corn. Always impress upon a tabetic patient that no corn should ever be cut ; it may be softened by an alkali, and rubbed with sand-paper, but should never be cut. The symptoms of tabes are thus, you see, numerous and varied ; and according to their grouping they present very different aspects. If you remember the leading features you will seldom have any difficulty in arriving at a diag- nosis. But there are certain special sources of error in diagnosis which it may be well briefly to point out to you. I cannot attempt even to glance at all the elements of the diagnostic problems presented by the disease, and must limit myself to a few which are either of special importance, are instructive, or are likely to escape your notice until the occasion arises for using the absent knowledge. In patients who are in the first stage of the disease, before ataxy has developed, and in many of whom it never does develop, uncertainty is sometimes caused by the limitation of the chief disease to the dorsal region. The pains are then felt in the trunk, and may be thought, and often are thought, to be due simply to intercostal neuralgia. But loss of sen- sation accompanies them ; and if you examine the legs DIFFICULTIES OF DIAGNOSIS. 153 carefully you will always find the knee jerk absent, even when there is no anaesthesia of the limbs, and no pains are referred to them. Such a state has also been mistaken for growth in the spinal bones, but in this, the terrible pain is especially related to movement of the spine, and that rela- tion is never present in tabes. Other difficulties in the recognition of the malady may be due to the unusual character of the symptoms, or to the preponderance of special symptoms which may bring the patient under treatment. Many ataxics come to a doctor first simply because they cannot easily pass water. When- ever that is the case, without local cause for the difficulty, always suspect this or a similar condition, and test the knee-jerk. In other cases the gastric crises have been thought to be due simply to primary gastric disorders. These difficulties are chiefly due either to imperfect knowl- edge or to incomplete examination. But there are other and more excusable diagnostic difficulties which arise from the occasional difficulty of distinguishing tabes from allied diseases. There are other morbid states which present almost the same symptoms. Especially is this true of chronic multiple neuritis to which I have already referred. When due to alcohol or arsenic it may be manifested by pains similar to some of those of tabes, as well as by loss of cutaneous sensibility, and also by loss of the knee-jerk, and distinct, even conspicuous, inco-ordination. This is true, as I have also mentioned, of the ataxic form of dipth- theritic palsy, although, in this, pains suggestive of tabes are usually absent. But in all these cases the chief indi- cation depends upon the detection of the cause, which you will seldom have difficulty in discerning, if only your scrutiny is properly directed by a knowledge of the possi- bility. Often also your opinion will be supported by the loss of pupil-action to light, which is so common in tabes, 154 LOCOMOTOR ATAXY. but which is very rare in the maladies that may be mis- taken for it. Another class of diagnostic difficulties is due to the fact that there are cases which present all the symptoms of tabes, but others in addition. The most frequent and most important of these are cases of general paralysis of the insane. As a matter of fact, these patients have tabes ; not only are its symptoms present, but its lesion exists ; but they have also the cerebral degeneration which causes the special symptoms of general paralysis. It is not a question of distinction; it is a question of recognising the combina- tion. As a rule, also, the common antecedent of pure tabes, syphilis, has also preceded the combined malady. Thirdly, there are cases which present some symptoms of tabes, only others are absent and are replaced by symp- toms of a different character. In the malady which we term ataxic paraplegia there is conspicuous unsteadiness of movement and in standing, but there is weakness as well ; the knee-jerk is not lost.; pains are slight; the pupil is not affected ; and, as the weakness increases, there is an increas- ing tendency to extensor rigidity in the legs; and, with less power of movement, its disorder necessarily sinks into the background. This malady is due to a combination of scle- rosis in the lateral columns of the cord, and in the posterior columns in the dorsal region, involving especially the as- cending fibres of the posterior median columns, the fibres which, as we have seen, probably convey impulses to the cerebellum. It does not involve the fibres which convey impulses to the grey matter, at the level of origin, which subserve the knee jerk. So the knee-jerk is not lost; the irritation of the nerves, which causes the pain, and their interruption, which causes the anaesthesia and the extreme ataxy, are absent. Wanting these features of tabes, the cases present the loss of power that is absent in typical and UNSATISFACTORY SYMPTOMS. 155 early tabes, and a sufficient examination, with adequate knowledge, renders the nature of the malady clear enough. In cerebellar tumour there is unsteadiness which reminds us of some cases of ataxy, and of the cases that I have just described. But there are not the other leg-symptoms of tabes, with one curious, but not frequent exception. Oc- casionally, the knee-jerk cannot be obtained in cases of cerebellar tumour. It is a curious fact, but we cannot yet give a satisfactory explanation of it. It is especially curious on account of the association with unsteadiness of move- ment in both diseases — an association which is probably not accidental, but has a significance at present obscure. In cerebellar tumour there is no anaesthesia in the limbs ; very often, instead of an absent knee-jerk, this is increased, and there is even a foot-clonus from pressure on the pyr- amidal tracts ; headache and vomiting are conspicuous features, and the latter does not occur in isolated "crises," as in tabes. Lastly, optic neuritis is extremely common, and atrophy, if found, is distinctly that which succeeds in- flammation, and quite different from the atrophy of tabes. Lastly, some hysterical girls present some unsteadiness in walking, and the question of this disease has arisen. But they have no other symptom of tabes. The ataxy is trifling and limited to the maintenance of equilibrium. It may not be easy for you to conceive that such patients should be thought to be suffering from tabes, but this error is sometimes rendered more easy by the fact that no knee- jerk is obtained. In many cases of hysteria, and, indeed, in many perfectly healthy persons, the knee-jerk seems to be absent ; sometimes many attempts fail to obtain it. This is, however, simply due to the fact that the muscles are not relaxed, and a very slight voluntary contraction of the flexors prevents the jerk. I believe the knee-jerk is never absent unless there is structural disease in muscle, nerve, or cord, or (rarely) within the skull. Some fifteen years 156 LOCOMOTOR ATAXY. ago I published a number of cases in which it was absent under normal conditions. All I can say now is that I should like, very much indeed, to examine these patients again, because I have no doubt whatever that I should now succeed in eliciting the jerk in every one of them. So, when I am told that the phenomenon is absent in a healthy person, I venture to doubt the fact, and in doing so I only treat the observations of others as I treat my own past observations. If the subject shuts the eyes, hooks the flexed fingers of the hands together and pulls on them, the jerk generally becomes distinct, though before it may have seemed absent. In a few, it may be necessary to let the legs hang vertically over the edge of a table, and it is always well to place your fingers on the hamstring tendons to ascertain their degree of relaxation. Yet, with all these precautions, I sometimes meet with cases in which I cannot obtain it on repeated attempts, and yet, next day, it is pro- duced in such typical form as to prove that its absence had been apparent only. Regard then with doubt even your own observations on cases in which there are no other symptoms usually associated with its loss. What of the prognosis of this disease ? First, remember that the malady involves very little danger to life, however extreme it may become, save by exhaustion due to laryngeal or gastric crises, or by the insidious development of some disease, such as pleurisy, which may not cause the warning pain that it produces with normal nerves. But one frequent and serious danger to life is directly related to the disease — that from the affection of the bladder, and impairment of the functions and structure of the kidneys, which I have already described. This danger may be prevented. On the other hand, unfortunately, actual recovery is even less likely than is death. The prospect, however, is not so gloomy as this state- ment may suggest. I speak of " recovery " in the medical PROGNOSES 157 and accurate sense of the word. Short of the disappear- ance of all the symptoms of the disease, they may and often do lessen to a degree that is practically unimportant to the patient, and they still more often cease to increase, and, though continuing, are tolerable. In the majority all cases, the disease, carefully treated, is not progres- ive This is true at whatever stage it is met with. Only when the symptoms are rapidly or steadily increasing (especially if this is in spite of treatment) must you be apprehensive that the affection will go on to grave disability-. On the other hand, in even.- stage you are not justified in holding out to the patient the anticipation of more than arrest and a slight degree of improvement More may be achieved, but you cannot reckon on it In the early stages, arrei: is almost tantamount to recover}', so far as the patient's con- sciousness is concerned. You cannot say that the patient has actually recovered, because if you test the knee-jerk even- year for twenty years, you will probably find it still absent. But, happily, the loss of the knee-jerk is not a symptom which enters into the patient's consciousness un- less : 5 assisted there by medical help. The pains, it is true, are not likely to leave consciousness untroubled, but they often become trifling and can generally be relieved. Remember, moreover, in connection with the pains, this very important fact, that tabetic pains, however severe, do not necessarily mean continued increase of the disorder. They seem, once excited, to go on as a neuralgia dees — without the morbid process becoming greater. The- are merely a persistent effect of the persistent alteration of molecular nutrition. It is unfortunate that, in considering any disease, we are compelled to leave to the last the subject that is first in importance — its treatment. I have already told you that, in spite of its relation to syphilis, nothing is to be looked for from anti-syphilitic treatment. In very rare cases :: 158 LOCOMOTOR ATAXY. early stage and rapid cause, iodide of potassium has apparently done some good. In many of these its influ- ence has been combined with that of other agents, and it sometimes does good in diseases that are not related to syphilis. Certainly your experience will soon demonstrate to you the accuracy of the general statement I have made. But if nothing is practically to be looked for from anti- syphilitic treatment, much is to be dreaded from its ener- getic use. When a patient has passed the stage of active syphilis, and is suffering from the sequelae due to its re- sidual influence, such treatment, and especially mercurial treatment, seems to intensify the tendency of the nerve elements to degenerate. I have seen some disastrous cases, in which a grave affection of this character has immedi- ately followed a thorough course of mercurial treatment, what they call abroad a " cure " — a term which bears to us more unconscious irony than it does to those who employ it. True, in many cases it may be wise to give the patient, for a few weeks, some iodide of potassium, because he is sure to be told that he ought to have had it, and it does not produce the harm of thorough mercurial treatment. It is often well to clear the therapeutic ground of any pos- sible source of future, though baseless, feeling that any- thing has been neglected, provided the measure involves no harm to the patient. Moreover, that which is true of full doses of mercury is not true of small doses. There seems to be some truth in the assertion that very small doses of mercury have a tonic influence on patients who have had syphilis, but who have ceased to be actively syphilitic. Such doses may, therefore, often be wisely given in com- bination with other agents. But, in treatment by drugs, it is essential to realize the fact that the state and process you are striving to modify and arrest is essentially a degenera- tive process, a process of defective nutrition, tending to structural decay. As far as we can discern, it presents no TREATMENT. 159 difference from the similar condition that is due to other causes. Further, the medicinal agents that sometimes exert a distinct influence on the process, seem to do so irrespective of its causation. Among these agents the first place must be given to arsenic. Its influence is, in many cases, conspicuous to no one more than to the suf- ferer. It is very interesting to note that we may trace a reason for the effect in the fact that arsenic has been proved to influence the nutrition of the peripheral nerves, especially the afferent nerves of the skin, which conduct sensation, and which also, as we have seen, determine nutrition ; probably it has a similar influence upon the deeper afferent nerves. It has to be continued, however, for two or three months, and then, after one or two months omission, it may, with advantage, be resumed. The form may be either the liq. arsenicalis, or liq. arsenici hydrochlorici, in a neutral or acid mixture, respectively, or the arseniate of soda in a pill. Phosphorus, which has so many points of similarity to arsenic, may be given alter- nately, and strychnia or nux vomica may be combined with either. Next in value to arsenic I am inclined to place a drug which I have used in many cases — seldom without benefit — chloride of aluminium. I do not know whether it has been used by others in such cases. It seems to have a combined tonic and sedative influence on the structures affected. Only yesterday I had a letter from a doctor about a patient who said he had been taking chloride of aluminium some time ago with great benefit, but had lately been trying a number of things. The doctor in- formed me that with none of these did he get any improv- ment, rather, his walking and his pains became worse, but as soon as he resumed the chloride of aluminium he again began to improve. It seems to have an especial effect in lessening the tendency to the pains. The dose is 160 LOCOMOTOR ATAXY. two, three, or four grains twice or three times a day after meals. It is fairly soluble, and you may give it in combi- nation with belladonna, Indian hemp, or nux vomica. Other tonics, as quinine, and sometimes iron, may be given with advantage in combination with arsenic or aluminium. Paroxysms of pain are most effectually relieved by antipyrin or acetanilide (once called anti- febrin). When superficial they are usually lessened by the application of chloroform to the skin, and very certainly, for several hours, by the hypodermic injection of cocaine at the upper part of the area of the skin to which they are referred. It does not influence the deep pains. No other local measures are of distinct use. Elec- tricity is not generally of service. If, indeed, the muscles are energetically faradised, so as to stimulate and wake up, as it were, the afferent nerves, the patients seem afterward to walk a little steadier ; but I have not been able to feel sure that permanent good has resulted from a course of this treatment. These, then, gentlemen, are the chief points with regard to the disease, to which I desired to direct your attention. I must hope that the facts I have mentioned, whether any of them are new to you or not, may be, at any rate, fixed in your minds by the relations I have pointed out, and those which have a direct relation to practical medicine may prove of service. We will now look again at the two patients whom we saw last week, and examine them in the light of what I have been saying to-day. I here show you the patient with the remarkable affec- tion of the cranial nerves — to an extent, indeed, that we very seldom see. He has, as you observed last week, considerable ataxy, loss of knee-jerk, loss of sensation ; he has even some loss of sensation in the region of the fifth nerve. His pupils are large, and do not act to light CLINICAL CASES. 161 or to accommodation. Accommodation is probably para- lysed. There is weakness of the right and left external recti muscles. The eyes are raised fairly well. The left converges much better than the right, although the right moves well in the lateral movements, — an illustration of the distinctness of the centres for the movement of the in- ternal rectus, with its fellow, in convergence, and that with the external rectus, in the lateral direction. There are conspicuous hollows in the position of the muscles of mastication, the temporals and masseters, and you can hardly feel any muscular contraction there. The motor branch of the fifth nerve on each side has evidently de- generated in extreme degree. The tongue is protruded well. He cannot bring his lips perfectly together, in con- sequence of weakness of the lower part of his face. He closes his eyes fairly well. He seems to close his glottis (as you observe by the explosive cough), although one of the vocal cords was found not long since to be paralysed. In the second patient there is a remarkable defect of sensibility upon the chest. He is suffering from the form of tabes to which I have already referred, in which there is greater damage to the nerve roots in the dorsal than in the lumbar region, so that there is a wide band of anaes- thesia around the thorax. He has no pain, only tightness round the chest. The irritation in the nerve roots has only caused this sensation of constriction. This sensation occurs very frequently when the root fibres are damaged and irritated by other processes. It is not a rare symptom, however, in locomotor ataxy, so that you must not im- agine, because you meet with it, that there is necessarily any focal lesion of the cord. Just below the nipple he ceases to feel a touch. Between the nipple and the umbilicus there is anaesthesia on the right side, but only in a much narrower band on 1 62 LOCOMOTOR ATAXY. the left. There is also a more extensive, wider affection of the nerves of touch than of the nerves of pain. In the first patient there was a loss of the plantar and cremas- teric reflexes, in the second the abdominal reflex should be lost in a greater degree on the right side than on the left. (On examination this was found to be the case.) It is distinctly greater on the left side, on which the sensation of touch has suffered most impairment. The loss of reflex action from the skin is related in greater degree to the loss of tactile than to that of painful sensibility. Here I must stop. Let me beg you, however, to go on. Supplement any new knowledge you may have gained to- day by reading a systematic account of the disease ; strive to connect the facts you thus learn with those you have just observed; and, above all, let the knowledge give point and effective power to the observation of the cases that are sure, sooner or later, here or elsewhere, to come under your notice. LECTURE XI. THE FOOT CLONUS AND ITS MEANING. Gentlemen : — The patient before you presents an unusual degree of increase in the symptoms associated with the name " tendon reflex," of which the knee-jerk and foot clonus are our special types. I have lately had occasion to draw your attention to the loss of these processes, and to the relation of the loss to the inco-ordination with which it is so often associated. The features of the excess of these phenomena, which I am going to show you, sug- gest that we may wisely consider their excess in relation to their nature. The process on which they depend has been, in my opinion, widely misunderstood, and a correct conception of its nature makes clear to us much that would otherwise be obscure. It is worth while, therefore, to consider its character, although in doing so we shall have to traverse ground familiar to many of you. I find, however, by experience, that the knowledge supposed to be familiar is often in a confused and shapeless state, and it is as useful to reconsider such a subject as it is to pre- sent the actual novelties that are welcomed with much more eagerness. What is the knee-jerk ? When the leg is in a certain posture, a tap upon the patellar tendon makes the muscles contract that end in it ; and, when the foot is in a certain posture, a tap on the Achilles tendon makes its muscles contract. It was at once assumed, not unnaturally, that the tap stimulated the nerves of the tendon, and that the Medical Magazine, Vol. IV., May, 1 895. 163 1 64 THE FOOT CLONUS AXD ITS MEANING. contraction of the muscles was a reflex action. At that time, however, the presence of nerves in the tendon had not been ascertained by the microscope ; the name em- ployed, however, involved their existence. A " tendon- reflex" action could not occur unless the tendons contain nerves. Hence, the point was at once investigated and the result was to prove that there are nerves in the tendon. Bat their investigations were wholly unnecessary to estab- lish the fact. May I ask each one of you now present to compress suddenly your Achilles tendon between the thumb and finger? If the squeeze is energetic you cannot possibly need the microscope to prove the presence of sensory nerves in the structure. The discovery, there- fore, that nerves could be seen in tendon was super- fluous, but it supplied the needful evidence that there was a mechanism for a "tendon reflex," and so it seemed to confirm the view that the contraction referred to is a reflex action. That opinion was embedded in the term "tendon reflex," and it has persisted to this day — a curious in- stance of the manner in which a name founded upon a theory may perpetuate the theory although it has been proved to be untenable. Within a year of these observa- tions, it was shown to be impossible that the contraction could be a true "tendon reflex." As you know, it is to be obtained in animals as well as in man. Tchirjew com- pletely isolated the patellar tendon ; he divided all its con- nections except those with the bone. He divided every nerve, and he found that this made no difference whatever to the effect of a tap on the tendon. That is an instance of the absolute facts of science, of the facts that cannot be got over. It alone shows that to whatever the phenomena are due they are not contractions due to the stimulation of the nerves of the tendon. There is another interesting observation that bears on this point and can be easily demonstrated. When there is WHAT IS THE KNEE-JERK? 165 moderate increase of these contractions, if the foot is gently pressed up so as to make the Achilles tendon a little tense, the muscular contraction can be obtained as well by a tap on the edge of the tendon as on the posterior surface. If the tendon, when tapped on the side, is sup- ported on the opposite side (by fingers against it), so that the tendon cannot move when tapped, no contraction occurs. Tap gently the tendon so supported, and then remove the support and tap again with the same force : the first tap has no effect, the second tap causes a contraction. The difference is that in the one case the tendon moves before the blow, in the other it does not. That is, unless the tap increases the tension on the muscle, it has no effect. This fact proves that the effect must be due to the in- creased tension, acting either on the muscle or on the tendon itself by its nerves. But the experiment I referred to, in which all connections of the tendon were divided, shows that it cannot be due to any effect on the nerves of the tendon, therefore the contraction must be due to the increase of the tension of the muscle. The result has never been disputed, it proves that the tap on the tendon acts through the muscle and not through the tendon nerves. With this fact we are able better to perceive the significance of another. We can only obtain these contractions when the limb is in such a position that the tap on the tendon in- creases the tension of the muscle. If the muscle and ten- don are relaxed the tap has no effect. There must be such a degree of tension as to enable the tap on the tendon to increase the tension of the muscle, or no contraction occurs. Let me here divert your thoughts to the fact I have recently pointed out to you in connection with locomotor ataxy, the fact that the muscle nerves are stimulated by two mechanical processes — by compression and by ex- tension. 166 THE FOOT CLONUS AND ITS MEANING. You may remember that I mentioned to you that it is the extreme compression of the over-contracting fibres that gives rise to the pain of cramp, and that when the muscle nerves are rendered hyperaesthetic by the effect of the cramp it is by tension that the pain is produced. The pain by compression during cramp, and the pain by tension in the after-state, afford us a most instructive illustration of the two modes in which these different muscle nerves are stimulated. We are, in general, unconscious of their exist- ence, their stimulation gives rise to no sensation. I use the term " afferent " because they do not normally influence consciousness. They are not " sensory." Nevertheless, the impulses act upon the motor centres in the cord, and, probably, also on the centre in the cerebellum influence their action by the state of tension or contraction of the muscle. To revert to our local "phenomena; " it is thus in har- mony with what we have learned in other ways that this sudden increase of tension should act upon the muscle and muscle nerves. The effect of a blow on the tendon is pro- duced through the muscle. So far as it is reflex, it is a muscle-reflex action, and not a tendon-reflex action. There may be error of name; that is, alone, of little consequence. You may call a pen by some other name without interfer- ing with its use, but it is different if the name involves a theory. You cannot use these phenomena, as an aid to understanding disease, if you call them " tendon reflex." At least it is necessary for every one first to unlearn the erroneous idea contained in the name, and, as a fact, a large number of persons never unlearn it and never gain the clear perception they might acquire of other phenomena of dis- ease. Let me now try to show you that a certain tap on the tendon is effective only when it can increase the tension of the muscle. In this patient, suffering from hemiplegia, MYOTATIC CONTRACTIONS. 167 I press up the foot. Unless the foot is pressed up so as to make the muscles tense a contraction cannot be obtained. For it, there must be gentle tension. Hence I have called the phenomena " myotatic " or " tense-muscle " contractions. When in excess, and the muscles are made tense, the very slightest sudden mechanical influence pro- duces the needful stimulation. I very gently tap the tendon from the side, first without support on the other side, and then supported by the finger tips so that the tap does not displace the tendon. You note that the tap, which in the first instance caused contraction, has no effect when the tendon cannot move — that is, when the tap cannot suddenly increase the tension on the muscle. You see the difference is clear. This fact, taken in conjunction with the other fact I mentioned, though the division of the nerves of the tendon prevents it, makes certain the fact that the increased tension does act upon the nerves of the muscle. Then comes another question which is, however, com- paratively unimportant. Is that contraction an immediate reflex effect of the stimulation of the afferent nerves of the muscle ? It is a difficult question, and the answer depends on difficult time measurements. When the interval be- tween the tap and the contraction is carefully measured it is found to be shorter than that which we can well con- ceive can be that of a reflex action. But I must ask you to allow me to pass this by. It seems to me probable that each contraction is not reflex. But this has nothing to do with the exclusion of the tendon nerves as the source of the afferent impulses. Yet I shall ask you to let me assume the theory that each contraction is not reflex, because I can better thus put before you the general principles on which I desire to insist, which help us most to understand the phenomena of disease and of health, and of which this special point is only one unessential element. 168 THE FOOT CLONUS AND ITS MEANING. Far more important than that is the way in which the theory I am about to mention harmonises with other facts and illustrates many phenomena of disease. We only get these contractions when the muscle is in a certain state of tension, and the theory I refer to is that that tension, by gently stimulating the afferent nerves of the muscles, in- duces a state of the muscular fibres in which they are excessively sensitive to the mechanical influence of the suddenly increased tension of the tap. The gentle tension causes the irritability. The irritability is reflex ; the con- tractions are local, and would not occur were it not for the irritability which is reflex. In connection with this, two other facts are of impor- tance. First, there is the condition called muscular " tone " or " physiological tonus," the curious state of slight con- traction of the muscle by which the muscle is always adapted to its posture, and always maintains a certain amount of contraction whatever the proximity or distance of its attachments. This depends upon the spinal cord. If the motor nerves are divided it ceases instantly ; it ceases instantly also if the sensory nerves are divided, and, therefore, it must be a reflex process. It is probable that this muscular tone, when a little augmented by an increased afferent impulse from tension, is the irritability which makes the muscle contract in response to the local stimula- tion. In most instructive harmony with this we find that whenever these phenomena become excessive, there is a tendency to tonic spasm, which is seen in such striking degree in spastic paraplegia, and, as you know, is always accompanied by such conspicuous excess of these contrac- tions. Compare the light which that throws on the symp- tomatic pathology of spastic paraplegia with the absolute darkness in which the " tendon-reflex " theory would leave it ; if, indeed, it were conceivable that the tendon-reflex theory could be held by any one who knows the facts. HOW PRODUCED. 169 But the name " tendon-reflex " goes on and conveys its theory to those who know nothing of the subject except the name, and thereby hinders them from all perception of the really instructive elucidation conveyed by the alterna- tive term the "muscle-reflex action." Therefore I never use the term " tendon reflex," nor have I done so for many years. I use the term " muscle reflex," or the general term which I have mentioned, " myotatic." That word, however, has not become much used. I am, however, consoled by the different fate of another term I suggested at the same time. At the meeting of the British Associa- tion at Cambridge I was talking with Professors Clifford Albutt and Westphal about the error of the term " tendon reflex " — which Westphal (who first pointed out the im- portance of the symptom) had always avoided. He pre- ferred the name " knee-phenomenon." I asked, "Why not call it the ' knee-jerk ' ? " Clifford Albutt so emphati- cally approved the term that I took the first opportunity of setting it going, and its use immediately became universal. The contraction which follows the tap is an instanta- neous one. The increased tension acts on all the fibres of the muscle; all contract. It is an instance of a simul- taneous single contraction of muscular fibres. When there is maintained contraction of a muscle, it depends upon the fact that a similar contraction occurs in different fibres not simultaneously, and the contraction is main- tained because there is never a moment in which a large number of fibres are not contracted. Clonic contractions are simultaneous ; tonic are not. If the muscular irritability excited by tension is con- siderably increased, and the tension is suddenly applied, it causes a single contraction ; as this ceases, the excitability of the muscle is renewed, and, if the tension is maintained, contraction occurs. That is the reason why, by main- taining the tension upon the muscle, a clonus is produced '5 170 THE FOOT CLONUS AND ITS MEANING. — a foot clonus or a rectus clonus. There is no essential difference between clonus and the knee-jerk, except that the irritability must be excessive to permit a clonus to be obtained. The patient whom I now show you, has an extreme excess of this irritability. Before pointing out its features, I may mention the conditions under which it becomes excessive. You remember what I said recently regard- ing the impulses from the muscles in locomotor ataxy, and the way in which their loss deprives the spinal centres and the brain of the guiding impulses which depend upon the state of the muscles, and so disturbs co-ordination. I indicate on the black-board, in the diagram before you, a couple of muscular fibres, and the connective tissue between them, and from between them a couple of afferent nerves fibres going up, through the ganglia on the posterior root, to the spinal cord. These fibres have different destinations. One passes through the postero-external column, ascending in it to the medulla, where it en'ds in grey matter, whence a path continues it to the middle lobe of the cerebellum; there its impulses are combined with others to guide the motor cortex of the brain. If the nerves are degenerated, the postero- median column degenerates up to the medulla. The other fibre seems to pass into the posterior cornu, and there, probably dividing, comes into relation with the processes of the motor cells, from which a motor fibre passes down to the muscle. This afferent fibre, acting on the motor cell, is a type of the fibres which determine what I call the muscle-reflex action, that is, from the muscle to the muscle. These determine the physiological tone and determine also the excitability manifested in the knee-jerk and, when excessive, in the clonus. But the motor cells form part of the voluntary path from the brain. This descends in the lateral column, in the " pyramidal tract," FUNCTION OF AFFERENT FIBRES. 171 the fibres of which seem to end by branches that are related to those of the motor cells ; thus the pyramidal fibres act through the motor centres of the cord, which are also influenced by the impulses from the muscles. We cannot say how much these afferent muscular im- pulses do in determining the form and degree of the action excited from the brain. They probably have a very great influence in facilitating the arrangement of muscular contractions, especially in standing and walking. Yet these pyramidal fibres acting on the motor centre seem also to restrain its reflex activity. We should not know this if it were not for the effect of the loss of those fibres when the pyramidal tracts are destroyed. There is an in- crease of the muscle-reflex action and of all the pheno- mena by which it is manifested. As you know, in such a case we have an excessive knee-jerk, but that excessive knee-jerk does not develop at once. If the cord is divided, there is instant excess of the superficial reflex action of that from the skin, but it is some time before there is excess of the muscle-reflex action, and it after- wards gradually increases. It seems as if the function of the pyramidal fibres was to exert a continuous gentle restraint on the muscle-reflex centre. When that influ- ence is withdrawn, there is a gradually progressive augmentation, a sort of functional hypertrophy, of the muscle-reflex action, which leads first to excessive knee- jerk, then clonus, then gradually increasing spasm, and then to the full condition of spastic paraplegia. At last there is a condition of spasm so intense as to prevent the contractions being obtained. I shall take some other opportunity of speaking of the fallacies there are in obtaining evidence of the presence of the knee-jerk when it is normal. I wish now to show you this remarkable feature of the excess of this muscle-reflex irritability. The patient before you is a man who has had 172 THE FOOT CLONUS AND ITS MEANING. myelitis in the dorsal region of the cord. I need not trou- ble you with his history. His symptoms afford evidence of perfect integrity of the lumbar enlargement, except for the secondary degeneration of the pyramidal fibres, due to the effects of the inflammation higher up. They have to some extent recovered, and can perform their function of conveying the voluntary impulse, so that he is able to move his legs, but so imperfectly that he is still unable to stand without support. He has much extensor spasm in his legs, the result of the excess of this muscle-reflex action, and the degree of this excess is manifested in a feature that is not often to be observed. Xot only is a clonus in the extensors of the ankle, the calf muscles, produced instantly by the gentlest passive flexion, but when he flexes the ankle himself, by voluntary contraction of the flexor mus- cles, the tension of their opponents sets up clonus in them. The ordinary way of obtaining clonus is suddenly to make the muscles tense, and the tension develops irritability and produces contraction without there being any interval which you can appreciate. But, as I have explained, when the first contraction is passing off, the maintained tension acts as a stimulus to another, and the time of each contraction is such that their frequency is about 6, 8 or 9 per second. When the patient flexes the joint you see there is developed a characteristic clonus, quite like that produced by passive tension. It is a feature you will seldom see, although I confess I cannot tell you why it is so rare for voluntary movement to do that which passive movement does so readily. It is, as a rule, only in the muscles of the calf that a clonus is obtained, not only because they are conveniently made tense in this way, but also because in them this re- flex action is normally very active and important. You may often get a clonus in the muscles which move the foot laterally, and also in the rectus femoris when the patella is ITS DIAGNOSTIC SIGNIFICANCE. 173 pressed towards the foot. I now press the patella straight down towards the foot and then tap it in the same direc- tion. You see that then a well-marked clonus is developed. When the patient is horizontal you can obtain the knee-jerk quite readily by depressing the patella and then giving a tap on the depressing finger in the direction of the patella, but not unless it is in slight excess. Not only can a clonus be obtained in the peronei ; I have even got it in the muscles of the great toe. It is often to be obtained in the flexors of the fingers in cases of hemi- plegia. I have even found it in the trapezius. With two other practical remarks I will conclude. First, remember that when a patient is in bed the best way of testing the knee-jerk, provided that you are unable to get it by pressing the tendon in the way described, is to let the hip and knee be flexed, and to let the foot rest upon your hand quite passively, and then the leg is in the most convenient position, with just the right amount of tension for the production of the contraction. This method is of extreme value in testing the knee-jerk. The other fact is that whenever the excess is in such degree as to enable the clonus always to be obtained, that is, when the irritability is so great that the continued ten- sion acts as a stimulus after each contraction, there is cer- tainly nutritional change in the muscle-reflex centre — there is persistent functional excess such as implies a nutritional change. This nutritional change is important. There has been much discussion as to the significance of a definite, distinct clonus, independent of any voluntary contraction of the calf muscles, such as you can readily perceive. I am certain of this, that for one case of error due to clonus being thought to be due to organic disease, when it was really due to a so-called functional state, twenty cases of error have arisen in consequence of the clonus of organic disease being disregarded, its significance being underrated. I 174 THE FOOT CLONUS AND ITS MEANING. could give you a long list of very instructive illustrations of the errors into which disregard of the foot clonus has led distinguished physicians. Yet one other point. In spite of the significance of such extreme increase as you observe in this patient, as proof of structural disease, it is no proof of irrecoverable disease. Never have I seen such intense excess, with such intense spasm, as was presented by a patient of whose case proba- bly most of you have heard, and in whom there was abso- lute motor and sensory paraplegia up to the level of the ensiform cartilage. In spite of the absence of cutaneous sensibility there was that terrible excess of the sensibility to the deep-seated pain of muscular spasm, so that the con- tractions caused an agony of such intensity that the man must have died of sheer pain in three months' time. The symptoms were due to a tumour pressing on the spinal cord ; the tumour was removed, and that patient has been, for years, free from the slighest trace of increase of knee jerk, or spinal symptoms of any kind. He daily walks miles, pursues an active life, and is the head of a big manu- facturing firm. Ten years have now elapsed since he was in the condition I have described, and his case should im- press on you the fact that no excess of myotatic irritability, and no degree of clonus or of spasm, preclude the possibil- ity of recovery. LECTURE XII. A CASE OF SYRINGOMYELIA. Gentlemen: — As the changed conditions of my work here prevent me from continuing the out-patient clinics which I have carried on for many years, I propose to try to find some substitute for them in showing you a case each Wednesday afternoon, and endeavouring to extract from its facts such information and instruction as it will yield us. I begin to-day with an example of an uncommon disease. Let me, at the outset, warn you against the frequent error of thinking that a rare disease can only yield you know- ledge which will be rarely needed. It is not so. Diseases which are uncommon cannot be studied without consider- ing common facts and without giving increased ability to recognise diseases which are often met with. The patient before you came into the Hospital a few weeks ago, presenting, as he presents now, two chief sets of symptoms : he has extensive loss of sensation over the hands, arms, and parts of the trunk and legs ; he has also muscular weakness, and muscular wasting in the right hand and forearm, and a little weakness in the left. The diseases in which loss of sensation is the chief symptom are not numerous. You meet with it as an iso- lated symptom in functional anaesthesia, in hysterical hemianesthesia, for instance ; you find it also in some cases of organic diseases of the brain which cause hemi- anaesthesia, and you meet with it in organic disease of the Delivered May io, 1893. Clinical Journal, May 31, 1893. 175 176 A CASE OF SYRINGOMYELIA. nerve trunks. But in all these forms, each kind of sensa- tion is involved ; the tactile sensibility is usually impli- cated even in greater degree than sensibility to pain. But in this patient — as in most other cases of the kind — the loss of sensation presents the remarkable feature that it involves sensibility to pain, and involves sensibility to temperature — to heat and to cold — but does not involve sensibility to touch. Such a combination, with the exclusion of tactile sensibility from impairment, is met with only in one other morbid process — degenerative changes in the peripheral nerves ; such, for instance, as are met with in many cases of tabes, and in a few cases of alcoholic and other forms of neuritis. In these, also, you may have sensibility to pain lost, sensibility to touch persisting, but their characteristic is that they involve the extremities of the limbs, either alone or in preponderant degree, and do not extend far up the limbs toward the trunk. But, in this case, and in other cases of the kind, the loss of sensation extends over the whole upper limbs, and involves a considerable portion of the trunk. Not only does it exist on the hands and fingers, but upon the chest as far as the fourth cartilage ; it involves the neck and even the back of the head as far as the cervical nerves are distributed ; it involves, in front, the lowest part of the abdomen on the right side. It also extends (an unusual feature) over the front of the thigh as far as the knee ; and behind — a point to which I ask your special attention — it is found over the whole sciatic area. Thus you see, by its distribution, it is entirely distinct from the only other morbid state in which this special loss is met with. Let me show the facts I have described. I prick the patient on the palms and fingers and arms with a point, and you hear that he feels it only as a touch. So also on the chest, but you perceive that on the left side this condition does not extend quite so far down as on the right side. We LOSS OF SENSATION AND MUSCLE-POWER. 177 examine the thermic sensibility by test-tubes containing hot and cold water. On the arms we find that he is unable to recognise either heat or cold, merely feeling the contact of the tube ; he has some recognition of heat on the upper part of the chest and the back, but not of cold. If we were to examine the leg we should find that the condition is essentially the same. Note the instructive fact that the sensitiveness to cold may be lost when that to heat is not ; it confirms the opinion that these two forms of thermic sen- sibility are subserved by different nerves. The second symptom which he presents is the muscular atrophy in the right hand. Observe the peculiar posture of that hand ; those of you who are familiar with the effects of muscular paralysis will recognise the position as that characteristic of paralysis of the interossei. The last two joints of the fingers are considerably flexed by the unop- posed action of the muscles which bend those joints, the long flexors; when he attempts to straighten the hand — to extend it — you see the effect in still more marked degree ; he cannot extend the middle or last joints of the fingers; he is able to extend the metacarpo-phalangeal joints by means of the long extensor, but the power of flexing these joints, and of extending the others, due to the interossei and lumbricales, is lost. This shows some weakness also in the long extensor. He is not able to extend the wrist together with the fingers. When he closes his fist, how- ever, the wrist can be well extended, because then the special wrist extensors are active, while they are inactive in extension of the wrist and fingers together. You see, more- over, the distinct wasting of the thenar and hypothenar muscles, and the hollow interosseal spaces, in striking con- trast with those of the other hand. We will now see what the electrical irritability of the muscles is ; this will tell us the state of nutrition of their nerves as well as that of the muscular tissue. You see 16 178 A CASE OF SYRINGOMYELIA. that a moderate faradic current causes contraction in all the muscles of the left hand and forearm, but in those of the right hand no contraction can be obtained with even a strong current. There is loss of the excitability that de- pends on the nerves within the muscles ; their state cor- responds with that which exists higher up their course, and if we test the nerve-trunks that supply these muscles we find their fibres equally insensitive. They have undergone degeneration. If we apply voltaism to them they are equally irresponsive. If we apply it, however, to the muscles themselves, we obtain a contraction, because to it the muscular fibres can react ; they do so, for instance, in the interossei, but only to a very strong current, and in the thenar and hypothenar muscles we get no response. There is thus a diminution of nerve and muscle irritability cor- responding to the wasting, with persistence of a very slight excitability in some of the muscular tissue. It is a condi- tion that results from, and is evidence of, a slow degener- ation of the motor nerves, such as results from slow gradual damage to their nerve cells in the spinal cord. This disease affords us a striking illustration of the great increase of diagnostic power that has come from the union of pathological and clinical observation. From its clinical characters it would not be possible to infer the exact nature of the lesion ; and what may seem more strange, however familiar we might be with the pathological change, if we knew it alone, we should be unable to infer from it that such symptoms would be produced as are present in this patient, and which attend the morbid pro- cess in the majority of cases. But let us see how far a study of the symptoms can lead us. Such an extensive differentiation in the loss of the different forms of sensibil- ity can have only one meaning. We have seen that it does not occur in degeneration of the terminations of the nerves in that distribution which we perceive in the case COMBINED STUDY OF SYMPTOMS. 179 before us. It does not occur in disease of the nerve trunks, which causes a general loss, corresponding to their distri- bution. But in the spinal cord we know that the paths for pain and touch are separate, because we know that various lesions of the cord — traumatic lesions, for instance — not uncommonly involve sensibility to pain, and leave sensi- bility to touch unaffected ; and we know also that the path of temperature must be near that of pain, because the two forms of sensibility are so frequently involved together. Therefore we can infer that there is probably a lesion of the spinal cord so situated as to implicate the paths of pain and of the thermic sense, and not that of touch. Beyond this we could not go except to infer, from the gradual progress of the symptoms — to which I shall refer later on — that it must be a lesion of chronic course. But the combination of clinical and pathological observations I have mentioned, has shown the remarkable fact that in most cases in which this combination of symptoms is met with, in any distinct and characteristic degree, the obtru- sive pathological feature of the morbid state of the spinal cord is the presence of a cavity or cavities within it. The disease is that which is currently known as " syringomye- it is sometimes called hydromyelia, but there seems a tendency to use the term syringomyelia to the exclusion of the other. These tendencies of nomenclature, I may remark in passing, are so strong that it is generally useless to attempt to oppose them ; we have to accept that name which is generally adopted, although the term " syringo- myelia," since it comes from a word meaning " trumpet," is not a very apt one. Cavities in the cord, most of which give rise to symp- toms more or less similar to those before you, are very various. The most common and the most typical is the dilatation of the central canal. It may be slight, and then may cause no symptoms ; it may be great, so great as to be 180 A CASE OF SYRINGOMYELIA. equal to the normal area of the cord, and by the pressure it exerts it may reduce the tissue of the cord to a narrow layer surrounding it. Lower down, there may be a normal central canal, or a central canal which is closed by a mass of nuclei, a condition which is often met apart from any disease. In other cases, with a normal or a closed canal, behind the posterior commissure there is a cavity — a slit- like cavity — between the two posterior columns, extending almost to the posterior surface of the cord. You will see examples of both forms under the microscopes. Further, we may have slit-like cavities, more or less distended and widened, in each postero-lateral column, or sometimes in one only. Similar cavities may be found within the sub- stance of the posterior horn, extending from the front to the back of the horn. Among the sections for your ex- amination are examples of both these changes, fissures in the posterior columns, in both posterior cornua, and in one only. Sometimes we find small slit-like cavities in the posterior part of the lateral columns, and of these also you will find an instance. We may even have a cavity, on one or both sides, at the inner part of the middle portion of the grey matter, near the base of the anterior cornu ; and by its enlargement, such a cavity may come to occupy a large part of the place which should be occupied by the anterior horn and its nerve cells. But another most important fact is illustrated by these sections and is generally to be observed. The cavities are not the only pathological feature. If you examine them closely, you will see that they are bounded by a layer of tissue different from that of the rest of the cord, a tissue resembling that which lies on each side of the central canal, and which occupies the furrow at the side of the posterior nerve roots, and which lies also in a thin layer between the pia mater and the cord, and is continuous with the neuroglia which penetrates and ramifies among the ORIGIN OF THE CAVITIES. 181 nerve fibres of the white columns. It is a relic of the embryonal tissue of the cord. It is similar to that of which the embryonic cord is composed, and which, in the process of development, slowly changes to nerve elements — nerve cells and nerve fibres. The presence of this embryonic tissue is evidence of the important fact that most of the cavities, certainly all those around which it can be recog- nised, and probably some others, are due to an arrest of development. They are due to the persistence of the early cavities, or to the breaking down of persistent embryonal tissue, or to both processes. This conclusion is confirmed by some other histological features which are found in the cases in which there is a distinct layer of this neuroglial tissue, and which may be sometimes found when this layer cannot be clearly seen, especially a peculiar sinuous mem- brane lining the cavity, or peculiar fringe-like processes extending outward from such a membrane. The origin of most cavities in arrested development is thus evident from their features. In the posterior commis- sure the epithelial lining may show that a cavity is the dilated canal. The embryonal groove is, you will remem- ber, at first closed at the posterior surface only; some cavi- ties seem to represent an arrest of development at that stage, because the cavity due to the enlarged central canal extends backwards up to the surface of the cord. In other cases, as in one under the microscope that I have already mentioned, the closure of the original cavity behind the central canal, to form the posterior commissure, has taken place, and then an arrest has occurred, so that the cavity is between the two posterior median columns. When a slit-like cavity is in the posterior part of the lateral columns, and sometimes in the posterior horns, or in the lateral columns, the adjacent neuroglial tissue, which should not exist there, shows that the embryonal structure has failed to develop, as it should have done, into nerve elements, iS2 A CASE OF SYRINGOMYELIA. and, probably, most of these cavities are due to its disin- tegration. But a cavity in the posterior horn, or inter- mediate grey matter, may end at the middle line, by a rounded space in the position of the canal, and here an epithelial lining may prove that the cavity is continuous with the enlarged canal. I show you such a specimen. You ma}- remember that, at first, before the posterior com- missure is formed, there is a short extension of the canal on each side. It would seem that such cavities have arisen by the extension of one of these processes, after the union behind. But not only does the persistent embryonal tissue remain in the amount which we should expect from arrested trans- formation, — it is sometimes in much greater amount, and has manifestly been increased by a process of active growth. You may perceive indications of that in one of the sections under the microscope, in which there are rounded projec- tions into a cavity, clearly due to arrested development. The process of growth may much increase the quantity of this tissue, and even make it preponderate sometimes in an extreme degree over the cavity. A cavity may even become obliterated, perhaps quite early in life, by the exu- berant hyperplasia, and a state of " central gliomatosis " may effect damage, and cause symptoms, like those that are produced by a distended cavity. The cavities formed by these processes, and likewise the overgrowth of tissue, are most common in the cervical region; sometimes, how- ever, they extend a long way down the cord, and they may occasionally be seen in the lumbar region. The distension of the canal may also pass up even to the fourth ventricle, which may be increased in size, and the process of growth of the residual tissue may invade the floor of the fourth ventricle, and extend even as far as the upper part of the pons. Finally, as an additional evidence of the develop- mental and coneenital nature of the changes we have con- LOSS OF PAINFUL AND THERMIC SENSATIONS. 183 sidered, it is not rare to find, in conjunction with it, some developmental defect in the brain. A few other forms of cavity are met with in the spinal cord, in rare cases, with which we need not concern our- selves just now. There may be a rarefaction of the grey substance, and a disintegration of it at some point ; but such cases are quite distinct from those which form our subject to-day. They are chiefly senile or due to manifestly acquired degeneration. I believe that a similar disintegra- tion of the white substance into a cavity only occurs in the neighbourhood of traces of undeveloped embryonal tissue. There, you may see that the nerve fibres and the interstitial tissue are alike scanty, and they may easily break down into an irregular cavity, the margins of which indicate its origin. You can readily understand that a morbid process which presents so many variations gives rise to symptoms that differ considerably in their character in different cases. In spite of these variations, in the great majority there is the loss of painful and thermic sensations. We do not yet know what mechanism in the morbid process gives rise to this peculiar symptom ; how far it is due to simple com- pression of the conducting tracts that convey these forms of sensibility, or how far it is due to damage to the fibres of the posterior commissure, by which we must assume that these sensory impressions cross. Again, when there is a cavity in the posterior horn, we can hardly conceive that the nerve cells, to which probably the nerve fibres first convey these sensory impulses, are not also damaged. These facts render it mysterious how sensibility to touch can so frequently escape. There are cases, however, in which all forms of sensibility have been affected. The affection of sensibility is commonly met with in the arms and parts of the trunk, very seldom, as in this patient, in the leg, though the legs are not seldom weak and present iS4 A CASE OF SYRINGOMYELIA. the characteristics of spastic paraplegia, a fact which we can understand when we consider the pressure which the pyramidal tracts must undergo when there is great disten- sion, or overgrowth of the embryonal tissue, or cavities in the lateral columns themselves. Muscular wasting, such as we see in this patient, is common, in one or both hands, sometimes chiefly in the shoulder muscles. It is appar- ently due to the damage to the anterior cornua by the enlarging cavities or compressing growth, and on the locality* of these its position depends. Other trophic changes are sometimes met with, the most frequent of which are lividity at the extremities of the hands, and inter- ference with the growth of the nails ; occasionally, especi- ally where there is muscular wasting in the upper part of the arms, there may be changes in the joints, especially in the shoulder and elbow, resembling those of tabes, and differing from those in chronic rheumatoid arthritis by the fact that the thickening takes place chiefly outside the capsule of the joint and not within it. Other symptoms, which, however, are rather rare, are the occurrence of ulcers and other changes in the nutrition of the skin of the fingers. More frequent are those which are due to the ascension of the affection to the medulla. The sympathetic may be involved by the lesion, so far as concerns the eyes, when the mischief extends to the upper cervical region of the cord ; and hence changes of the pupil-contraction have been sometimes noticed. The bulbar nerves occasionally suffer, and also those for the external ocular muscles, the nuclei of which may be reached by the distending force. In this patient there is weakness of the external recti, pro- bably due to the distension of the fourth ventricle. There is also nystagmus, which has the same significance, and is not at all rare in this disease. One other pathological fact should be mentioned to com- plete our general survey of the more salient features of this DEGENERATION OF PERIPHERAL NERVES. 1S5 strange inalaay. In several rases a erenerati :r. has 'iter f:und ::: the peripheral cutaneous nerves. It has been zhiehy 1 coked :":r v.- hen there have been trtphic zhanves, and has tnen teen constant. It seems scarcery n.-te.y tnat the soeeial -ens : ry lass is due to this, since the ae venerate a nerves are of the extremities, and the loss of sensation is so extensive. Such a malady, so peculiar in its manifestations, is not lively:: ': e mistaken, prtvided :nly that its features are knew::. There are in been inly v: disease- vith v.-hich this is likely t: be :::::: u nded, provided you knov its fea- tures. One of these is chronic cervical pachymeningitis, in which there is much thickening of the cervical dura mater, and damage to the nerve roots. In such a case you have wasting and loss of sensibility in the arms, but the loss of sensibility involves all forms, is less ex:ensive. and is preceded by much pain. These are distinctions which seldom fail. The other condition is one which cannot be entirely separated from the disease we are considering, be- cause it seems to be, in part at least, due to the same morbid process; it is the affection which has become known under the name of " Morvan's Disease," or " Painless Whitlows." in which whitlows form on the fingers, and produce ulcers which do not heal for a long time, sometimes indeed only when the finger-ends have been lost. The pal nlessness is b::e t: thelzss :f sensibility t: z air.. :::::::: inly in ::::;:;:: :- ti in vith liss if sensibility t: tennaerature. : at ".nth::: t less t: teach. I n:ay nnentitn. as a striking instance cf the analgesia a:::n:aany:ng this affect! :::. that the shaalaer joint has been resected without pain, although : anaestheti : •as v:i 11 ; rvan's Disease " is f:ana :i be fee:: cavi- ties in the cord, similar to those in the condition of syringo- myelia, but combined, apparently in all typical cases, with extensive chronic degeneration, or degenerative neuritis, of the peripheral nerves. Its chief distinction is the presence 1 86 A CASE OF SYRINGOMYELIA. of these painless whitlows, and there is usually a greater limitation of loss of sensibility to the lower portion of the limbs. This is thus rather a variety of syringomyelia than a separate disease. These, then, gentlemen, are the main features of syringo- myelia ; and the chief characteristics by which it is mani- fested. It is a disease for which, you will be prepared to hear, we are able to effect little by medicinal treatment, though we may be able to give temporary relief to its symptoms. This patient complained greatly, on his ad- mission, of distressing pain in the head, probably due to increased pressure within the cavities of the brain. This symptom has been greatly relieved by antipyrin. Elec- trical treatment may maintain for a time the nutrition of the muscles, but when their wasting depends on a morbid process that destroys the motor cells, we cannot hope for any regeneration of the nerves to reward our efforts ; we cannot anticipate any restoration of the path by which the will can act upon the muscles. But the future may per- haps have in store a ray of light to lessen the present practical gloom. Where the physician fears to tread, the surgeon sometimes steps in with the best results, and we must humbly bow in grateful recognition of the good he has succeeded in effecting in recent years, the lives he has saved, and the hindrance to the progress of disease which he has produced, the relief he has given, when we could not save or relieve or retard. But I am not aware that the surgeon's skill has yet been applied to this affection.* It seems to me that it is quite possible that, in some cases, surgery may be able at least to effect an improvement in * Since this lecture was delivered I have found a report of one case that was operated on in the manner suggested, without ill effects, with transient amelioration, but no permanent benefit. The case was, however, unfavourable, because the chief symptoms were in the legs, and cystitis ultimately caused death. (Abbe and Coley, Journ. Nervous and Mental Dis., July, 1882.) PROGNOSIS UNPROMISING. 187 the condition of the patient, and a retardation of the pro- gress of the affection. The exposure of the cord would readily enable him to ascertain whether the process was the distension of a cavity by liquid, or the presence of a morbid growth. A central growth, extending high and low, could not be touched, but it may prove to be different when the chief element in the lesion is a distended cavity. The draining away of the fluid contents might enable a cavity to contract, and when the amount of circumferential growth is not too large, the relief from the slow compres- sion of the adjacent structures, would probably permit their recovery to a considerable extent, because no effects pass away so readily as those which are produced by slow com- pression, when that compression has been removed. The malady we have been considering is thus one of those developmental diseases on which we, as practitioners, can only look with feelings which are the reverse of grati- fying. It must be so. Such affections are the result of morbid tendencies, perhaps inherent in the germ, certainly beginning in the embryo, continuing in childhood, and be- coming active and aggressive in a later period of life. Their activity seems often to coincide with the completion of the process of growth ; but over the morbid tendency we are as powerless as we are over growth itself, or as the monarch of old was over the waves of the advancing tide. The ad- vance of these diseases is sometimes rapid, sometimes slow; sometimes, like the waves on the shore, they may seem for a time to give no evidence of onward progress ; they may even seem to recede, but it is only for a time. Before long a fresh advance carries their effects beyond the highest point they had before attained. But here, alas, we find that our simile fails us. The tendency of these diseases knows no retrocession. It is a tide which flows, but never ebbs. LECTURE XIII. THE TREATMENT OF MUSCULAR CONTRACTION. Gentlemen : — You are familiar with the fact that conse- quences of disease are sometimes more serious than the maladies from which they result. I do not now refer to the cases in which one definite disease is produced by another, such as the paralysis due to the poison that is the result of diphtheria, which causes such grave after-effects. I desire to draw your attention only to some simple results of disease, results that are the necessary consequences of the primary affection, — which we may almost describe as the natural effects of unnatural states. But, although they are normal consequences of abnormal conditions, they reach, by their simple increase, a degree which constitutes so definite a disorder that we are inclined to regard it as a morbid process. A morbid process it is, so far as its effect is concerned, although it is not a morbid process in its nature. It is disease only because it is an excess — because, in the absence of the normal limitation, it passes on to a condition which disorders normal action. This being the nature of the contractions of which I am about to speak, I must point out to you their causes and mechanism, that you may understand them, and that you may discern their nature when you encounter them ; that you may know what they mean, and that you may know how to remove them — to remove them if it may be, to lessen them if they cannot vanish, and, above all, to antici- pate their occurrence, to discern the conditions in which The Clinical Journal, March 6, 1895. 188 TOXIC CONTRACTION IN ORGANIC DISEASE. 189 they will develop and yet in which they need not develop, to prevent them in the strictest sense of the word — to come before them, to stand in their way, to obviate them. Such a secondary process, arising from the unrestrained action of processes which are normal, is the cause of many of the results of disease which are most difficult to remove. Indeed, removal is only possible when the normal restraint can be renewed. The process of diminution even then is very slow. The renewal of restraint is gradual, and has to act on a tendency that has become far greater than normal, and, moreover, has a tendency to increase. This has first to be arrested before diminution of the abnormal degree can be achieved, and the normal restraint re-established, if its recover}- is adequate. I use these general terms because the conditions de- scribed, and the influences referred to of restraint on the one hand, and excess on the other, may be perceived in many morbid states, and their discernment enables us to understand better many processes and many consequences which outlive their causes. They often seem like indepen- dent maladies when they are only properly to be regarded as runaway vitality. For they depend really on the normal power of living structures to adapt themselves, by their nutrition, to conditions different in character or only in degree from those of health. This process, as I say, may be met with in many affec- tions, but in none more clearly than in the condition to which I want to direct your attention to-day, the muscular contraction met with in many paralysing maladies of the spinal cord and of the nerves. The tonic contraction of muscles that occurs in organic disease may be divided into two chief forms, easily distin- guishable, and important in their significance — that which yields to an attempt to extend the muscle, and that which cannot be overcome. r 9 o THE TREATMENT OF MUSCULAR CONTRACTION. I propose to consider to-day only the form that cannot be overcome. The calf muscles are often thus contracted as I show you in this patient, who is recovering from mul- tiple neuritis. When the foot is pressed up by the hand, placed against the sole, the calf muscles resist, the contrac- tion yields a little, and the foot is brought almost or quite to a right angle with the leg ; but however long you con- tinue the pressure the foot goes no further. That is the contraction which depends upon tissue-changes in the muscle, causing it to be structurally shorter than normal. It is of very great practical importance. You may remem- ber observing it in the boy with idiopathic muscular atrophy whom we examined last week. The first point for us to consider is to what it is due. It is met with especially in muscles that are less para- lyzed than their opponents, and is most common in the calf muscles, which extend the ankle joint, when the flexors in the front of the leg are paralysed. But we also meet with it when both sets of muscles are paralysed, and also in cases in which there is no paralysis, provided extension of the foot on the leg has been maintained for a long time. It occurs also in the flexors of the hip and of the knee if, from any cause, the patient lies in bed for a long time with these joints flexed. If you scrutinise the conditions in which it is met with, you will find that it is due especially to posture — that even paralysis is only effectual by induc- ing a certain posture. You know that there is an extraor- dinary capacity of the muscles to adapt themselves to pos- ture. Whatever the position of a limb or joint, the muscles must always continue in a certain state of slight gentle tonic contraction — "physiological tonus" or " tone " as it is called. If a limb is moved passively, the muscles which are relaxed do not wrinkle up ; they remain contracted in exact proportion to the approximation of their attachments, and the muscles which are extended elongate, but remain CONTRACTION OF THE CALF MUSCLES. 191 in similar tonic contraction in exact proportion to the in- creased distance between their attachments. That capacity for adaptation to posture is evidently one phase of the condition which physiologists call " tone," and which de- pends upon the connection of the muscles with the spinal cord. From the nerve cells, in the anterior cornu, motor fibres proceed to the muscle. From the tissue between the muscular fibres, other nerve fibres pass up to the pos- terior cornu of the cord, and are connected, by a mysterious interlacement of fibrillar, with the branching processes of the cells of the anterior cornua. On the integrity of that arrangement this adaptation to posture and this muscular tone seem to depend. In consequence of that, if the attach- ments of a muscle are permanently and constantly approxi- mated in consequence of posture, the changes which, as you know, are always going on in every structure and every tissue of the body, alter the muscular fibres and the interstitial tissue, in accordance with the diminished length, so that after a time you are unable to extend the muscle — unable, that is, by any attempt which you may make at the time. We meet with this in cases of unequal paralysis, and also in cases in which one posture is constantly adopted. This effect of paralysis may be always traced to its influence in causing a certain position of the parts. That is the great cause of the various forms of talipes that result from the unequal palsy of muscles in infantile paralysis. The most common of these forms, that to which I want specially to draw your attention as typical of the others, is the contrac- tion of the calf muscles when those in front of the leg are paralysed. We meet with it also as the result of posture, determined by pain, and also of posture determined by simple comfort. It is probable that when there is infantile palsy with this sequel, the contracture of the muscles is facilitated by the fact that they also have suffered a little, that there are 192 THE TREATMENT OF MUSCULAR CONTRACTION. slight interstitial changes, nuclear multiplication, with in- creased development of new fibrous tissue which contracts. But remember it is not a consequence of the palsy; it occurs without it — I mean it is not a consequence of the slight palsy of the contractor muscles ; it is the conse- quence of the persistence, for a long period, of the state that results from the capacity of the muscle for adaptation to posture. It is a grave inconvenience ; it hinders the recovery of use of the legs for standing and walking in patients who have otherwise gained sufficient power. It often needs division of tendons in order to permit the wrong posture to be rectified ; but it does not always need it. In many cases, unless there is shortening of the leg, the persistent, long-continued extension involved in the act of walking elongates the contracted muscle to its proper length. And that is the case in all conditions in which perfect recovery of the damaged structures causing paralysis is possible. It is conspicuous in adults — in whom no hindrance to growth can occur. In multiple neuritis, for instance, we have much greater palsy in the muscles in the front of the leg than in the sural muscles. The latter contract, in con- sequence of the posture permitted by the weakness of the flexors of the ankle. It is a disease which passes away under good conditions ; and however great that contrac- tion, it always in time yields to the attempt to stand and walk. Tenotomy is never needed. Tenotomy is only indispensable when the condition is due to a state from which recovery is impossible. In infantile paralysis it is frequently needed, because infantile spinal paralysis depends upon destruction of the grey matter of the cord ; recovery is impossible then, and without return of power in the opponents the contraction that results cannot be removed. But that which is difficult to remove may be prevented. The great cause of this contraction of the calf-muscles, A METHOD FOR PREVENTING CONTRACTION. 193 permitted by the affection of the muscles in the front of the leg, is the action of gravitation in determining the posture of the foot as the patient lies. The foot " falls," i. e. f always sinks into a position of extension. The calf-muscles become shortened, first by active, extensible contraction, but this becomes fixed by nutritional changes, and ultimately is absolute by structural change. But the falling of the foot need not occur. It is a point of great practical importance, and is almost entirely neglected. There never need be this shortening of the calf-muscles from loss of power in the anterior tibial muscles. It is only necessary to keep the foot always up, and the contraction cannot occur. It may be kept up either by some support below the sole, such as a plank or a thick sandbag, or else by some traction of the foot. Support from below cannot, however, be maintained in adequate continuity. I have tried every form of support that could be devised, and have been compelled to fall back on traction. The force exerted need not be great ; a very gentle elastic traction, long continued, will keep the muscles adequately extended, and prevent the occurrence of this most troublesome result. But the contrivance of such traction is a much less simple matter than it may appear. There must be a place from which it is exerted. It must come from some part near the knee. However gentle it is, by long continuance it becomes annoying, then painful, and ultimately unbearable. It can be borne, indeed, for a very little time when there is tenderness, as there is always in the condition in which it is most needed — cases of multiple neuritis. I tried traction from the upper part of the bed- stead, but the changing posture of the patient made this useless. At last, after many trials of various forms, an idea occurred to me of a plan by which the traction could be exerted from the upper part of the leg near the knee, which has proved perfectly successful. I described what I desired to our excellent resident medical officer, Dr. Whit- 17 194 THE TREATMENT OF MUSCULAR CONTRACTION. ing, who speedily carried it out in perfect detail, with a suc- cess which is most gratifying. I show you here the apparatus. There is a leather sheath for the leg, almost meeting, laced together in front by a cord passing round the hooks. It may go above the knee or stop short of it. There is a similar sheath for the foot, and the two are connected by an isthmus of leather about an inch wide. From rings at the The upper figure represents the shape of the brown paper pattern, and of the leather, cut in correspondence with it. The dimensions are ap- proximate, varying with each leg. D is the narrowing for the knee, that the leather may not press on the bone at the sides. B is the excavation on each side for the ankle with the narrow connecting piece of leather, which is the special feature of the contrivance. This, I to I ^ inches wide, and no longer than is needful, affords, lengthwise, adequate support to make the pull from above act down on the foot-piece on which the pull acts. At H and G are the rings for the cords. The cords are here shewn only as going to the lower rings; if there is contraction of the knee they should be carried through these rings and fastened to the other rings above the knee. If there is no contraction of the knee it is not necessary for the leather to extend above the knee. end of each sheath cords extend, in which India rubber is interposed. A strong, common India rubber band answers perfectly well, for the elastic force, as I said, does not need to be great ; it needs only to be constant. The slightest constant traction is effective, but constancy, under varying mechanical conditions, can only be secured by elastic trac- tion. It is remarkable how well the muscles bear such UTILITY OF THE APPLIANCE. 195 traction if only it is gentle. To the more sensitive skin it soon becomes unendurable, but the impressions from the afferent muscle nerves only act on consciousness, so as to cause " sensation," when excessive, and even long-continued gentle extension seldom causes muscular discomfort. The new feature which it embodies is this. The part of the apparatus for the \zgfrom which the traction is made is continuous with that for the foot on which the traction is made. The two are continuous by a narrow piece of the leather, which will bend so as to permit the traction to increase the flexion of the ankle, and so to overcome the shortening which has occurred, and which yet presents longitudinal resistance. This resistance is enough to make the source of traction, and that on which it is exerted, effectively one. Instead of the pull being from the leg itself, it is from the apparatus that the force acts on. Con- tinuous traction from the leg itself — which, however slight, becomes unendurable by persistence — is thus avoided. The traction is exerted from a part of the same structure as that on which it is exerted. This piece of leather, narrow and short as it is, affords sufficient support, and no pressure is caused on the leg itself. Another advantage of this apparatus is that it is easily made by any practitioner. A piece of leather must be cut out, of the shape which I now show you — two ovals con- nected by a narrow piece. It is best first to take a sheet of brown paper and cut out a pattern to fit the leg and foot of the patient. The leather must then be softened in water, warm at the last, so that the leather is warm when it is applied to the leg. It must be bound on the leg by cord or bandage, and then allowed slowly to dry and harden. The leg should be swathed to prevent chill from the drying. If thought desirable, oil silk may be placed next the skin. When dry and hard it is taken off. It has only to be lined with wash leather, and any shoemaker will insert in it the 196 THE TREATMENT OF MUSCULAR CONTRACTION. necessary hooks and rings. A piece of cord and an elastic band complete it. It may be made in rough but perfect efficiency, with holes and tapes or cords, and a penny elastic band for traction. Mr. Hawksley, the well-known surgical instrument maker of Oxford Street, has made a copy of this apparatus, finished off with all due skill, and therefore more sightly than that which I show you, although it could not be more effective. He has been good enough to place it in his shop for any one to examine. As the only apparatus known to me that is really effective for preventing a grave evil, and even removing it when it has occurred, at least when it has not reached an extreme degree, and as one that is easy to make in perfectly serviceable form, I think it deserves to be widely known. It should be adopted in all cases in which a patient has to be kept in bed, and the weight of the foot leads to persistent extension of the ankle-joint, whatever be the cause of the condition, if there is the slightest evidence of commencing contraction of the extensors, or if there is even reason to suspect that it is likely to come on. But I have found that it produces another effect which I did not anticipate. When the leather is carried a little above the knee, as it is in the form shown in the figure, any tendency to flexion of the knee is counteracted, and with it flexion of the hip. When a disease that can be recovered from leads to such flexion, prevention is of great import- ance. When it occurs as the sequel to extensor spasm, treatment has little chance of doing good. The flexor spasm then says unequivocally, " Hope is at an end." LECTURE XIV THE INFANTILE CAUSES OF EPILEPSY. I. Gentlemen: — Many pleasant and useful memories remain to me of the two years' apprenticeship to a country surgeon by which I commenced the study of medicine. One inci- dent often comes to my mind in connection with the subject of infantile convulsions. I remember spending some hours in a cottage, one summer afternoon, watching a child in " teething fits.". The child was kept continuously in a warm bath. I do not know that my presence there was of much service to the patient, but it is of service to me now. It brings before me, very forcibly, the changes which have come to our conceptions in the thirty years that have elapsed. Fits, associated with dentition, were then always due to " cerebral congestion," excited by the irritation of the backward teeth, and it was necessary to keep the blood in the vessels of the skin that there might be less in the intracranial vessels. I remember well how the colour of the skin testified to the continuous action of the process, and yet how the convulsions continued without the slight- est alteration, although the " derivative " influence of the bath was so marked. It was, indeed, some years before this time that the true relation of convulsions to dentition was made clear by Sir William Jenner. But knowledge filtrates slowly through the profession, or at least its diffu- sion was then slow in rural districts, even among intelligent Clinical Journal, September 5, 1894. 197 198 THE INFANTILE CAUSES OF EPILEPSY. practitioners. The lectures on " Rickets " of Sir William Jenner not only marked, but made a turning-point in pro- fessional knowledge. The facts he demonstrated have now for long become general knowledge, and their chief source is lost to sight. The perusal of his lectures cannot, how- ever, be too strongly recommended, and, happily, their forthcoming reprint will make it easy. No one now dreams of associating the common convul- sions of retarded dentition with congestion of the brain, although a direct influence on the brain by the irritation of the process of teething is still sometimes thought to be the sole element in causation. In general, however, the influ- ence of the process of the eruption of the teeth is relegated to its proper place, as merely a possible excitant in a few cases. I hope it will not be very long before the progress of medical knowledge, and the improvement in medical edu- cation, makes " congestion " everywhere a morbid state that needs definite discernment, and not merely the equivalent for a negation. This may be, perhaps, too much to hope for. The assumption of " congestion " of this or that organ, of the liver in dyspepsia, or of the brain in hypo- chondriasis, is so easy ; its assertion is so precise, so satis- factory, and it enables treatment to produce such a definite result, that the assumption will not readily be regarded as a thing to be proved. As far as can be judged, its end is not yet. Of course, I am speaking only of the active con- gestion, which is constantly invoked to explain that which is not understood, not of the passive mechanical congestion, which is quite a different process, and is ignored when it exists as frequently as active congestion is assumed when it is absent. The service which Sir William Jenner did was to demon- strate that the condition of rickets is a general retardation of development, with various secondary, necessary results, DENTITION CONVULSIONS. 199 and that the association of convulsions is with this general state, of which the backward teething is only one of the manifestations. The convulsions are a consequence of the retarded de- velopment which occurs so often towards the end of the first year. This epoch is one in which development is readily checked. Growth, indeed, goes on, but the progressive development of the tissues is hindered; their elements are still produced, but do not present the progressive develop- ment that should coincide with general growth. It is the epoch at which the character of the food-supply undergoes change, often unwise in suddenness, or does not undergo the change that is natural. It is a period also, when much functional capacity passes into functional use. Any general illness or prolonged digestive disturbance may have, as a result, distinct signs of this malady, but in considerable degree it is the result of preventible causes, which it would be foreign to my present purpose to discuss. The influence of rickets in causing such fits is of great importance in connection with epilepsy, because there is no doubt that such convulsions are a frequent element in the causation of that disorder. They leave behind a residual disposition to the like morbid action, which may be contin- uous in its results or may become active at a later period of life. Between the cases in which dentition convulsions have ceased after a few months and convulsions have recurred as epileptic fits at puberty, on the one hand, and, on the other, cases in which the infantile convulsions did not cease, but continued into life-long epilepsy, we have every gradation. There are cases in which the attacks went on for a few years, ceased for a year or two, and then returned, and cases in which the interval of freedom was five or six years. Indeed, every variation of interval is met with. It is thus impossible to doubt that the dentition con- vulsions are a definite element in the causation of epilepsy. 200 THE INFANTILE CAUSES OF EPILEPSY. So constant, moreover, is their association with the defec- tive development which we call rickets, that it is impossible to doubt that the prevention of rickets would have a con- siderable influence in the prevention of epilepsy. One question to which I desire especially to direct your attention is this. How does the state of defective develop- ment cause the liability to convulsions ? I believe that they can be adequately explained by the simple fact of retarded development. This is conspicuous in all parts of the system. The morbid conditions that are characteristic of rickets can be, for the most part, explained by the per- version of nutrition, which is a necessary result of the retardation of the process. In the nervous system we know that, at birth, the state of the various structures differs, that some are structurally perfect, while in others a considerable amount of developmental change has still to occur before full functional capacity can exist. It is not necessary to describe the facts in detail, because the gen- eral character of the relation is alone important. The nerve structures which are lower in function, and in part in posi- tion, are developed before the higher. This is true of the motor elements and reflex centres. The lower centres are under the control of the higher, and their activity in early infancy, when the higher centres are less developed, is manifested in the restless, aimless movements of the child, and the conspicuous activity of the reflex processes. At birth these are developed to such a degree as to enable the mouth to suck, and the thorax and larynx to co-operate in producing too-familiar sounds. The process of develop- ment goes on, and as structure is complete, function be- comes established in more perfect degree. But the estab- lishment of perfect function follows structural completion, and it is long before the influence of the relative develop- ment ceases to have an influence on the action of the nervous system. Hence it is that all through infancy and DEFECTIVE DEVELOPMENT. 201 early childhood reflex processes are so active. Their early predominance is only slowly reduced into subordination as the function of the higher centres becomes established in effective degree. It must be remembered that the motor centre in the cortex of the brain, from which impulses proceed directly to the spinal cord, is a centre relatively low, and its control by higher structure is a comparatively late event. Not only does this relativity of development and function explain the excessive spontaneous and reflex movement of infancy, but it explains to some extent the readiness with which the motor cortex acts in the sudden, violent manner which gives rise to convulsions. It is sudden and violent, although incomparably slighter than the convulsions of later life. It is important to remember the fact. Without doubt many of the attacks in infancy which consist only of tonic spasm, would in an adult con- sist of violent clonic spasm as well as of tonic spasm. The same activity in the well-developed centres for the larynx gives rise to the familiar " laryngismus stridulus." The hindrance to development, which acts upon all structures, of necessity has most effect upon those which are least complete. Where structure has become perfect, and function is established and is being fixed by use, its influence has little effect. Where structure is not yet perfect, and where it has just become perfect, but function is only just assuming its adequate form, the retardation has an influence in proportion to the imperfection of structural and functional development. It must be remembered that the retardation has an influence on growth and nutri- tion, and the effect on the finer nutrition which is related to function may be considerable, even after structural growth is at an end. Hence the relation of the lower cen- tres to the higher, the excessive activity of the lower, which is so conspicuous in early infancy, is reproduced by the influence of rickets. The excessive activity of early life, 18 202 THE INFANTILE CAUSES OF EPILEPSY. moreover, becomes definitely morbid insubordination. The greater capacity for functional action of the lower centres needs the proportioned control which should come from the due development of the higher, and this is wanting. Hence the uncontrolled activity is manifested by morbid action, possibly contributed to by the general failure of nutrition, which involves a less perfect form of action even in the centres in which development is complete and function established. Thus we can understand, without difficulty, the morbid action of the reflex centres which causes such a state as the contracture in the hands and feet and the laryngismus stridulus of rickets, and also the occurrence of the characteristic convulsions which prob- ably depend upon the lower cortical centres of the brain, those of the motor region. With the restoration of the normal process of develop- ment which comes from the treatment of the general state, the higher centres acquire increased control, and, in most cases, the tendency to convulsion ceases. But the sudden spontaneous overaction has left its residual effects. These may be so strong that there is no cessation of the fits, or only a brief cessation, or a recurrence under favourable conditions. Thus we have the series of cases to which I have already referred, which demonstrate the relation of such convulsions to ordinary epilepsy. The residual influ- ence, with its consequent tendency to persist or to recur, is certainly influenced by the mysterious functional state which is due to inheritance. The precise nature of this we do not know ; we can only guess it from its effects. It seems to be a defect in the chemical composition or the molecular structure of the nerve centres, in consequence of which nerve force is liable to be released without the stimulus which alone should excite this liberation. The tendency is seen in its special and isolated operation only after infancy is passed, but during this period it seems to INHERITED TENDENCY. 203 co-operate with other conditions, and seems to increase the relative liability to convulsions in rickets. A slight degree of hindrance to development suffices then to give rise to the convulsions. Still more markedly does the inherited tendency dispose to the residual disposition to recurrence which is left behind by every fit, to which the persistence of convulsions is due, and still more again does it dispose to a recurrence, in consequence of this residual effect (which never entirely ceases) at some period in later child- hood or in youth. All parts share in the defective nutrition and in the hin- drance to their development. The bilateral symmetry of the frame involves the equal affection of both sides, and hence it is that the convulsions which are due to these causes are general. This is an important fact, as I shall hope to show in the next lecture. The persistent convul- sions which may earn.- on the malady from infancy to adult life are of the same general character. Often, indeed, the establishment of the control of the higher centres, when it does not arrest the process of discharge, modifies its form in a way we cannot yet understand, and attacks of " minor " character replace the convulsions, to be succeeded, at the time when epilepsy is most prone to develop, by convulsive attacks of greater severity. The convulsions of rickets, as I have said, in the slightness and tonic character of the spasm in a large number of cases, are nearer to minor than to major epilepsy. This fact is important because attacks of petit vial may succeed the convulsions of rickets and be unnoticed until spasm is again added. I need not say anything of the treatment of these con- vulsions, because it is well known to you all. But one point it may not be superfluous to mention. The bromide which is given to check the tendency to convulsions while the general treatment is re-establishing the proper course of development, should not be too hastily discontinued 204 THE INFANTILE CAUSES OF EPILEPSY. when the fits cease. The residual tendency must be remem- bered, and this may be lessened, I believe, by continuing the treatment for some months after the attacks have ceased, and the time should be still longer when there is reason to think, from the fact of inheritance, that this ten- dency is likely to be profound. LECTURE XV. THE INFANTILE CAUSES OF EPILEPSY. II. Gentlemen: — The convulsions which we considered in the last lecture as a cause of epilepsy — those that occur during the period of dentition and are the result of the hindered development of rickets, are general. The later convulsions of epilepsy, which are developed from them, or disposed to by them, are also general. They are the char- acteristic convulsions of what is called " idiopathic " epi- lepsy. The convulsions of rickets are moderate in severity, and when they are continued into epilepsy, the attacks often become slighter, until they are merely minor attacks with loss of consciousness ; there may be only the slightest indication of spasm, or there may be no trace of spasm after a time. Such minor attacks may continue for a few years, and then again muscular contractions may be combined with the loss of consciousness ; increasing in severity, the seizures gradually assume the features of severe epileptic attacks. But the practical lesson to be learnt from the facts which we have considered is this : in a case of epilepsy which dates from convulsions in the period of the first den- tition, moderate in degree and general, whether the attacks have been continuous or interrupted by a few years' free- dom, you are justified in considering that the case is one of idiopathic epilepsy, founded on the convulsions of rickets. But there are two other forms of epilepsy in which the Clinical Journal, September 12, 1894. 205 206 THE INFANTILE CAUSES OF EPILEPSY. disease can be traced to infantile convulsions. The first of these is especially important. It differs from that we have considered in three features. The epileptic convulsion is one-sided, at any rate when moderate in degree, and if there is an aura, it is in one limb or in the face on one side. The spasm can generally be observed to commence locally in the hand or face. When both sides are convulsed, one is affected before the other, except in the most violent attacks. In these there may be no perceptible interval between the affection of the two sides. Inquiry regarding the character of the infantile convulsions will usually elicit the fact, when their character is known, that these also were confined to one side. The second feature is that the first attack of in- fantile convulsions was of great severity. Often, indeed, there was a series of convulsions, one after another, for several hours. Sometimes such a series was the only infan- tile attack ; often initial attacks were followed by others for a month or two, which ceased to recur when a few years had passed. Complete continuity of occurrence from in- fancy to adult life is less common in these than in the cases first considered. The third feature is that the first convul- sions often occur during some acute illness, or soon after a fall, or in a state of general physical prostration. This is not always the case, nor is it always true that there are several consecutive convulsions at the onset. But the first fit is seldom slight in degree, and, if the facts can be accu- rately ascertained, the first fits are known to have been unilateral. The unilateral character of the convulsion means uni- lateral instability of the motor structures of the brain. The commencement in one limb means local instability in a certain part of these motor structures. This part differs considerably from the rest in its morbid functional ten- dency. Such local change excludes a general cause, which would act on the whole brain. The suddenness of the ORGANIC DISEASE AT THE CORTEX. 207 onset indicates a sudden development of the instability — that is, a sudden change at one spot. This is, however, equivalent to the assertion that the convulsions are due to a local lesion in the motor structures, because a sudden change means organic disease. Without doubt, this is the fact in the class of cases we are considering. There is organic disease at the cortex of the brain. Its occurrence is indicated by the initial convulsions. The effects of the process on the adjacent tissue, which is slightly changed, is such as to induce a permanent alteration in nutrition and function. In consequence of this, the structures "dis- charge " from time to time. The disease may be either in or only near the motor region. If it is in the motor region, and is more than minute, there is loss of power — hemi- plegia — the amount of which depends on the extent of the lesion. It may be such as to be persistent through life, or it may be transient and pass away in a few months. When the disease is only near the motor region, any initial loss of power may be too slight to be recognised. In an infant slight weakness on one side usually escapes recognition, and considerable paralysis may be unnoticed if the child is at the time prostrate from some general illness. When there is no lasting palsy, and when there is no history of initial palsy, we may nevertheless obtain some indica- tion of it. You are familiar with the aspect of cases of infantile hemiplegia, you know that they frequently present a peculiar inco-ordination of voluntary movement, especially conspicuous in the peculiar ataxy of the hand. When this is present in the arm, you will seldom fail to discern that in a slight smile there is more movement of the face on that side than on the other. You may often detect this when there is no other indication of hemiplegia. Corresponding to the side on which the convulsions begin, or to which they are limited, it is of very definite signifi- cance. Its significance, indeed, is far greater and more 208 THE INFANTILE CAUSES OF EPILEPSY. trustworthy than you would conceive from the trifling nature of the symptom. It is only in a minority of these cases that we find persistent hemiplegia. The tendency to " discharge " in the motor structures is due to damage ; the hemiplegia is due to destruction. The tissues which are destroyed cannot give rise to convulsion. Hence, when the damage is too slight to cause persistent loss of power, or is situated outside the motor region but near it, so that the nerve elements in the motor region are damaged but not destroyed, the tendency to convulsions is met with in chief degree. The actual lesion in these cases is still a matter of con- jecture to an almost strange degree. Perhaps, however, it is not so strange. The lesion is comparatively slight, and vitality in early life is strong and has a marvellous power of resisting the influence of local disease, readily as it often yields to a general morbid state. Although the motto of the Pathological Society asserts that death is not silent, the truth is relative. Death has her secrets still, and, when death does not unfold the closed roll, the secrecy often baffles the most careful scrutiny. Not only is scrutiny baffled, but effort is tantalised, and the region of inference is a field of contest. Hence, opinions differ widely as to the character of such sudden lesions. Doubtless we shall not have to wait much longer before facts assist us. At present the opinion which seems to myself to deserve most weight is that there is a sudden occlusion of a small surface vein by clot, with the consequent intense congestion and hemorrhagic softening of the region of the cortex. In the softened region the nerve elements are destroyed ; an in- durated contracted area ultimately represents the disease. There is not usually the actual softening that is met with in arterial occlusion, or in traumatic injury, but on the margin of the chief destruction, in every form, there is a region of slighter damage, and it is, no doubt, from this that the discharges proceed. THE RESIDUAL DISPOSITION. 209 Remember that the function of the motor nerve struc- tures depends on their ability to release a considerable amount of nerve force, without any appreciable interval of time after the stimulus reaches them. There must be a large amount of "latent energy" in the structures, which the stimulus at once releases as active force. In the struc- tures in which the nutrition is deranged by the previous process, the energy is not retained in its latent form until the proper stimulus releases it; it accumulates and escapes without any influence, that we can recognise, acting upon the structures to cause the release of the force. This we call " discharge." The process, once set up, spreads widely in proportion to its energy. If this is slight, it is limited to the part in which it began; if greater, it spreads through that side of the brain ; if greater still, to the opposite hemisphere. And note this — wherever it occurs, wher- ever it spreads — in every place the discharge leaves behind it a residual state disposing to its repetition. This resid- ual disposition is, in health, the foundation of all our habits, of all our education, of all our acquirements, muscular and mental. It is the physical basis of memory, and it is the basis of such functional disease as that which we are now considering. It facilitates the spread of the discharge, so that the convulsions after a time, become general more readily than at first. Remember this, because we shall presently see that it has an important practical application. Lastly, there is yet a third class of cases in which epi- lepsy has its origin in infancy. There are cases in which the first symptoms can be traced back, not only to infancy but through infancy — back to the earliest period of sepa- rate existence. They are cases in which there were con- vulsions during the first two or three days of life, or, at least, convulsive twitchings and other indications, such as difficulty of swallowing, of grave impairment of the brain. These cases occur in the children who are the " first-born," 210 THE INFANTILE CAUSES 7 EPELEPS* who enter the world with greater difficult than those which follow them, and whose birth, as : can generally ascer- tain, was long and tedious, often needing the aid of instru- ments to terminate it soon enough to save the new life. You know that such children pass out of infancy into an imperfect physical childhood, with weakness of the legs, sometimes also of the arms, with irregularity* of the move- ments of the hands, and often also with convulsions. You have heard the cases described, and they have doubtless been pointed out to you as examples of the condition for which I have proposed the name of "birth-pals The symptoms are the result of damage to the cortex of the brain, commonly the effect of meningeal hemorrhage. When the damage is slight there may be no symptoms of paralysis, but its effect may be manifest by the occurrence of convulsions, either in childhood or in later life. Do not forget this because, although it is not common for epilepsy to be the sole effect, some cases, otherwise mysterious, can be thus explained. You will meet with fev. cases of epilepsy in which the fits can be traced back to infancy which do not fall into one of these three classes, and you will seldom have any difficulty in discerning to which class a given case belongs. Of the small remainder, most are associated with con- genital mental defect, and occur in families with neurotic disposition, and in which other instances of idiocy can be heard of. In these the malady must be ascribed to a congenital imperfection of the nerve tissue, of which the convulsions and the mental defect are both consequences. Such cases are usually self-evident In the others, those which I have considered in this and the last lecture, the diagnosis is determined by the character of the fits and of the initial convulsions, and the circumstances under which these occur. But there is one important effect of the in- fluence of attacks in disposing to their recurrence. A e IS OPERATION ADVISABLE 211 long continuance of convulsions, which begin locally and spread to the other side, the spread becomes facilitated by the residual disposition, and its influence on the whole brain may be such that minor attacks occur quite like those of idiopathic epilepsy. They are typical attacks of the form which is understood by the term petit mal. You must not, therefore, let the occurrence of these have weight against other evidence that the original cause was a local lesion. A still more important question is connected with this residual disposition, and its action in augmenting the ten- dency to discharge of structures far away from the initial seat of the process. The question to which I refer, is that of the removal of the part from which the discharge pro- ceeds, the extreme instability of which is the source of each convulsion. The operation seems at first sight a promis- ing one, but the condition with which we have to deal differs much from the state in which there is some remov- able disease outside the nerve tissue, and acting upon it, such as depressed bone. Should the disease proceed from tissue which has recovered from partial damage, and which is in the vicinity :