Digitized by the Internet Archiv in 2010 with funding from H dee University of Toronto I) hy, i ng ; =“ - ’ oo. x 4 é ‘ = <= ! ; ‘ THE HARVEY LECTURES Delivered under the auspices of THE HARVEY SOCIETY OF NEW YORK Previously Published. FIRST SERIES . . 1905-1906 SECOND SERIES . 1906-1907 THIRD SERIES. 1907-1908 | FOURTH SERIES . 1908-1909 | « eee 231 The quadrilateral near the isthmus: .:.........:..+.: 75+ ape 231 Motor aphasia without involvement of corpus striatum.............. 232 Transcortical motor aphasia:....3. 2.0 -45.04-2 oe =) eee 232 Pure motor aphasia: :. ...425 biel. epee. eye} = or 233 Transitory motor aphasia with extensive softening of Broca convolution 233 Transitory motor aphasia with a slit of degeneration in subcortex of left anterior central gyrus: . ..:... -. 2... 2.22.55. a2 +. 4 eee 234 Cortical softening of right Broca region and adjoining part of insula in a left-handed. woman: 22: )2.5..2)....-. 5s. 820 2 234 Negative cases. - 2... 0.6 Fives eds soe cen ee ee 235 Case of permanent word-deafness and paraphasia.................... 236 Transcortical sensory aphasia.......5..25..-..5.5.-.5: eee ee 236 Case of anomia and paraphasia. ... 22.0... 22: a2 is ..2 2 so eee 237 Occlusion of retrocentral vessels. .. 2... ...0. 222... 08-42 ee 237 Gordonier’s case of agraphia; frontal lobe tumor.................... 238 Ferrier’s map of cerebral centres... 2... 2.252 95... 5< +6 ae 238 Maps of the cortex according to differentiation of structure by Brod- mann, Campbell, Flechsig, and E. Smith. ........: 5.2252. 238 Vogt’s record of electrical stimulation in monkeys................... 239 Integrity of the knee of the internal capsule in the brain.............. 239 Brodmann’s map of lateral view of hemisphere............:......... 239 Brodmann’s map of mesial view of hemisphere...................... 240 Degenerated fibres of sagittal marrow near a wound of the brain...... 240 Defects of visual field of supposed origin in lesions of calearine cortex... 241 Typical occlusion of the posterior cerebral artery with right homony- mous: hemianopala, 05. io. ek eee les « See a er 241 Mind-blindness in bilateral symmetrical occlusion of the postparietal ATGOLY ssc eidle ieee et baw NS gy Snelite se eee ae 241 Brodmann’s subdivision of the island and transverse temporal gyrus.. 242 Bilateral destruction of transverse temporal and auditory zone; com- plete deafness with preservation of reading. ....-.<.....25 eee 242 Diagram of lesions of corpus callosum, cortex, subcortex, and capsule. . 243 THE PROBLEMS OF EXPERIMENTAL NEPHRITIS * PROF. RICHARD M. PEARCE University and Bellevue Hospital Medical College, New York UR present knowledge of nephritis is the result of the methods of clinical observation, pathological anatomy, and experimental pathology, successively applied. By means of the first of these, Richard Bright, in 1827, demonstrated that albuminuria and dropsy had an intimate relation to certain pathological changes in the kidney. Studies in pathological anatomy during the following years led to the differentiation of several types of nephritis, and, finally, to a classification based on morphological alterations. I do not think it an exag- geration to say that clinical observation has added little of essential importance to Bright’s original conception of eighty years ago, or that pathological anatomy has added little to Weigert’s classification, which has been generally accepted for thirty years. Bright’s views, it is true, have been amplified, certain phases of the relation of renal disease to cardiovascular disturbances have been more clearly understood, and much negative evidence concerning uremia and cedema has accu- mulated; but little has been added by clinical methods to our knowledge of the interrelation between a kidney lesion and its manifestations. The methods of pathological anatomy have given a classification, based on careful study of the gross and minute lesions of nephritis, and with these have been correlated in a more or less satisfactory way clinical manifestations and changes in the urine. This most important period of anatomical study began in 1851, with Frerichs, who considered all forms of nephritis as stages of a single process, beginning as an acute * Harvey Lecture, Delivered Oct. 30, 1909. bo 17 18 HARVEY SOCIETY nephritis and ending as the small granular kidney; the period terminated with Weigert, who, in 1879, demonstrated conelu- sively that Frerichs’ stages do not represent the successive changes of a single lesion, but are distinct types of nephritis, caused by various injurious substances acting during varying periods of time, and representing the varied reactions of kidney tissue thus influenced. Weigert’s view is the one held to-day. More recent studies by improved histological methods have added to our knowledge concerning certain details, especially in regard to the glomerular changes, the sequence of lesions, and certain unusual types of nephritis, but the methods of pathological anatomy offer no promise of an interpretation of the important problems of this many-sided disease. The application of the experimental method to the study of renal disease is not a recent development. For many years experimental lesions of the kidney have been utilized, and with gratifying results, in the study of the sequence of the histolog- ical changes occurring in nephritis. With such studies, essen- tially anatomical in nature, have been combined, in recent years, investigation by methods which allow an interpretation of changes in function, upon which morphological studies throw no light. Such investigations necessarily demand the methods of chemistry and physiology ; and we have witnessed in the past few years the curious spectacle of pathologists turning from the methods in which they were trained to those of the physiolo- gist and chemist in which presumably they had, originally, little or no training. Investigation by such methods is termed ‘‘experimental pathology’’ merely because the pathologist, despairing of the anatomical method, has seen fit to adopt them in the study of altered function. It is to such methods that we must look for an advance in our knowledge beyond that which has been possible by the methods of clinical medicine and patho- logical anatomy; and if the pathologist is criticized, as fre- quently happens, for appropriating the methods of other sciences and for applying to the field of endeavor thus created the term ‘‘experimental pathology,’’ it is sufficient to point out that the physiologist and the chemist, as well as the pharmacologist PROBLEMS OF EXPERIMENTAL NEPHRITIS 19 who shares the same methods, have with few exceptions limited themselves to the field of normal function. That nephritis has been one of the principal objects of attack by these methods is in part due to the importance of the disease, and in part also to the fact that the kidney lends itself very readily to experimental study. And, moreover, although the results of experimental study may not always be applied to explain disease in man, it must be evident that, owing to the peculiarities of the structure and function of the kidney, results of experimentation with this organ have a very definite applica- tion. Thus, some aspects of etiology, the almost specific action of certain substances in picking out certain kidney structures, the character of acute lesions and the relation of these to chronic lesions, questions of repair and regeneration, the matter of cast formation and the source of albumin, are problems which, when elucidated by animal experiments, can readily be transcribed to explain similar problems in human nephritis. But aside from these, the experimental method offers hope, in part already realized, of a solution of the more prominent problems of renal cedema, of anuria, the question of the relation of renal disturb- ances to hypertension and heart hypertrophy, and the most important, though at present the most hopeless, problem of uremia. Here I may at once eall your attention to the fundamental problem of experimental nephritis, that is, the influence of the glomerulus as contrasted with the influence of the tubule. This enters into all phases of renal pathology, in some partially elucidated, but in most still a matter of doubt and speculation. The dual structure of the kidney is responsible for the difficulty which we have of interpreting the physiology as well as the pathology of this organ. We are familiar with glands in which different types of cell are concerned in the elaboration of differ- ent chemical substances, and with those in which cells are modified to produce an internal, as contrasted with an external secretion, but the kidney stands alone as an organ with two widely different structures, having for a common object the elimination of a single fiuid representing the products of 20 HARVEY SOCIETY metabolism. This is not the place to discuss the significance of this structural peculiarity and its bearing on the function of the kidney, though it must be considered in what follows. It may be permitted, however, to point out here that the glomeru- lus, as has been emphasized by Beddard, is a structure without analogy elsewhere in the body except, perhaps, in the choroid plexus of the brain; and that the urinary tubule differs from all other gland tubules in its length and complexity. On these peculiarities of structure, coupled with the peculiarities of the renal circulation, depends the power which the kidney has to remove from the blood-stream the fluid and solids which con- stitute the urine. If we disregard the one synthetic process of which we have positive knowledge, the formation of hippuric acid from benzoie acid and glycin, the essential function of the kidney is one of elimination, with the important feature that the resulting fluid contains all of the soluble components of the blood except its protein constituents and dextrose—in a differ- ent percentage, it is true, but still the same substances. If we accept departures from normal elimination as evidence of disturbance of kidney function, the problem of experimental nephritis is to determine the part played in this disturbance by glomerulus and tubule, respectively. This may be done by the use of physiological methods which graphically demonstrate alterations in vascular reactions and by comparing such results with those obtained by chemical study and eventually corre- lating both with the anatomical changes. By such studies of simple phases of the problem of nephritis, enough has been accomplished to warrant their continuance with the prospect of adding essentially to our knowledge of renal pathology. The study of experimental nephritis may be expected, how- ever, to do more than explain the sequence and significance of pathological changes. By producing lesions which affect only certain structures as the glomeruli or the tubules, or but certain portions of the tubules, we may expect not only to solve some doubtful points in the physiology of this organ, but also to obtain data of considerable importance to the pharmacologist and therapeutist, thus bringing the work home to the clinician. PROBLEMS OF EXPERIMENTAL NEPHRITIS = 21 As a single example may be given the study of the effect of diuretics on the diseased kidney as compared with their effect on the normal. Our knowledge of the latter action is fairly com- plete, but we have very little knowledge of the former. The study of vascular dilatation and contraction in the kidney, the elimination of water, the general composition of the urine, the chloride-regulating mechanism and many other points, in dis- tinetly tubular and distinctly glomerular forms of nephritis, which we are now able to produce, should yield practical infor- mation of great value. Some information in regard to these matters we now possess, but before it can serve as working knowledge, extensive chemical and physiological studies of various forms of nephritis must be made from the pharmaco- logical point of view. I have gone somewhat into detail in this introduction, not only for the purpose of demonstrating the value of the study of experimental nephritis, but also for the purpose of showing that the results of such study are of interest to every one con- cerned with the problems of normal and abnormal physiology— to the physiologist, the chemist, the pathologist, the pharma- ecologist, and the clinician. And, in order to maintain interest, if it has been aroused, I shall deal briefly with the methods of inducing nephritis, the character of the acute lesions, and the relation of these to chronic lesions, attempting to set forth clearly the types of experimental lesions known as tubular and glomer- ular. Time thus saved will be devoted to the more interesting questions of altered function. ETIOLOGY AND CHARACTER OF THE EXPERIMENTAL LESIONS In speaking of the etiology of nephritis in man, excluding, of course, lesions due to the localization of bacteria, we use, owing to our inexact knowledge, the phrase ‘‘soluble toxie sub- stances reaching the kidney through the cireulation.’’ So, in experimental nephritis a direct nephritic poison must be capable of absorption, of solution in the body fluids and of causing injury to the renal cells when given in doses so small as not to cause death through its other actions. An indirect poison acts Be HARVEY SOCIETY through products formed by blood or tissue destruction, as with the hemolytic poisons; here the action on the kidney is second- ary. If we exclude Siegel’s experiments on the production of nephritis by the application of cold, all forms of experimental nephritis are caused by substances falling in the above elassifica- tion. ; According to Sollmann, all metals, so far as they have been studied, cause nephritis, though some act only in corrosive doses or when given intravenously. Other nephrotoxic substances are aloin, coal-tar products, alcohol, anesthetics, oxalates, can- tharidin, essential oils, snake venom, ricin, abrin, bacterial toxins, hemolytie poisons, and nephrotoxic immune serum. Of these some act diffusely, while others affect the tubules or the glomeruli separately. Only such as have a more or less definitely circumscribed action are of value in producing exper- imental nephritis. Thus in the group affecting tubular epithe- lium with httle or no primary glomerular injury, we may place, as most important, uranium nitrate, the chromates of potas- sium and of ammonium, and corrosive sublimate. Of those affecting glomeruli especially, the more important are arsenic, cantharidin and snake venom. All of these latter have some slight effeet on tubular epithelium, probably secondary to cir- culatory disturbances dependent on the glomerular injury, but the latter lesion is so marked and so evidently primary that they are usually referred to as glomerular poisons. Another agent of value in experimental work is diphtheria toxin which combines glomerular and tubular injury. All of these cause the appearance of albumin and easts in the urine; only uranium nitrate produces cedema. Although I have, thus far, used the terms ‘‘tubular’’ and ‘‘olomerular ’’ in reference to these poisons, they may more definitely be denominated, respectively, ‘‘ epithelial’’ and ‘‘vas- cular’’ poisons. Until recently this division was made on ana- tomical grounds, that is, on histological evidence of degenera- tion, necrosis, exudation or cell proliferation, but the study of nephritis by physiological methods has brought out evidence of the existence of functional glomerular injury of extreme grade PROBLEMS OF EXPERIMENTAL NEPHRITIS~= 23 accompanied by little if any anatomical evidence of vascular lesion. These methods have also shown that nephritides due to agents formerly supposed to act only as tubular poisons, pre- sent, in the late stages of intoxication, definite evidence of vascular incompetency. It is necessary therefore to describe briefly the lesions pro- duced by the more important nephritic poisons. This descrip- tion will be limited to those poisons especially discussed in this address. It, however, by no means exhausts the list of sub- stances which may be used. The anatomical changes due to uranium and to the chro- mates are, in the early stages, confined essentially to the tubules, especially the convoluted tubules, and consist of granular or fatty degeneration and definite necrosis often affecting large groups of tubules. Corrosive sublimate causes similar lesions involving especially the ascending loops of Henle and charac- terized also by the deposition of lime salts. In these typical forms of tubular nephritis no anatomical lesions of the glomeruli are evident in the early stage, but in the late stages an ill-defined thickening + of the capillary walls may sometimes be seen and evidence of vascular disturbance is shown by phys- iological methods. The glomerular form of nephritis varies. Arsenic, which acts through paralysis of the capillaries, causes little or no ana- tomical change in the glomeruh. The capillary loops may show slight thickening, the vessels may be overfilled, and the nuclei may stain peculiarly. Exudate into the glomerular space is usually absent, though a shght amount of coagulated serum may be present. By physiological methods, however, it is shown that despite the absence of anatomical lesions, serious vascular injury is present. Tubular involvement is slight and usually difficult to demonstrate. Cantharidin causes a glomerular nephritis involving both the tuft and the capsular space. The lesions of the capsule have been variously described as desquamative, as consisting of * In the uranium lesion, Christian has described hyaline droplets in the capillary loops. 24 HARVEY SOCIETY a leucocytic exudate, and as due to the presence of epithelial cells pushed up into the capsule from the convoluted tubule. Lyon has recently emphasized this latter view and also describes degenerative changes in the convoluted tubules and ascending loops of Henle with necrosis of the latter. Functional tests demonstrate serious vascular injury. The venom of the rattlesnake, as I have recently determined, causes a very remarkable glomerulonephritis of the exudative type. Single large doses or repeated small doses cause an exudation of serum and fibrin in both the capsular space and the glomerular tuft. This exudate is usually but not always hemorrhagic. Leucocytes are not prominent, but occasionally are present. The tubular changes are slight or entirely absent. Diphtheria toxin is the best example of those poisons which combine both epithelial and vascular injury. Hyaline thrombi are found in the glomerular capillaries and small arterioles of the cortex in acute and intense intoxication. The vessel walls show hyaline changes and, in the later stages, cyst-like hemor- rhages in the tuft (Lyon). Leucoeytes are abundant in the tuft and slight necrosis may occasionally be seen (Flexner). With these changes are found extensive degenerative and necrotic lesions of the convoluted tubules and the ascending loop of Henle. Undoubtedly, the lesion in both tubular and glomerular nephritis occurs in that portion of the kidney through which the poison is eliminated, though this has not been definitely demonstrated except in the case of uranium.’ From this, and our knowledge of the elimination of iron through the convoluted tubules, it seems probable that nephritis, due to the salts of various metals, is an indication of injury at the point of elimination. The peculiar involvement of the loops of Henle in the corrosive sublimate lesions supports this * Schneider working with Petromyzon fluviatilis injected uranium solution in the muscle of the back, and also subcutaneously, and found that by the use of a fixing fluid containing potassium ferro- cyanide, picric acid, and hydrochloric acid, the uranium was precipi- tated as a brownish-yellow deposit in the epithelium of the tubules. Ts PROBLEMS OF EXPERIMENTAL NEPHRITIS~= 25 view. It is not too much to hope that by careful study of such localized lesions experimental nephritis may eventually con- tribute to our knowledge, not only of altered function, but also of the normal physiology of the kidney. The glomerular lesions likewise must be considered as a special manifestation of a general injury to capillary struc- tures; the intensification of that action in the glomerulus being due to concentration of the poison at the point of elimination. At present, then, we are familiar with several poisons which affect either tubule or glomerulus, respectively, the injury being recognized sometimes by anatomical changes, sometimes by functional disturbances, and sometimes by both. The futility of judging of altered glomerular function by anatomical changes alone is best illustrated by Takayasu’s his- tological study of the kidneys utilized by Schlayer and Hed- inger in their investigations of disturbances of function in various forms of nephritis. This work will be discussed in detail later. Here it is sufficient to state that in arsenic and cantharidin nephritis characterized by constant and severe disturbance of vascular reactions, the glomeruli presented exu- dative lesions in only 2 per cent. of the kidneys examined, and this anatomical condition reached a degree comparable to the functional disturbance only in those kidneys showing total insufficiency. Proliferative lesions of tuft or capsule were not demonstrable. The only frequent lesion was increase in size of the glomerular nuclei and an indistinct outlining of the capil- lary walls, due, apparently, to an ill-defined thickening. The nuclear changes, moreover, occurred in tubular as well as in glomerular nephritis. Such results would appear conclusively to establish the possibility of serious functional disturbance with little or no evidence of structural lesion. To this problem I shall return in the discussion of altered function. This brief description summarizes the more important types of acute injury caused by irritants acting directly on the kidney. It remains to discuss the relation of these to the pro- duction of lesions which may be termed chronic nephritis, or are accompanied by manifestations characteristic of the chronic 26 HARVEY SOCIETY disease in man. The production of such a condition has been the object of nearly all work on experimental nephritis, and until recently with no success. Lyon, who worked with can- tharidin, diphtheria toxin, and corrosive sublimate, with the object of following acute lesions to their termination in chronie, found that acute lesions rapidly disappear and that the kidney returns to normal. Such has been the experience of many other investigators, and it has incidentally served to strengthen the clinical observation that an acute nephritis, if the causative agent be no longer active, may go on to cure without the devel- opment of subacute or chronic lesions. This, however, is a phase of experimental nephritis which, in view of the very recent statement of Miiller based on clinical observation and supported by the pathological studies of Léhlein, should again be investigated, and especially with regard to the matter of glomerular lesions. Muller expresses the opinion that a chronie nephritis may be the result of an acute lesion with a progres- sive course marked by acute exacerbations, or, on the other hand, there may be complete cessation of symptoms for many years with eventually a contracted or indurated kidney due to healing by scar formation. Lohlein, as the result of a very careful study of selected material, has shown that many individuals dying of chronic nephritis present a definite history of an acute nephritis, fol- lowed by a quiescent period of several years, before the appear- ance of the chronic lesion responsible for death. His conelu- sions are based more especially on the kidneys of searlet fever and acute coceus infections, in which he found inflammatory elomerular changes which seemed to be the starting-point of the fibrotic tufts and thickened glomerular capsules characteristic of the later developing chronic nephritis. Such observations are not new. Others have reported isolated instances of a chronic nephritis following the acute lesion of searlet fever. Thus Handford describes such a condition after scarlatinal nephritis in a child 12 years old, in whom the chronic condition developed three years after the acute; and Councilman describes a chronic interstitial nephritis with heart hypertrophy following scarlet PROBLEMS OF EXPERIMENTAL NEPHRITIS = 27 fever, in a child of 2 years. Similar findings have been reported by Leyden, Mann, and others. Lohlein’s extensive and thor- ough study, however, brings the problem once more prominently before us, and, coupled with the observation of Muller, makes it one of much importance. It would seem possible that by the use of a substance like venom, which acts as a definite glomeru- - lar poison and causes exudation and very striking endothelial destruction, experimental evidence of chronic nephritis fol- lowing a single injury could be added to the clinical and patho- logical evidence now at hand. Despite this possibility, it must be admitted that the experi- mental study of nephritis supports the more common concep- tion of the etiology of chronic nephritis in man, that is, that it is a gradually developing lesion due to the long-continued insidious action of some ill-defined toxic substance. With the possible exception of the recent experiments of Dickson, the results obtained have been neither constant nor of such nature as to justify the term of chronic nephritis. Certainly if we take as a criterion a persisting lesion of the kidney charac- terized during life by elimination of albumin and casts, and histologically by changes involving glomeruli, tubules, and con- nective tissue, nearly all experimental efforts can be excluded. If we include cedema as a necessary corollary, chronic nephritis has not been produced experimentally. Some of the methods which have resulted in lesions approaching chronic nephritis are, however, worthy of mention. Ophitils, who investigated this subject, came to the conclusion that the best results could be obtained with lead, and, by the prolonged administration of a lead salt, he produced in guinea-pigs and dogs a definite sclerosis. The urine, however, did not contain albumin and casts. The same objection holds for experiments with many of the other metals (Petroff). The experiments of Ehrlich and of Levaditi with vinylamin show that the primary necrosis of the papilla of the kidney caused by this substance may be followed by cortical injury with increase of connective tissue and considerable contraction. In a few of these experiments, in which mice were used, cedema, 28 HARVEY SOCIETY hypertrophy of the left ventricle, and albuminuric retinitis were observed, with characteristic changes in the urine. Such changes, however, were not constant. The value of these exper- iments, moreover, is slight, for the diffuse nephritis followed destructive lesions of the papilla leading to mechanieal obstrue- tion, and were not due to a primary injury of cortical struc- tures caused by a circulating poison, though it must be admitted that Lindemann has described the production of such injuries by the use of this substance. Occasional positive results have been obtained with a variety of substances, as cantharidin (Aufrecht), oxalic acid and oxamide (Ebstein and Nicolaier), potassium chromate (Ophiils), and uranium nitrate (Siegel). I have myself found, in the course of a study which had for its object the production of cedema in the dog, a typical contracted granular kidney as the result of continued injections of potassium chromate and nephro- toxic immune serum. Chronic lesions, however, cannot be pro- duced constantly by such methods and occasional positive find- ings, in view of the frequency of spontaneous lesions, must be regarded with suspicion. Or, to look at it in another way, these oceasional positive results may have been due to the acci- dental presence of some secondary factor, as some metabolie or circulatory disturbance, necessary to the production of chronic nephritis. It was with this possibility in mind that Dr. Haven Emerson investigated experimentally the relation of circulatory disturbances to chronic nephritis. He recognized that, while a variety of causes are known to be responsible for, or contribute to, chronie interstitial changes in various tissues, there is almost constantly associated with them a circulatory disturbanee, usually a venous congestion. It might be objected that such a disturbance is the result and not the cause of productive lesions in man, but Emerson’s experiments are nevertheless of value, in that this hypothesis was, for the first time, investigated. The influence of vasodilators and vasoconstrictors was tested by inhalation and by subcutaneous and intravenous injection. Inhalation experiments during a period of half a year caused the appearance of degenerative parenchymatous lesions with PROBLEMS OF EXPERIMENTAL NEPHRITIS~= 29 sight connective tissue changes. Though these experiments were few in number, the results, due apparently to disturb- ances of circulation and nutrition, suggest that with this back- ground, the long-continued administration of a renal irritant in small doses might result in the fairly constant production of ehronic nephritis. In this connection Caro’s observation that nephritis occurs in eats five to eight days after extirpation of the thyroid is suggestive. In other words, the evidence at hand supports the theory that chronic nephritis should readily be produced as the result of an irritant action associated with, or causing, circulatory and nutritional disturbances. This is in accordance with our clinical and pathological knowledge of chronic nephritis in man. In accord with this view, also, is Bradford’s suggestion that the many failures to produce chronic nephritis are probably due to the fact that we have no irritant capable of causing in animals a condition analogous to acute nephritis with cedema as seen in man. This statement was made in 1904. Such a sub- stance we now possess in uranium nitrate, which, as Richter showed in 1905, causes a very definite acute tubular nephritis with the occurrence, when an excess of water is administered, of edema of the subcutaneous tissues and accumulations of fluid in the serous cavities of the body. Uranium nitrate has come into general use as one of the most satisfactory of nephritic poisons, and Dickson, during the past year, has shown that its prolonged administration causes chronic nephritis in a large percentage of the animals treated. Unfortunately, his choice of experimental animals did not allow a study of cedema. If rabbits, in which cedema is readily produced, had been used instead of guinea-pigs, and the animals placed under condi- tions favorable to the production of cedema, it is possible that his results would have been the most satisfactory yet reported. As it is, he has shown (1) that prolonged administration of uranium nitrate causes a progressive ‘‘subchronic’’ nephritis; (2) that a series of six or seven acute attacks results in exten- sive fibrotic changes, with, in some instances, granular atrophy ‘and associated polyuria; (3) that single injections not infre- 30 HARVEY SOCIETY quently cause more or less severe fibrosis with occasionally granular atrophy; and (4) fluid, in small amounts, was found in the serous cavities of a few animals. These experiments are of great importance in connection with what has been said about the influence of circulatory dis- turbaneces in the production of chronic nephritis. Uranium nitrate, in addition to its very decided action on renal epithe- lum, also causes very definite vascular disturbances. Several investigators have been forced to this conclusion, as Heineke and Meyerstein and Dickson. Recently, I have ealled attention to the necessity of assuming such an action in order to explain certain phases of the ceedema caused by this substance. Final proof of this vascular injury is furnished by Schlayer and his associates, who have shown, by physiological methods, that although uranium primarily affects the tubules, there occurs a stage of glomerular injury characterized by dilatation of the vessels and decreased permeability. This will be discussed later in connection with cedema, but these observations serve here to indicate the value of uranium in combining the toxie¢ effects apparently necessary to the production of chronic nephritis by causing not only structural changes, but cireula- tory disturbances also. Thus may be summarized briefly the methods which have been employed in producing nephritis experimentally, the char- acter of the acute lesions, and the relation of these to chronic conditions. Such a statement is necessary as a preliminary to the discussion of functional disturbances. FUNCTIONAL DISTURBANCE The study of anatomical changes in experimental lesions adds little to our knowledge obtained by the investigation of human material. By applying physiological methods on the other hand, we may correlate disturbance of function with any state of anatomical change and thus obtain information which clinical and pathological studies fail to give. The kidney lends itself, perhaps more than any other organ, to investigation by physiological methods. The very abundant PROBLEMS OF EXPERIMENTAL NEPHRITIS 31 blood-supply with its intimate relation to the function of the kidney, the close relation of function to general blood-pressure, and the infiuence of the circulation on diuresis are conditions which readily allow the application of methods, the results of which may be graphically registered. Changes in kidney volume dependent on general blood-pressure or on the influence of its own independent vasomotor system may be measured by the oncometer, and the results for the normal compared with those in animals with experimental nephritis. Likewise a simultaneous study of diuresis allows of the determination of the changes in the elimination of fluid. The injection of various substanees influencing blood-pressure or diuresis yields infor- mation concerning the reaction of the kidney to these stimuli, and by their use it is possible to differentiate between the dis- turbances due to a glomerular and to a tubular nephritis. Further information concerning disturbances of function due to tubular or to glomerular lesions, respectively, may be gained by the use of phloridzin, and by correlated studies of the protein and salt elimination. Some information, as the result of such investigations, especially in regard to diuresis, is offered by pharmacological studies, but the most comprehensive study of this kind has been made by clinicians, by Schlayer and his associates, and deals particularly with the vascular reactions in the two types of nephritis. Their work is based on the assumption that the vascular reactions of glomerular nephritis should differ from those of tubular nephritis and that this difference should be readily determined by the action of certain stimuli, the effect of which would be to cause either contraction or dilatation of the vessels. These changes, through decrease or increase of the kidney volume, would be readily recog- nized with the aid of the oncometer. It was necessary to choose stimuli the effect of which would be but transient and which wou!d cause no injurious after-results, thus allowing a series of observa- tions on the same animal within a comparatively short space of time. Furthermore, as the conditions of experiment were such that observation on the same animal before and after the development of nephritis could be made only in short-period experiments, it was necessary to demonstrate that these stimuli exerted a constant effect on normal animals. 32 HARVEY SOCIETY To test the capacity of the vessels to contract they used sensory stimulation (tobacco smoke in the nose or transient suffocation) as an example of effect through the vasomotor centre, and adrenalin as an example of the effect of peripheral contraction. Each of these methods produced a transient diminution of kidney-volume with an inerease at the same time in general blood-pressure. Caffeine and strong salt solution were used for the purpose of producing dilatation of the renal vessels. In connection with all these conditions the rela- tion of diuresis to vascular changes and the power of phloridzin to cause glycosuria were also studied. In brief, the study was one of the reaction of the renal vessels to various stimuli and the relation on the one hand to general blood- pressure and on the other to diuresis. Necessarily, much depended on the uniformity of the control experiments, and for this reason rabbits of the same breed and similar weight were chosen, and with the exception of adrenalin, all substances were injected in definite ratio to body weight, and all but phloridzin, intravenously. Sensory stimulus and adrenalin (1 drop of 1 per cent. solution in 0.5 e.c. normal salt solution) inerease blood-pressure with a corresponding fall in kidney-volume. In each instanee this effect is transient, the normal condition being resumed in a very short space of time. On the other hand, 5 per cent. salt solution (5 ¢.c. per kilo) and 5 per cent. caffeine solution (2 ¢.¢. per 1.5 kilo) cause a marked dilatation of the renal vessels with strong pulsation and immediate diuresis, the general blood-pressure remain- ing unchanged. At the end of the experiment, phloridzin was given subcutaneously; this caused a moderate diuresis with glycosuria but without increase in kidney volume or in general blood pressure. These results were always obtained with normal animals, and the degree of reaction with each stimulus was practically the same. With such observations as controls, a study was undertaken of animals with various forms and differing stages of toxic nephritis. Potassium chromate and corrosive sublimate were used for the production of tubular nephritis, and arsenic, eantharidin, and diphtheria toxin for vascular nephritis. Schlayer’s opinion concerning tubular nephritis is based on 21 experiments with chromate and 15 with corrosive sublimate animals. The reactions to the various stimuli in the early stages of nephritis so produced do not differ markedly from the normal. It was found that the animals eliminated a larger amount of urine than do normal animals, which is in accord PROBLEMS OF EXPERIMENTAL NEPHRITIS — 33 with Weber’s observations, and also that diuretics led to a still greater flow, as had also previously been demonstrated by Hellin and Spiro. The vascular reactions differed from the normal only in degree; the power of the vessels to contract after sen- sory stimulus and adrenalin was slightly increased and the power to dilate was greater also, to about the same extent. Phloridzin acted as normally, that is, caused polyuria and glycosuria. The results with corrosive sublimate were similar except that the polyuria before the administration of diuretics was not so marked. In both forms, epithelial lesions were very promi- nent, but no anatomical changes were evident in the glomeruli. In short, the early stages of a tubular nephritis with albu- minuria and cast secretion and severe anatomical changes in the tubular epithelium offer no physiological or anatomical evidence of vascular injury. Before taking up the late stages of tubular nephritis, the reactions of vascular nephritis, for the sake of sharp contrast, may be described. Cantharidin and arsenic nephritis offer the best examples of this type. Severe vascular disturbances come on very quickly. In cantharidin nephritis, the early polyuria characteristic of the chromate lesion is absent. Within four to eight hours the effect of sensory stimulus and adrenalin is much less than in the normal, and after the administration of diu- retics the power of the vessels to dilate decreases and with it diuresis. As the nephritis proceeds to severer degree, or if larger doses of the irritant be given, the power to contract after sensory stimulus and adrenalin becomes minimal and dila- tation and diuresis become slight or cease entirely. Under such circumstances phloridzin produces no diuresis and no glyco- suria. The lesions due to arsenic are similar to those of cantharidin except that the general blood-pressure falls more quickly and remains at a lower level. This is to be explained by a greater peripheral capillary injury or perhaps by more intense action on the vasomotor centre. This comparison is very instructive. A tubular nephritis 3 34 HARVEY SOCIETY with extensive epithelial destruction and a urine rich in albu- min and casts give no physiological evidence of vascular dis- turbance except a slight polyuria and a slightly heightened response to vascular stimuli. On the other hand, in a glomeru- lar nephritis with little or no evidence of anatomical injury to either tubules or glomeruli, and with comparatively slight albuminuria and cast excretion, we find that the capacity of the vessels to contract and dilate is greatly altered, and with this a corresponding inhibition of diuresis, which may go on to total insufficiency. These observations demonstrate for the first time the pos- sibility of primary injury to glomeruli and tubules, respectively, and offer a sound experimental basis for the conception of a vascular as contrasted with a tubular nephritis. But how, ask those who object to the direct application of experimental evidence to the problem of human pathology, is this to help us in explaining the majority of renal lesions in man? We admit its value from a pharmacological point of view. We admit also the possibility of primary glomerular injury and primary epithelial injury, and also that oceasionally the glomerular lesion, as in searlatinal nephritis, may remain the predominating lesion, and, on the other hand, that the acute renal lesions of certain intoxications, as cholera, eclampsia, and to a certain extent of diphtheria, may be purely epithelial lesions; but what is the bearing of this experimental evidence on those forms of nephritis in which both glomeruli and tubules are involved, and, most frequently, it would seem, the tubules first and more seriously? This question is a proper one, and while it cannot be fully met as yet, it is, I believe, answered in part by the studies which Schlayer and his associates have made of the later stages of tubular nephritis. They find that the late stages occupy a middle position between early tubular and typical vascular nephritis, and in severe forms may simulate the latter. The reaction to sensory stimulus and adrenalin remains practically normal, but the power of dilatation and diuresis, after the administration of diuretics, decreases grad- ually, and in severe late stages, that is, after two to four days, PROBLEMS OF EXPERIMENTAL NEPHRITIS = 35 dilatation is very slight or absent and diuresis does not occur. Phloridzin no longer causes glycosuria. These changes may be accompanied by slight histological alterations in the glo- meruli, but the condition is, essentially, a functional glomeru- lar disturbance following tubular injury. That this secondary glomerular involvement is a true vascular disturbance and not the result of compression of the glomeruli, due to the retention of urine in tubules blocked by casts, Schlayer and Hedinger have shown by experiments in which the ureters were ligated. Under such conditions no vascular disturbance resulted. Thus, these investigators have demonstrated not only tubular and vascular nephritis as experimental conditions, but have shown that the former may develop into the latter. The relation, however, of the late glomerular changes to the early epithelial changes cannot be explained without more complete experi- mental evidence. That the late vascular injury is due to the original poison is doubtful, but the possibility must be con- sidered, in view of the fact that in Schlayer’s experiments with diphtheria toxin, a gradually developing nephritis of the tubular type passed, after only twenty hours, into the typical vascular type. Again, it is possible that the tubular nephritis may cause the development of secondary poisons, consequent on metabolic disturbances in other organs, and capable of affecting the glomeruli. In this connection must also be considered the matter of the ‘‘give and take’’ of renal function recently emphasized by McCrae. This theory assumes the possibility of the glomeruli taking over in part at least the function of the tubules. It is possible that substances normally passing through the tubular epithelium are, when the latter is destroyed, eliminated by the glomeruli, the endothelial cells of which may be more susceptible to injury by such substances than is the tubular epithelium. These are some of the problems suggested by Schlayer’s work, which await the verdict of further experimentation by physiological methods. During the past year I have been inter- ested in certain phases of these problems, and have repeated Schlayer’s experiments, using the dog rather than the rabbit,

ae yy Sew oo oes THE MAMMALIAN RENAL TUBULE 1A7 tubules which le nearest the developing pelvis, generations of renal vesicles and renal tubules developing external to them. The most fully developed tubules are, therefore, those found deepest in the developing cortex and they attain a marked degree of development and differentiation at a time when new renal vesicles are still developing from the neogenic zone of the periphery, a point to which I shall have occasion to refer again. I may also refer briefly to the fact that it is generally believed that certain of the first-formed renal tubules degenerate after they have attained a certain degree of development. As above stated; there are usually developed in the human kidney four primary collecting ducts of the first order, while in the adult kidney somewhat over one hundred large collecting ducts ter- minate in the pelvis. It has been learned, more particularly through the careful investigations of Hauch,°® that the increase in the number of collecting ducts which terminate in the pelvis is attained, as development proceeds, through a proximal expan- sion of the terminal end of a collecting duct of a certain order, this expanded end being then taken up into the wall of the developing pelvis, the branches of the said duct then opening separately into the pelvis. It is estimated that the collecting ducts to about the fifth division are thus taken up into the pelvis. Accompanying the reduction of the collecting ducts, there is stated to take place a reduction or degeneration of certain of the first-formed renal tubules which for instance in the human kidney toward the end of the second month are relatively large, with well-developed renal corpuscles and tubules (Felix). The question is of especial interest to me since these degenerating tubules may furnish an explanation for the existence of certain irregular vascular branches to which reference will be made in discussing the vascular supply of the kidney. It has been stated that the anlage of the different parts of the renal tubule may be recognized at a relatively early stage of development and that these parts may be definitely deter- mined as soon as epithelial differentiation is observed. It has seemed to me that not only could the different parts of the 118 HARVEY SOCIETY renal tubule be thus early recognized, but that the relative posi- tion of the different parts was also determined at a relatively early stage. The parts of the S-shaped tubule from which the proximal and distal convoluted portions are developed le above the lower S-curve from which is developed the capsule of the renal corpuscle. It forms for each tubule, as develop- ment proceeds, a coil-complex situated above the respective renal corpuscle. The loop which forms the anlage of the medullary loop rests, as will be remembered, in the concavity of the lower S-curve. As the loop elongates, it passes down over the developing renal corpuscle or the tubule connected thereto, and grows toward the pelvis. The upper end of the proximal and distal arms of the medullary loop are thus from the beginning in close relation with the renal corpuscle of the respective tubule, especially the upper end of the distal arm which lies from the very beginning nearer the developing vascular pole of the renal corpuscle, a relation which is per- manently maintained and attributed by Hamburger and Peter to the fact that the upper end of the distal arm of the medul- lary loop is fixed in this position by means of branches from the efferent glomerular branch, a point which is not always confirmed in reconstructions and may be more apparent than real in maceration preparations. The upper end of the prox- imal limb of the medullary loop in the region where it leaves the coil-complex is generally in close relation with the upper end of the distal limb, sometimes lying to the inner side of it— toward the collecting duct—sometimes over it and not often to the outside of it, as would be seen from reconstructions ; there- fore also near the renal corpusecle of the respective tubule, retaining thus in later stages of development the relations borne in earlier stages. In all stages of development it may be seen that the two limbs of the medullary loop are quite parallel and take quite a direct course toward the renal pelvis or the apex of the renal pyramid. Development also shows that the medul- lary loops of the tubules first formed extend, from a time soon after their formation, to the neighborhood of the renal pelvis or to the apex of the renal pyramid, as soon as this develops. THE MAMMALIAN RENAL TUBULE 119 The medullary loops of the several generations of tubules which develop later terminate at various levels in the medulla and in general terms it may be stated higher up in the medulla for each successive generation of tubules. In tracing the development of the renal tubule by means of reconstructions, it becomes evident, if one notes the anlage of the different parts, that the proximal limb of the medullary loop les nearer the collecting tubule than does the thicker, the distal, limb. ‘This appeared to me to be the more typical relation and was so interpreted in reconstructing the diagrams of the renal tubules shown in Fig. 6. Peter,® in a recent and most painstaking and important contribution, which I shall con- sider more fully and freely draw upon in succeeding pages, has taken exception to this statement, he finding that the distal or thicker arm is nearer the collecting duct, although he states in discussing this point with reference to the human kidney (see page 211), ‘‘Doch wird diese Lagerung nicht stets beibehalten ; da die Schleifen nicht selten langsgedreht sind, so kann sich das Verhaltnis umkehren, auch konnen beide Schenkel nicht radiar, sondern tangential in der Saule des Markstrahles zum zentral befindlichen Sammelrohr legen. Fiir die Rinden- schleifen gar lasst sich tiberhaupt kein Gesetz herauslesen; sie sind bisweilen derartig ineinander verschrankt, dass sich kein Schenkel als innerer oder aiisserer bezeichnen liisst.’’ Peter courteously excuses this error in interpretation on my part on the ground that my observations were based on renal tubules from kidneys of embryos and the new-born, and that I did not examine the renal tubules of kidneys of full grown animals. One may question, however, the validity of such an argument, for the reason that in the kidneys of embryos in later stages of development and in the new-born, from which certain of my reconstructions were made, the first formed renal tubules, those situated in the deeper part of the renal cortex, have attained a degree of development and are surrounded by a connective tissue of such a state of organization that it does not seem reasonable to suppose that the relations of their parts are materially altered in further development and growth. In 120 HARVEY SOCIETY Fic. 6.—Diagram of three renal tubules and their relation to a collecting tubule. A, of a tubule, the renal corpuscle of which is situated in the lowermost portion of the cortex; B, about the middle of the cortex; C, in the outer portion of the cortex. m, renal (Malpighian) corpuscle; v, vessel porta; n, neck; pc, proximal convoluted portion; es, medullary segment; dil, descending limb; al, ascending limb of medullary loop (loop of Henle); dc, distal convoluted portion; 7, junctional tubule; c, collecting tubule. THE MAMMALIAN RENAL TUBULE 121 Peter’s text-figure LVIII, page 340, reproduced in Fig. 7, giving a scheme of the form of a mammalian renal tubule, in the long tubule to the left of the figure, the relations of the upper ends of the proximal and distal arms of the medullary loops are, it seems to me, correctly given, or, as one may inter- pret such a relation through development,—namely, the prox- imal arm nearer the collecting tube. In this same tubule, just below the renal corpuscle, the two arms of the loop are rotated so that the distal, the thicker arm, lies nearer the collecting tubule. I would not imply that Peter would have it under- stood that such a rotation takes place in the medullary loop of every renal tubule, nor that it would take place very late in development, perhaps after birth, as may be implied in the eriticism of observations based on embryological material. It should be understood that only in a general way can the course, structure, and relation of the different parts of a mammalian renal tubule be represented by way of scheme or diagram. For, as concerns form and relations of the proximal and distal convoluted portions and their relation to the renal corpuscle and its relation to the upper ends of the proximal and distal arms of the medullary loop and the relations of all these parts to the collecting tubule, each renal tubule presents slight differences and variations. And it has seemed to me that a fundamental type-form of a mammalian renal tubule, the sequence of its parts and their relation is more readily and more accurately ascertained through development than by a study of segments of renal tubules obtained by maceration and subsequent isola- tion, even though such maceration and isolation is carried out as successfully as was done by Peter, judging from his results as shown in his excellent figures. In very early stages of develop- ment of the renal tubule, at a time when the medullary loop is in anlage, the proximal arm lies nearer the collecting tube. The same is true in certain reconstructions of slightly older stages, made by myself and Stoerk. But even in such stages, one may find renal tubules in which the two arms of the medullary loop are not in radial position with reference to the collecting tubule, but in a tangential position, in which the distal, the thicker, 122 HARVEY SOCIETY Fic. 7—Scheme of course of renal tubule of mammalia, A\S, outer stripe; JS, inner stripe; AZ, outer zone; /Z, inner zone; R, cortex; black, renal corpuscle; stippled, proximal convoluted portion with medullary segment (Hauptstiick); cross-lined, inter- mediate segment (eigentliches Schaltstiick), distal convoluted portion; cross-hatching, thicker, darker part of Henle’s loop; clear, lighter, thinner part of Henle’s loop, part of - distal convoluted portion (Zwischenstiick) and collecting tubule. After Peter. THE MAMMALIAN RENAL TUBULE 123 arm of the medullary loop may be regarded as nearer the col- lecting tubule. It seems to me that Peter has selected, as concerns this point, tubules showing one extreme, in his recon- struction of a scheme showing a mammalian renal tubule, while the writer has selected the other extreme, basing his diagram of a mammalian renal tubule on what appeared to him as the type-relation as ascertained through development. Both state- ments, it would seem, should be interpreted in this light, with the added statement that there are undoubtedly numerous renal tubules in which the two arms of the medullary loop hold a tangential position with reference to the collecting tubule, with respect to which one or the other arm may be slightly nearer, or neither of which may be regarded as nearer. We may discuss somewhat more fully the different parts of what may be regarded as a fully developed mammalian renal tubule and the relations which these parts bear to each other, to the collecting tubule, and also their position in the kidney substance. In naming the different parts of the renal tubule, it has been the custom to begin with the renal corpuscle, naming the parts in sequence to the end of the collecting ducts. This is contrary to the established custom of naming the tubular parts and secretory compartments of other glands with per- sistent ducts, in which the designation and description begins with the large ducts and proceeds toward the periphery, fol- lowing the course of their development. A number of observers, among whom may be mentioned Stohr, as quoted by permission by Peter, have suggested that the parts of the renal tubule be named beginning with the large collecting duct and passing to the renal corpuscle. Bailey and Miller (‘f Text-book of Embry- ology’’) have reversed the generally accepted meaning of prox- imal and distal, since ‘‘in development it is more convenient to speak of the terminal part of a tubule as its distal part.’’ Attention has already been drawn to the fact that the secretory part of the renal tubule, which has its anlage in the renal vesicle, begins its development with the differentiation of the renal corpuscle and that its epithelial differentiation proceeds from the renal corpuscle to the collecting tubule. For this reason, 124 HARVEY SOCIETY therefore, and for the added reason that the fluid that enters the renal tubules through the renal corpuscles is subjected to alteration by way of addition thereto and resorption therefrom in its course through the renal tubule to the collecting tubule, is it more consistent to name the parts of the renal tubule in the order of their sequence from the renal corpuscle, proximal and distal having reference to this sequence. The renal tubule begins, therefore, with the renal corpuscle (Malpighian cor- pusele), which is of spherical or of round-oval form in the human kidney and presents a urinary and vascular pole, which lie about opposite, and consists of a double-walled capsule, the glomerular capsule (Bowman’s capsule), usually spoken of as the invaginated end of the renal tubule, though it is not devel- oped by invagination, and of the glomerulus. The glomerular capsule is easily traced through its development. The outer lamella, the capsule proper, consists from the time when the tubular anlage has reached the S-stage of a single layer of pavement epithelium. ‘The inner lamella consists at this stage ofa layer of cuboidal or short columnar eells, sharply con- toured with large nuclei. These cells, as development proceeds and the glomerulus differentiates and the whole renal corpuscle attains its spherical form, assume the character of pavement epithelial cells which in the fully developed renal corpuscle are often difficult to make out clearly, since the very thin and transparent cells which surround the glomerulus are indis- tinetly bounded and have nuclei which in sections often resemble the nuclei of endothelial cells. The epithelial layers of the two lamelle are continuous at the vascular pole, a fact easily ascertained through development. From Johnston’s ™* reconstructions of the glomerulus of a human kidney (a child of three months) we learn ‘‘that the afferent vessel of the elomerulus, after entering the capsule of Bowman, immediately divides into five diverging branches. which with their sub- divisions and with the efferent vessel form an almost spherical tuft of blood-vessels.’’ The tubule is attached to the renal cor- puscle by a short though often not distinct neck continuous with the outer lamella of the glomerular capsule, the epithelium THE MAMMALIAN RENAL TUBULE 125 of the tubule which becomes shorter in the neck extending for a variable though relatively’ short distance into the outer lamella. The first portion of the tubule, which I shall designate as the proximal convoluted portion with the medullary seg- ment, is for the greater part markedly convoluted. This portion, often known as the tubulus contortus of the first order, Peter deseribes as the convolute, forming with the medullary segment,—a segment which in part forms the end segment of Argutinski or the spiral tubule of Schachowa,—the main seg- ment (Hauptstuck), terms which do not seem to me to ehar- acterize this portion of the renal tubule as definitely as the term proximal convoluted portion with medullary segment. This portion of the tubule, as will be remembered, develops from an arched tubular segment, recognized soon after the S-stage and, although it increases greatly in length and becomes mark- edly convoluted, one may recognize in this convolute, both in reconstructions of tubules representing various stages of de- velopment, as also in Peter’s excellent figures of maceration preparations, the primary arch, often with two or three second- ary curvatures, each of which presents a variable number of minor curvatures. It may be emphasized that while it is pos- sible to ascertain a certain type-form for this tubular segment, especially if traced through development, each tubule presents individual variations. The main or primary arch may be low with prominent secondary curvatures or may extend for a relatively long distance toward the periphery to return again toward the renal corpuscle, a fact which was not clearly recognized by me at the time J was making reconstructions of renal tubules and was revealed rather by accident in preparations of human and other mammalian kidneys injected with celluloid with a view of obtaining corrosion preparations of the blood-vessels. Now and again, and especially in material not perfectly fresh, the corrosion mass broke through the glomerular vessels and enter- ing the renal tubule extended in this for a variable distance. After macerating and washing away the kidney substance, there would be found numerous easts of the lumen of renal tubules, each attached to a glomerulus, the form and course of 126 HARVEY SOCIETY | each tubule, as far as injected, being preserved. Such tubular casts are easily studied under the stereoscopic binocular and the course of the tubules traced, since representing the casts of only the lumen of the tubules, the convolutes are not compact. Fic. 8.—Outlines of celluloid casts of glomerular capsule and parts of proximal convo- luted portion of the renal tubules of the human kidney. In Fig. 8 are given a number of corrosion preparations obtained in this way, from an adult human kidney. The outline of the renal corpuscle with vascular and renal pole, the former shown in only a few of the tubules, the course of the tubule as it leaves THE MAMMALIAN RENAL TUBULE 127 the renal corpuscle and its further course can here be ascer- tained. Probably not the entire proximal convoluted portion was injected, certainly not the medullary segment; the general arched form, with one or several secondary curvatures and numerous minor curvatures, however, are shown. The main mass of the convolute, formed by the proximal convoluted por- tion of each tubule, lies peripheral to its renal corpuscle, that is, toward the peripheral part of the cortex. Now and then one or several small loops may lie central to it. The proximal con- voluted portion of each tubule forms a fairly compact coil. The medullary segment varies in length. It is longer for tubules situated in the peripheral parts of the cortex and extends for a short distance into the medulla as will be stated more definitely in giving the relative position of the different parts of the renal tubule. The characteristics of the epithelium lining the proximal convoluted portion and medullary segment, essentially the same throughout, are so well understood that they may be dealt with very briefly. The relative scarcity of the nuclei, the indistinct ¢ell boundaries, the rodded protoplasm of the basal portion of the cells, the striated free border of the eells and the fact that the protoplasm stains more readily in eosin, erythrosin or Congo red than do other tubular segments found in the cortex are all distinguishing features of the epithelium of this tubular segment. The medullary segment of the proximal convoluted portion is followed by that portion of the tubule which extends for a variable distance into the medulla to return again to the region of the renal corpuscle of the respective renal tubule, thus forming a long loop, gen- erally known as the loop of Henle, for which, however, I would suggest the term medullary loop, even though I am aware of the fact that not in all cases, though this is the exception, does it extend into the medulla. This tubular segment was named by Von Kolliker as Henle’s loop after its discoverer, who, however, had an erroneous conception of its course and its relations to other parts of the renal tubule. For this reason and for the sake of consistency, since the B. N. A. has dropped all personal names, though long used to designate certain 128 HARVEY SOCIETY tubular portions or parts of kidney substance, the term medul- lary loop is suggested. To characterize the parts of the loop more fully one may speak of its proximal (descending) arm or limb, its distal (ascending) arm or limb, and the erest of the loop. The qualifying adjectives, proximal and distal, as here used by me, have been adopted by Peter. It is the proximal arm of the medullary loop which contains the narrow segment with flattened epithelium as first described by Schweigger- Seidel and accepted by the majority of writers and again estab- lished by myself, in reconstructions and by Peter in maceration preparations, after Stoerk had stated the contrary, basing his conclusions also on reconstructions. The transition from the medullary segment of the proximal convoluted portion to the thin segment with flattened epithelium is rather gradual in the human kidney (Peter) and takes place in the peripheral part of the medulla. This segment of the tubule has a rela- tively small diameter, about 20 » as compared to about 60 » for the proximal convoluted portion, though a relatively large lumen. The epithelium is of a thin pavement type, with rela- tively large oval nuclei, which reach from top to bottom of the cell and may cause the cell to bulge into the lumen. Two or three nuclei may be met with in one cross-section of this seg- ment of the tubule, often assisting in distinguishing it from capillaries, which have fewer nuclei, a point which Peter also recognized. The length of the thin and transparent segment of the proximal arm of the medullary loop varies with the length of the loop. In the reconstructions at my disposal, including tubules from new-born mammals, the thin seg- ment ceased before the crest of the loop was reached, this being formed by a thicker tubule with darker epithelium, such as is characteristic of the distal arm. In the diagrams of the tubules given in Fig. 6 it was so represented. Peter states, however, and I think correctly, that the crest of the loop is not always formed by the thicker, the distal, arm of the loop. It is necessary, as he has shown, to distinguish between long and short loops, this depending on the extent to which the loop enters the medulla. The crest of the loop in the long medullary THE MAMMALIAN RENAL TUBULE 129 loop is in the thin transparent segment, which extends for a distance on to the distal arm, while for the short loop, those ending in the peripheral part of the medulla, the crest of the loop (in the human kidney) always falls to the thicker and darker, the distal, arm of the loop, the thin segment of the proximal arm being in a short loop relatively short and may be lacking entirely in a very short loop. I am glad to correct the diagrams of tubules given in Fig. 6 to this extent; the source of error is, I think, readily seen. In even the late stages recon- structed there was evidently not complete epithelial differen- tiation; the epithelium forming the crest of the loops re- constructed, and which had the appearance of the epithelium of the distal arm, is rather to be regarded as an embryonic epithelium, the region representing a growth zone. It has seemed to me that the same explanation and interpretation might be given to the mixture of the two types of epithelia,— thin and transparent and thicker and darker,—observed by Peter near the end of the medullary loops in the kidneys of children, but never observed by him in the adult. The distal arm of the medullary loop extends from the crest of the loop to the renal corpuscle of the respective tubule, its upper end often arching over the renal corpuscle in the region of its vas- cular pole. This tubular segment is lined from the region where the thin transparent epithelium of the medullary loop ceases to the region of the renal corpuscle by an epithelium which I have characterized as presenting essentially the same structure throughout. The diameter of this tubular segment is about 30» (Peter). It is lined by a short columnar epithelium in which cell boundaries are indistinct, possessing slightly gran- ular protoplasm with indistinct striation in the basal portions and staining somewhat less deeply in eosin or erythrosin than does the epithelium of the proximal convoluted portion. There is no striated free border. The nuclei are relatively large and stain readily. Peter has called attention to the fact that the distal end of the distal arm, the portion near the renal cor- puscle, presents an appearance and structure which differs somewhat from that of the lower or proximal portion of this 9 130 HARVEY SOCIETY arm. In the rabbit, the distal portion of the distal arm has a smaller diameter and much lower epithelium. In the human renal tubule, the distal portion of the distal arm of the medul- lary loop presents in macerated preparations a diameter which is slightly larger than the proximal portion of this arm, while the lumen of the distal portion is distinctly larger, this at the cost of the height of the epithelium. Peter recognizes in this distal segment of the distal arm of the medullary loop a dis- tinct tubular segment, and it is represented as a distinct tubular segment in all of the schemes of renal tubules given by him. Whether this is justified seems to me to be questionable, espe- cially as concerns the human renal tubule. He himself states as concerns this point with reference to the human renal tubule: ‘“Der Zellbelag hat sich ganz erheblich verdiinnt, und ohne dass sein Charakter auffallend wechselt, wird dadureh doch das Aussehen des distalen Schenkels nicht unbetrachtlich verandert : Das ganze Kanalchen wird einmal heller; ferner bedingt die Dunne der Wand, dass das Rohrchen von anderen benachbarten Gangen eingeengt wird und sich der Gestalt seines Querschnittes nach diesen richten muss’’ (see page 166).8 The tubular seg- | ment following the distal arm of the medullary loop, which ends, as has been stated, near the vascular pole of the renal cor- puscle, assumes a very irregular course and may be spoken of as the distal convoluted portion, although this portion is much shorter and not so distinctly convoluted as the proximal con- voluted portion. Peter uses the term ‘‘Schaltstiick’’—inter- calated portion or tubule, to designate this tubular segment, which he further divides into a relatively short segment, known as the ‘‘Zwischenstiick’’—intermediate portion, which differs structurally not at all or only slightly from the distal end of the distal arm of the medullary loop and the intercalated por- tion proper—‘‘eigentliches Schaltstiick,’’ the beginning of which is determined in preparations macerated in hydrochloric acid, by the presence of minute dust-like crystals, which give this portion a characteristic dark appearance in isolated prepa- rations when viewed by transmitted light. The term ‘‘Schalt- stiick’’—intercalated portion—does not seem to me to be well THE MAMMALIAN RENAL TUBULE 191 chosen, since it has been customary to designate a narrow tubu- lar segment following the secretory alveoli or tubules of many glands by this term, which has thus acquired a somewhat dis- tinetive meaning. For this reason, the term distal convoluted portion, which is consistent with the other terms used, would seem to me more desirable. As has been stated, the distal con- voluted portion, which I would have extend from the place where the renal tubule breaks contact with the renal corpuscle near its vascular pole to its junction with the initial collecting tubule, is not nearly as long as the proximal convoluted por- tion—only about one-fourth to one-third the length of the latter. The convolute formed by the distal convoluted portion is not nearly so complex as that formed by the proximal portion, con- sisting frequently of one or two main loops with secondary curvatures, often having the form of a zigzag (Peter). In this segment the tubule shows irregularities of contour, enlarge- ments and constrictions, and irregular bulgings and bud-like appendages, the latter not so apparent in reconstructions (see figures of reconstructions by Huber and also Figs. 37a and 38a, Pl. V of Peter’) as in maceration preparations and it is a question whether or not a certain amount of the irregularity in contour in this tubular segment as seen in preparations mace- rated in hydrochloric acid is not due to compression caused by a. contraction of the surrounding proximal convoluted tubules which have a firmer consistency. This variation in the contour of the distal convoluted portion is quite apparent in sections. The lumen is prominent and often of irregular form. The nuclei of the lining cells are relatively large and stain readily while their protoplasm does not stain as readily in eosin or erythrosin as in the cells of the proximal convoluted portion. Peter states that the faint basal striation evident in the cells of the distal arm of the medullary loop is not evident in the eells of the intercalated segment proper, ‘‘eigentliches Schalt- stiick.’’ The peculiar dust-like crystals, seen in preparations of this segment after maceration in hydrochloric acid, are not observed in sections. The loop or zigzag formed by the distal convoluted segment usually hes to the outside of the convolute 182 HARVEY SOCIETY formed by the proximal convoluted portion and, as stated by Peter, usually on the side turned toward the collecting tubule. The main loops of the distal convoluted portion are often found near and a little above the renal corpuscle of the respective tubule. Now and again a loop of the proximal convoluted portion covers a loop of the distal convoluted portion and a loop may penetrate the coil complex formed by the proximal convoluted portion, as is shown in the left upper tubule of Fig. 6, which is drawn after a reconstruction. Peter feels that such an arrangement is impossible, stating ‘‘Auch ist eine derartige Durchflechtung von Hauptstiick und Schaltstiick, wie sie das obere linke Konvolut zeigt, in welchem das letztere durch eine Schlinge des Hauptstiickes hindurchschlipft, undenkbar’’ (see page 343).* It requires only a slight rotation outward of the proximal arm of the medullary loop of the tubule under discussion to expose the distal convoluted portion, not more rotation than is permitted in the long loop to the left in Peter’s text figure LVIII. The distal convoluted portion is continued as the initial collecting tubule which ends in the collecting tubule of the medullary ray and forms the end of the secretory portion of the renal tubule. There is a gradual transition of the epithelium of the distal convoluted portion to the initial collecting tubule and it is somewhat difficult to state where the one ends and the other begins. It is also impossible to state with any degree of certainty whether the initial collecting tubule develops as an outgrowth from the collecting tubule or is differentiated with the other parts of the renal tubule from the renal vesicle, since soon after the fusion of the S-shaped renal tubule with the ampulla of the collecting tubule, all trace of place of fusion is lost. The epithelium lining the thicker portion of the medullary loop, the distal convoluted portion, and the first portion of the initial collecting tubule presents many points of similarity. Through- out these parts, the cells are of a low columnar type with indis- tinct cell boundaries. It is true that the vertical diameter of the cells varies somewhat, the cells being somewhat shorter in the distal part of the distal arm and the cells of the distal THE MAMMALIAN RENAL TUBULE 133 convoluted portion present a somewhat more irregular inner border and stain less deeply in eosin. The differences in shape and structure are, however, so slight that it has not seemed to me necessary to recognize distinctive types with perhaps dis- tinctive functions. We are, it seems to me, justified in recog- nizing four distinct types of epithelium in a renal tubule, if we include the glomerular capsule as a part of the renal tubule. First, the flattened epithehum of the glomerular capsule; second, the epithelium of the proximal convoluted portion with the medullary segment; third, the pavement epithelium of the medullary loop; fourth, the epithelium of the distal arm of the loop, or from the region where the pavement epithelium ceases, the distal convoluted portion to the first part of the initial collecting tubule. Concerning the first three types, there can be no question. Slight variations in shape of cells and in their structure are met with in the different portions of the tubular segments, the epithelium of which is grouped under type four, variations which are more marked in certain forms than in others. In the rabbit, for instance, the distal segment of the medullary loop presents a relatively small diameter and the epithelium of the distal convoluted portion possesses an epithelium which resembles that of the proximal convoluted portion (Peter). It would seem to me, however, that the differ- ences in the structure of the epithelium of the tubular segments grouped as having the epithelium of type four are not suffi- ciently marked to warrant the recognition of further types. It is also of some importance to note that these four types of epithelia of a renal tubule are met with not only in a mam- malian renal tubule, but also in the renal tubules of certain of the other classes of vertebrates. I cannot at present speak of the renal tubules of the bird’s kidney; a reconstruction of this type is under contemplation. In the renal tubule of the rep- tilian kidney, four parts with characteristic epithelia are recog- nized. In Fig. 9 is shown a reconstruction of the renal tubule of the kidney of a turtle (Chrysemys marginata) in which the renal corpuscle, which is relatively small, with its renal capsule with distinctive epithelium is joined to an arched tubule which 134 HARVEY SOCIETY corresponds to the proximal convoluted portion with its char- acteristic epithelium. This is followed by a short narrow seg- ment, not in the form of a loop, but slightly convoluted, with a low epithelium which corresponds to the epithelium of the thin arm of the medullary loop of the mammalian renal tubule. The remainder of the tubule corresponds in epithelial lining to the distal arm of the medullary loop and distal convoluted portion of the mammalian renal tubule. In Fig. 10 is shown a Fic. 9.—Reconstruction of renal tubule of the kidney of a turtle. reconstruction of the renal tubule of a kidney of a frog, there- fore a mesonephric tubule in which the four distinctive regions possessing characteristic epithelium may also be observed. It would seem evident, therefore, that it is necessary, in consider- ing the function of a renal tubule, to take cognizance of the fact that the renal secretion from the time when its presence in the glomerular capsule is noted to the time when it reaches the collecting tubule meets with at least four types of epithe- THE MAMMALIAN RENAL TUBULE 135 lium, each of which presumably has a specific function to per- form. Starling states, ‘‘In all organs of the body whose func- tions have been investigated by physiologists, it has been found that a difference of function is invariably associated with a difference of structure, so that interdependence of function and Fic. 10.—Reconstruction of the renal tubule of a frog’s kidney. structure has become an axiom. We are therefore justified in founding theories concerning the physiologic function of an organ on a purely anatomical study of its structure, although the complete establishment of such theories must ultimately be afforded by physiological investigations.’’ Before leaving the 136 HARVEY SOCIETY consideration of the form and structure of the mammalian renal tubule, a word may be said concerning the collecting tubules. Their function is very probably merely that of conveying the fluid received from the renal tubules to the pelvis of the kidney. As was stated, the collecting tubules are developed through budding and dichotomous division of the primary collecting tubules, which bud from the primary renal pelvis. The central division of the collecting ducts, or, to state it in other form, the central union of the collecting ducts, takes place in about the inner half of the medulla. In the outer part of the medulla, the collecting ducts show no or very little division. In the cortex, the collecting tubules receive the initial collecting tubules, through which the renal tubules are united to the definite collecting tubules, the details of which vary in different forms and need not be discussed here, but are discussed very fully in Peter’s excellent account. The collecting tubules throughout are lined by a short columnar, sharply contoured epithelium not to be confused with the epithelium of any por- tion of the renal tubule. A knowledge of the relative position of the different parts of the renal tubule in the kidney substance is essential to a correct interpretation of sections of this organ and Peter’s recent contribution has very materially increased our knowledge in this regard, especially as concerns the relations of the differ- ent parts of the medullary loop. In the following presentation, I have made free use of the data which he gives. It is a well- known fact that the renal corpuscles, the proximal convoluted portions, the distal convoluted portions, the upper ends of the medullary loop, especially the distal arm and the initial collect- ing tubule, are found in the cortex and between the medullary rays, constituting what is often known as the labyrinth of the cortex. In the medullary rays are found the medullary seg- ments of the proximal convoluted portion which pass from the region of the renal corpuscle of a respective tubule to the medullary ray, further the distal end of the distal arm of the medullary loop and the cortical collecting tubules. The medulla of the kidney, whether of a simple or lobulated kidney, THE MAMMALIAN RENAL TUBULE 137 may be divided with reference to the position of the different parts of the medullary loop into distinct zones which have a definite and constant relation to the parts of the medullary loop, showing distinctive structure, as has been shown by Peter, who must be recognized as advancing very materially our knowledge of the structure of the kidney in this regard. He has shown that one may recognize in the medulla an inner and outer zone and that the latter may be further divided into an inner and outer band or stripe. He has characterized these zones and bands somewhat more fully for the kidney of a rabbit. In a sagittal section of a fresh rabbit’s kidney, the cortex presents a brown color, while the greater part of the medulla, beginning with the apex of the renal pyramid, pre- sents a grayish-white color and a somewhat transparent appear- ance. This zone is relatively broad, measuring 9.5 mm. in a kidney measuring 15 mm. from periphery of cortex to apex of renal pyramid. This constitutes the inner zone of the medulla. Peripheral to this zone there may be recognized a zone of a yellowish to a reddish color about 2.5 mm. broad and bounded externally by the cortex, forming the outer zone of the medulla. In this outer zone there may be recognized two bands or stripes of about equal width, the inner having a more yellow color, the outer a redder color, and known respectively as the inner and outer band or stripe of the outer zone. In the human kidney about the inner one-half of the medullary substance of a renal pyramid presents a grayish-white color and constitutes the inner zone, which is not sharply bounded toward the outer zone. The inner and outer bands of the outer zone are not as distinct in the human as in the rabbit’s kidney, except in the kidneys of children and young individuals. The outer band has a width of about one-fourth of the width of the outer zone. These zones may also be recognized in tissue fixed in Zenker’s, Miiller’s, and Van Gehuchten’s fluids and are especially clear in tissues macerated in hydrochloric acid. Peter has further shown that the boundary line between the inner and outer zones represents the region of transition of the narrow seg- ment of the medullary loop with pavement epithelium to the 138 HARVEY SOCIETY thicker segment with darker and thicker epithelium in the long medullary loop, those extending to the deeper parts of the medulla; this transition occurs, it will be remembered, in the distal arm of the medullary loop. In the inner zone of the medulla there are found, therefore, only collecting tubules and medullary loop segments with flattened epithelium. The boundary lne between the inner and outer band of the outer zones of the medulla is found in the region of transition of the epithelium of the medullary segment of the proximal convo- luted portion to the thin flattened epithelium of the proximal arm of the medullary loop. This transition for long and for short medullary loops takes place at about the same level and marks the inner boundary of the outer band of the outer zone of the medulla. The crests of the short medullary loops are found in the outer zone and for the rabbit in the inner band of the outer zone. In the human kidney, the inner band of the outer zone contains thin proximal arms of the medullary loop, the distal, thicker arms of the medullary loops, the crests of the short medullary loops, and collecting tubules, while the outer band of the outer zone contains the medullary segments of the proximal convoluted portions, the thicker distal arms of the medullary loops and now and again a crest of a short medullary loop (Peter). The relations as here stated as existing between the different portions of the medullary loop and the medullary zones and bands pertain not only to the medullary substance of the kidneys of rabbit and man, but also to other mammals. For comparative investigations it is, however, necessary to remember the following facts which I have taken from a sum- mary given by Peter * (see page 285). Carnivora (cat and dog) possess only the long medullary loop, those extending into the deeper parts of the medullary substance; in the rabbit, the proportion between long and short loops is as 3 to 2; in the sheep, the proportion is 1 long loop to 2.3 short loops; in man, 1 long loop to about 7 short loops; in the pig, the short loops greatly predominate. The general relations of the renal tubule and the relative positions of the different parts of the medullary loop to the medullary zones and bands are shown in Fig. 7, THE MAMMALIAN RENAL TUBULE 139 giving a diagram of the course and relations of a renal tubule of a mammal (Peter’s text-figure No. LVIII). Having thus considered the course and structure of the renal tubules of mammalia, we may now consider their relation to the renal vessels and in doing this, cognizance shall be taken more particularly of the terminal vascular branches and their relation to the different parts of the renal tubules. That the vascular supply of the kidney differs in many respects from the vascular supply of other glands with persistent ducts and external secretion is well understood. In the majority of glands of this type, the arterial branches follow in the main the duct system, to terminate in capillary networks which surround the secretory compartments, the vascular branches developing in relation with the duct system and in the majority of such elands one may recognize more or less clearly defined vascular units. No such fundamental relations pertain with reference to the blood-vessels of the kidney. The course and relations of the main branches of the renal artery are well understood, thanks to the very successful corrosion preparations of Broedel and others. It is not my purpose to consider these further than to state that the renal artery, on entering the pelvis of the kidney, courses, after further division, in the peripheral part of the medulla. These main branches, having a course which in the main is parallel to the surface of the kidney or the renal lobule, therefore describe ares with convexity outward. They are known as the arcuate arteries. From the convex side of such an arcuate artery there arise at relatively short intervals branches which form acute angles with the parent stem and pass with slight inclination toward the cortex. The length of these branches varies and from their cortical sides are given off at short intervals branches which pass more directly toward the cortex, subdividing further and giving origin to numerous branches which radiate toward the periphery of the cortex and may be known as the radial branches (interlobular arteries). The arcuate arteries ultimately terminate in smaller branches which also end in radiate arteries. Small arterial twigs, which shall be designated as the afferent glomerular branches, arise 140 HARVEY SOCIETY from all the branches of the renal artery, beginning with the arcuate branches; from the latter and from the main stems relatively few, from the radiate arteries, however, and from the branches from which they arise, there are given off numerous afferent glomerular branches. As is well known, the afferent glomerular branches divide to form the capillary plexuses of the glomeruli, these in turn uniting to form the efferent glo- merular branches, which in turn again break up into capillary networks. There has been much discussion as to whether all the terminal branches of the renal artery constitute afferent elomerular branches and whether certain terminal branches may not end in eapillaries without being connected with glo- meruli, and the chief controversial question concerns the vasa recta of Henle and Donders or the arteriole rect of other authors. Several views are current in the literature pertaining to the origin of these arteriole rectze. According to one view, they arise from the efferent glomerular branches of the renal corpuscles lying nearest the medulla, a view early expressed by Bowman, also by Gerlach, Von Kolliker, and Ludwig. showed that antirennin appears in a wave-like manner in the blood of goats injected with rennin, only that here the acme was reached quite early. According to Von Dungern ® the precipitins that form in rabbits after the injection of the plasma of certain sea animals also describe a more or less typical curve, the latent period after the primary injection of the quantities he used lasting about five days. Forssman and Lundstrom ‘* obtained a typical curve for botulismus antitoxin after a single injection of toxin, as did Madsen and Walbum & for antiricin and Famulener ® for the antitoxin against the bacterial hemolysins in beef-broth eultures of vibrio naskin (vibriolysin) and Staphylococcus aureus (staphylolysin). In rabbits injected with ox blood, FORMATION AND FATE OF ANTIBODIES 153 Bulloch ?° traced a typical curve for the specific hemolysin and numerous instances, many from personal observation, might be given of the wave-like rise and fall described by the antibodies for red corpuscles that develop in response to the injection of alien blood. Numerous researches have established definitely that the antibodies, especially the agglutinins, for typhoid bacilli, colon bacilli, cholera germs, and dysentery bacilli after a single injection in the normal animal of the respective antigen in various modifications all give typical curves.1t This also holds true for the pyocyaneus bacillus,!? the glanders bacillus,'* and paratyphoid bacillus B."* This enumeration, which makes no claim to completeness, must include, however, a general statement in regard to the opsonins. One outcome of the recent investigations of this antibody by Wright *® and others is the demonstration that the course of newly-produced opsonins in the blood does not differ essentially from that of other newly-formed antibodies. THE EFFECT OF QUANTITY AND MODE OF INTRODUCTION OF ANTIGEN ON PRODUCTION OF ANTIBODIES The height and duration of the simple curve naturally vary much, depending on the kind and amount of antigen intro- duced, the place of introduction, and in large measure also on the individual animal. General statements only can be made in regard to the relation between the quantity of antigen intro- duced at one time and the range of the resulting curve. In the case of many antigens it seems that up to a certain point the larger the amount of antigen the animal tolerates without serious disturbances the greater the production of antibodies and the longer the time before the normal level is reached again, but in the case of many antigens the optimum dose, as meas- ured by the resulting amount of free antibody, may be far below the maximum quantity the animal can stand. Certainly the yield of antibodies does not appear to increase with the same ratio as the quantity of antigen is increased. As I shall 154 HARVEY SOCIETY point out, the power of the cells to take up and dispose of antigen soon reaches its limit. Now the optimum dose of anti- gen as measured in free antibody probably should not exceed this limit because the antigen in excess of this limit to an extent may remain in the body fiuids and by uniting with the anti- body as it is produced prolong the period of latency and reduce the amount of free antibody. Relatively small quantities of certain bacteria and of alien corpuscles cause a greater output of specific antibody on intra- venous than on subcutaneous injection. Friedberger and Dorner 7° found that 300,000 to 900,000 goat corpuscles injected intravenously in rabbits would raise the specific hemolytic power of rabbit serum 5 to 20 times above normal, which was considerably more than the rise obtained on subcutaneous injection. Mertens secured 20 to 150 times as much antibody on intravenous as on subcutaneous injection of the same quan- tity of cholera germs killed by heat. In other cases the intra- venous injection does not seem to be so productive of antibodies as the subcutaneous. This is true of diphtheria toxin and Simonds had greater response as measured by the opsonic index on subcutaneous than on intravenous introduction of killed streptococci in rabbits. Probably the intravenous injec- tion of relatively small quantities of antigen is the preferable method in many cases when it is attempted to study experi- mentally under comparable conditions the effect of various factors on antibody formation. In animals previously subjected to the action of a certain antigen the mechanism of antibody production may be espe- cially sensitive to that antigen and respond to proper doses more promptly and freely than is the case in the fresh animal. This may be the reason of the quick rise in opsonin noted by Wright and others to occur sometimes on injection of specific vaccine in chronic infections. When a series of successive injections of antigen are given at varying intervals in suitable animals more or less complex antibody curves are obtained which differ much from the simple eurve. It is out of question to consider in detail the many FORMATION AND FATE OF ANTIBODIES 155 schemes that have been and are used to secure a maximum con- centration of antibodies in immunization. Speaking generally, it is the rule in immunization with toxins and bacteria to begin with relatively small and harmless quantities and to reinject with increasing quantities, carefully graded in order to avoid severe reactions and prolonged depressions of antibody pro- duction, at intervals of a few days, three or four or more, over a considerable period or until the desired antibody concentra- tion is reached in the blood.* In the case of horses immunized with diphtheria antitoxin great variation exists in the power to produce antitoxin and sooner or later the power is lost to be recovered, if at all, only after intervals of complete rest. Rarely an animal is discovered in which a sort of high antitoxic equilibrium is established that continues for months without much change either on bleeding or injection of toxin.’ Expe- rience has shown that usually it is most advantageous to bleed animals in the course of forced immunization eight or ten days after the last injection as the antibody concentration is likely to be high at that time. What may take place when the antigen is injected in increasing quantities every three or four days is indicated by the agglutinin curves obtained by Joérgensen?® in animals immunized in that way with typhoid bacilli. This curve appears to consist of a number of superimposed simple curves of gradually increasing height up to a maximum; the decline which eventually sets in and continues in spite of the continua- tion of the injections is marked by elevations of regularly decreasing height. Deutsch found that weekly intravenous injections of swine erysipelas in gradually increasing quanti- ties gave a series of curves each with the apex approximately 8 to 10 days after the corresponding injection, the maximum being reached in the fourth month.1® And Famulener *° by repeated injections of increasing doses of vibriolysin and *The possibility of producing a prolonged or “ cumulative ” negative phase in ease of established infection by too large and too frequent doses of vaccine, and its dangers to the patient, are em- phasized by Wright. 156 HARVEY SOCIETY staphylolysin every second, third or fourth day for about six injections produced gradually increasing but zigzag curves in which the maximum was attained about ten days after the last injection. On the other hand, the curves obtained by Klien ** in rabbits injected at intervals of about five days with progressively increasing quantities of typhoid bacilli and by Meakins*? in rabbits injected with increasing quantities of dysentery bacilli, staphylococci and streptococci at more irregu- lar intervals show a steady increase of the antibodies up to a certain point when some antibody or other would fall behind while others would continue to increase, their ultimate fate not having been traced. Whether the differences between these curves and those of Jorgensen and others are dependent on differences in technic must be left unsettled. On daily injections subcutaneously in goats of typhoid bacilli and cholera germs in constant doses Jorgensen obtained agglutinin curves with a somewhat prolonged latent period, the maximum being reached a few days later than after a single injection; the curve now maintained a high level for a number of days and then gradually declined in spite of the continua- tion of the injections. Von Dungern was not able to materially modify the course of the precipitins in the blood of rabbits injected with crab plasma by repeated injections during the latent period. I have made observations on dogs injected subcutaneously every day for several months with the same quantity of goat blood. In one set the quantity was 1 ¢.c. of a 10 per cent. suspension of goat blood per kilo of dog, in another one thou- sandth that amount, and in a third one two-thousandth. In all the content of the blood in antibodies reached a fairly high level on about the tenth day, which represents some prolonga- tion of the first and second phases of the simple curves obtained from a single injection of any of the doses given. And in the case of the smaller quantities the concentration reached was much greater than obtainable with a single dose, at the same time as it was maintained at a fairly constant high level for months. With the larger doses, however, the concentration at ‘pooyq 7808 YIM payelul sBop Ut saposndsoo yBod 10J UISA] JO UOIWBULLOJ OY} Aq po}BAYSN][! SV SUO{pUOD JUEJEY Tp Jepun uononpoad 4 poqyuuy FORMATION AND FATE OF ANTIBODIES +fiuojo@uside Go fiop cubs Uo y “fop fuara fignoeuoynogne , uaipo0{U) 4ajjo flop puocsas va, *Bop (Duis0u Way pesnyeuoiy PUD flop Puosae Uo fup prAq , |} Pong ibs jp iunowp eu0e YIM *T quvug 157 158 HARVEY SOCIETY the acme was not any greater than usually obtained with a single injection of that dose; in the third phase of the curve definite rises occurred, but on the whole a steady decline seemed to be taking place. There are then various procedures that may be used to secure an accumulative and continued production of antibodies. With small doses of antigen it may be accomplished by daily injections of constant quantities, at least in certain cases. With increasing doses the indications are that in many cases the best results are obtained when the injections are made toward the end of the second phase of the simple curve, the new curve starting, so to speak, from one of the levels of the previous curve. It appears, however, that animals under the continuous influence of antigens eventually lose the power to produce anti- bodies and that in cases where different antibodies are formed at the same time this loss of power may occur earlier for some antibodies than for others. McClintock and King ** were able to produce a considera degree of immunity in animals by the oral administration of toxins at the same time as digestion was inhibited by suitable means, but as yet the exact course of the antibodies in the blood under such conditions has not been traced. THE SIGNIFICANCE OF THE ANTIBODY CURVE From the foregoing statements we are warranted in con- eluding that the antibodies which form in response to antigens in animals of different species appear to pass in and out of the blood in a similar and typical manner, at least when the con- ditions are relatively simple and uncomplicated. The close similarity in formation of various antibodies in different species indicates that the mechanisms of immunization are governed by the same law and constitute physiological entities. Taking the curve as a whole it represents the rise and fall of free anti- bodies in the blood after immunization. To explain the rise and fall of diphtheria antitoxin Salomonsen and Madsen assumed that production and destruction of antitoxie substance take place at the same time. Assuming this to be true of anti- FORMATION AND FATE OF ANTIBODIES 159 bodies in general, then we may regard the antibody curve in any part of its course as representing the balance between produc- tion and loss of antibody. On the basis of this view it is evident that during the second part of the curve the amount of anti- body that is being produced exceeds increasingly the amount that is being lost and consequently antibody accumulates in the blood and other fluids. When the acme is reached production and loss are equal, but the production soon falls so that it no longer makes up for the loss and the curve gradually sinks until a permanent level is reached. The primary fall, so often noted in the curve, was first observed (Ehrlich) in animals that were reinjected during the declining phase. It also occurs, though apparently not always, when the blood normally con- tains antibodies for the antigen injected. By Ehrlich the fall was regarded as the result of neutralization by the antigen of antibodies in the blood, but in many instances the amount of antigen introduced hardly seems enough to account in this way for the fall observed. And there is room also for the con- ception that the antigen, at least in part, promptly is bound by cells whose power to produce antibodies for a period may be hindered soon to be resumed at increased rate. In the mean- time the temporary depression of function in conjunction with possible neutralization by antigen of antibodies in the blood registers itself in the primary fall. THE RELATION OF NEWLY-FORMED TO NORMAL ANTIBODIES AND THE EFFECT OF THE PRESENCE OF PREFORMED ANTIBODIES ON THE ACTIVE RESPONSE TO ANTIGEN The blood of various animals, including man, normally contains small amounts of a large number of antibodies. As in the case of many animals, the blood of normal human beings contains lysins, agglutinins, and opsonins for many different bacteria and red corpuscles. Other antibodies are also present, such as antitoxins. Small amounts of diphtheric antitoxin are present in the blood of new-born infants. These so-called normal antibodies which appear to arise in the course of normal 160 HARVEY SOCIETY metabolism usually maintain a fairly constant balance between production and loss in health. In many instances they have been found to possess specific affinities for antigenic substances and there are grounds for the view that they are not essentially different either in structure or otherwise from the correspond- ing bodies that are produced on immunization. One impor- tant reason in favor of this view, it seems to me, is the fact that the primary fall below normal may occur when an antigen is introduced in a normal animal is specific, that is, affects only the preformed or normal antibodies for the particular bac- terium or corpuscle injected. Against the view that in immu- nization there is produced an increased quantity of some pre- existing antibody or antibodies may be urged the fact that instances occur in which the antibody produced by immuniza- tion appears to lack the normal analogue. Thus normal goat blood is said not to contain antivibriolysin which is produced when goats are injected with vibriolysin, and Von Dungern failed to find any precipitin in normal rabbits for the plasma of certain sea animals. It is possible that in cases of this kind minimal amounts only of the antibody occur normally. That diphtheria toxin in certain animals fails to give rise to antitoxin on intravenous injection has been aseribed (Dzierz- gowski) to neutralization of the toxin at once on its injection by antitoxin normally in the blood. (That subcutaneous injec- tion of toxin does evoke the formation of antitoxin has been accounted for by local production.) Leaving aside the question whether the explanation is correct or not, the facts with refer- ence to the effect of preformed antibodies on the response to other antigens are quite different from those assumed to be true with respect to diphtheria toxin in this explanation. The facts are that normally the blood may contain vastly more anti- body than sufficient to completely neutralize the amount of antigen which when introduced in the blood nevertheless gives rise to the formation of antibody. Thus botulismus antitoxin is formed on intravenous injection of toxin, notwithstanding | the fact that the blood of the animal normally in a few cubic centimetres contains antitoxin enough to wholly neutralize FORMATION AND FATE OF ANTIBODIES 161 in vitro the toxin injected. In view of these facts we may assume that the cells which produce antibodies in many if not most instances have a strong affinity for antigenic substances. In actively immunized animals we know that even when newly- formed antibodies are present in the blood and other fluids in relatively large amounts reintroduction of antigen may increase anew the output of antibodies. In this case there is no indica- tion that neutralization of antigen takes place to the extent that seems possible. It does look, however, as if the affinity for antigen on the part of the cells, charged with forming anti- bodies becomes augmented in the course of the active perform- Cuarrt II. emer t | BW RS ED I I (1 a co ASSSSeREEP PEELE EEE EEE HEH | DLENE SES) ee sheen INSEE NS seanecaagern a Pee ee eee CELE ee Fe lo eee hs ele ee aa a aS aaa ERS SRR PERERA __. . JUEOR RES SeE SERRE Passive immunization by transfusion and subsequent injection of antigen. ance of their function. It has been found further that the intro- duction of antibody at the same time as antigen, separately or mixed with it, may give rise to antibody production. Special methods of immunization based on this principle have been devised by Calmette and by Besredka. The injection of rabbits with mixtures of typhoid bacilli and agglutinating rabbit serum was found by Nicolle to incite the formation of agglutinins. Recently Theobald Smith ** has demonstrated that well-marked active immunity may be induced in guinea-pigs by mixtures of diphtheria toxin and antitoxin that produce no local lesions and no constitutional disturbances. In addition to the more II 162 HARVEY SOCIETY obvious explanation of such phenomena we have the one urged by Forssman in his attack on Ehrlich’s theory, namely, that the antigen proper and the substance that unites with the antibody are different substances. But this view, although it might appear to simplify matters, does not seem to me to harmonize well with some of the results obtained on active immunization of passively immunized animals. In animals injected with typhoid or cholera antiserum, Jérgensen and Madsen ** failed to obtain any evidence of active production of antibodies on subeutaneous injection of the respective bacteria in quantities that in fresh animals result in typical curves. In this case the antigen appears to become neutralized. Tallquist ** obtained analogous results on combined passive and active immunization against vibriolysin. He found, however, that if the toxin is injected intravenously in the passively immunized animals an antibody curve of small range may be obtained; furthermore, that an active curve results if the antigen is introduced intra- peritoneally or subcutaneously immediately after the antiserum is injected into the circulation ; but if the preformed antibodies have circulated in the blood for two hours or more no active curve results if antigen is injected extravascularly. It is note- worthy that in these experiments Tallquist used an alien anti- serum for the passive immunization. We may conclude that under certain conditions of passive immunity the circulating antibodies do hinder union of antigen with antibody-producing cells. This fact would speak against the view advanced by Forssman. That antibodies are formed in passively immunized animals when antigen is introduced directly into the blood indicates again a special affinity for antigen. When the antigen is injected subcutaneously the process of absorption may give the antibodies a better oppor- tunity to bind it firmly. THE FORMATION OF DIFFERENT OR SISTER ANTIBODIES FOR THE SAME CELL In many investigations on antibody formation after the injection of substances, some of which contain distinct anti- gens, attention was first centred on the study of some single FORMATION AND FATE OF ANTIBODIES 163 antibody, e.g., of agglutinin on the injection of typhoid bacilli or of lysin on the injection of alien red corpuscles. In such eases several kinds of antibodies are developed; typhoid bacilli and alien corpuscles in suitable animals give rise to specific lysins, specific agglutinins and specific opsonins, and probably also other specific bodies. In some eases different antibodies appear to be increased in the same proportions and to describe parallel curves. This is true of the lysin, agglutinin, and opsonin for goat corpuscles in the blood of dogs injected with goat blood. But the serum and the other fluids of dogs injected once with rat corpuscles may give marked rise in the agglutinin and opsonin for these corpuscles but not in the lysin. After a single injection of dead typhoid bacilli in man the agglutinin does not fall at the same time as the lysin and the opsonin. Lack of parallelism has been noted also in the course of immunization by repeated injections at intervals of increasing doses of antigen. At first the antibodies may all increase at the same rate, the agglutinins later falling behind, as in Klien’s experiment with typhoid bacilli,** or the lysins and opsonins may decrease while the agglutinins are still increas- ing, as in Meakins’ experiments with dysentery bacilli.2* In rabbits progressively immunized against staphylococcus and streptococcus Meakins observed a steady rise in opsonin, but only an insignificant increase in agglutinin and lysin. This asymmetry in the curves of the antibodies educed in the same animal suggests that we are dealing with distinct sub- stances, the production of which is dependent on similar yet not identical mechanisms. MULTIPLE IMMUNIZATION The influence of the simultaneous or successive introduction of distinct antigens on the production of the corresponding specific antibodies is of interest because of its bearing on certain aspects of mixed infections. Several writers *° conclude that in multiple infections and immunization the production at the same time of different agglutinins takes place without any change in the onset, intensity, and duration of the production 164 HARVEY SOCIETY of any single agglutinin which is claimed to proceed as in infee- tion or immunization with the single bacterium. On the other hand, Iversen *° and Kraus found that second- ary infections, especially pneumonia, in the course of typhoid fever depress the agglutinin curve. In rabbits infected with typhoid bacilli or with rabbit septicemia, Friedberger *! found a considerable diminution of the production of amboceptors for cholera germs. I have repeatedly observed that in dogs in the course of immunization against alien blood, the development of pneumoniec infection may suspend almost completely the production of antibodies for the alien corpuscles (see Table I). TasLe I. Tue InurBitive Errect or INTERCURRENT PNEUMONIA ON THE FORMATION OF SpecirFic Lysin In Docs INJECTED witH GoaT BLOoop. | 0.1 ¢.c. 10 per cent. goat blood per K. | 1.0 c.c. 10 per cent. goat blood per K. subcutaneously daily subcutaneously daily Days ut ae Healthy dog | Pneumonic dog | Healthy dog Pneumonic dog ua eee ee Sanne BENS a eee’, 1 24 24 | 24 24 2 24 24 | 6 6 3 48 24 | 12 12 4 } 96 24 | a4 hd 5 | ae se | 96 24 6 3072 192 Sie a 7 6144 96 | 768 48 8 ts 35 1536 96 The figures give the dilution of the heated serum at which lysis of goat corpuscles ceases, guinea-pig serum being used as complement. Working with dead typhoid and colon bacilli Jorgensen ** found that injections in the second phase of the curve—3rd to 9th day—of either the antigen already introduced or of some different antigen produces a relatively small development of antibodies. Analogous facts are recorded by Von Dungern in regard to precipitins. I find that the simultaneous introduction in dogs either subeutaneously or intravenously of two or three different kinds of alien corpuscles in moderate doses may result in the production of less antibody for any one corpuscle than when the same amount of that kind only is injected. Further, that in dogs injected with alien corpuscles the injection of a different corpuscle a few days later may give less antibodies for FORMATION AND FATE OF ANTIBODIES 165 either corpuscle than are usually obtained when but one injec- tion is made. This apparent inability of the dog when influenced by more than one antigen to produce any of the respective specific antibodies to the same extent as any single antibody is produced under the influence of the corresponding antigen only, might be taken to indicate that largely the same mechan- isms or groups of cells produce all antibodies. It would be different if distinct groups of cells were charged with the exclusive production of definite antibodies. Crippling of the mechanisms of antibody formation by the simultaneous or suc- cessive introduction of different antigens might be urged in explanation of the gravity of many mixed and _ secondary infections. The lessened resistance to streptococcus and other infections of the acute infectious diseases, notably smallpox and scarlet fever, and of tuberculosis may depend in large measure on the inability of the cells charged with the produc- tion of antibodies to respond freely to the stimulus of more than one antigen at a time. And the aggravation of the primary disease, e.g., tuberculosis, on the occurrence of a secondary infection may depend on depression in the manufacture of tuberculous antibodies by the- antigen of the new infection. This possibility is illustrated by the suspension in a tuber- culous person of the tuberculin reaction by an attack of measles. Here we can assume with Von Pirquet ** that the measles anti- gen suspends the production of the antibodies on which the tuberculin reaction depends. In connection with this may be cited the observation by Ransom that in a ealf the loss of tetanus antitoxin, introduced in kindred serum, was quite eradual until the 45th day when tuberculous toxins were injected. This was followed by fever and rapid fall in the tetanus toxin—a possible hint to the effect that the loss of antibodies may be hastened under such conditions. ON THE DISTRIBUTION OF ANTIBODIES IN BLOOD AND OTHER FLUIDS In his classical experiments on immunity through inherit- anee and nursing, Ehrlich ** demonstrated that in active immu- 166 HARVEY SOCIETY nizatiow certain antitoxins appear in the milk in considerable quantities, and Brieger and Ehrlich ** constructed our first antibody curve by means of data obtained by measurement of the tetanus antitoxin in the milk of immunized goats. Since then other antibodies have been found in milk and the oceur- rence established of antibodies in transudates and other fluids besides the blood, as, for instance, of destructive substances for typhoid bacilli in the thoracic lymph by Meltzer and Norris *° but syétematic study on the distribution seems not to have been made until recently.*7 In a number of instances it is found that CuHart III. Daus i |__| ls ste re soe rae aE Re aa aa Ta REREAD ase hese CECE EEC eee ERE REPRE REARA RRA BERERREEARS Bees a BESS 400 CSRS ERe [ABE oeeo BES TT TT AT Vettel [TTT Tot TFS Stee eta SB. @2R 5 Re BE 4An wane Bf el BP4" 2 RRReeses” ADDRESS Res SeeSeolo Pan Poo ant a = Pe ‘Dy DS ee Bebe FEE EEE fey 7 EREBER BEER BES EEE i a ERESaa PRE Specifia agglutinin in blood and tymph of dogs injected with rat blood. normal antibodies are most concentrated in the blood, less in the thoracic lymph, and still less in the neck lymph, while traces only occur in the cerebrospinal and pericardial fluids and aqueous humor; and at the height of the immunity curve this relative concentration remains practically the same, at least so far as blood and lymph are concerned. For the purpose of closer study of the distribution of anti- bodies in the course of immunization Dr. Carlson and myself selected dogs which we injected intravenously with the blood of FORMATION AND FATE OF ANTIBODIES 167 white rats and of goats. Only one injection was made, the amount being 1 ¢.c. of a 10 per cent. suspension per kilo of the weight of the dog. The animals so injected were killed at different intervals after the injection and the amount of anti- bodies in the blood, lymph, ete., carefully determined. In this manner estimations were secured of the antibody content of the fluids examined at various stages of immunization. The results show clearly that so far as the blood, the lymph from the thoracic duct, and that from the neck are concerned the changes in the concentration of the antibodies during the course of active immunization run practically parallel. This applies to the lysin, agglutinin and opsonin for goat corpuscles, and to the agglutinin and opsonin for rat corpuscles, no inerease demonstrable taking place in the lysin for rat cor- puseles. The concentration in the two lymphs is about the same and always somewhat lower than in the blood-serum, but in the case of lymph as well as blood, composite curves,—in which the abscisse mark the day after injection on which the dog was killed and the fiuids collected,—correspond quite accurately with the simple curve obtained by determinations at short intervals of the content in antibody of the blood of the same animal. We may say then that in dogs injected with goat and rat blood the newly formed antibodies in the lymph appear and disappear at the same time and describe the same wave-like curve as that in the blood. As regards the cerebrospinal fluid and aqueous humor of dogs injected with goat blood, there was a complete absence in these fluids, at any stage of the reaction, of agglutinin. Lysin and opsonin, however, were present in both the fluids during the period of high antibody content in the blood and lymph, but only in traces. It is noteworthy that in dogs injected with rat blood, opsonin only was demonstrable in the cerebrospinal fluid in which it gave a typical antibody curve suggesting that an easy passage exists for this antibody. (The aqueous humor was not studied in the rat dogs.) In dogs transfused with the blood of actively immunized dogs, the antibodies may be detected in the thoracic and neck 168 HARVEY SOCIETY lymph in from one-half to three hours and very soon the same relative concentration of antibodies in lymph and blood is established as in normal and in actively immunized animals. Hence it seems probable that in active immunization the dis- tribution depends on the relative antibody content in blood and lymph rather than on place of formation of antibodies and that the rate of passage into the lymph probably is in part dependent on the concentration in the blood. ANTIBODY FORMATION IN ACUTE INFECTIOUS DISEASES Specific antibodies are produced in microbic diseases in response to the action of the infecting agents which contain antigenic substances. Naturally the degree and manner of formation of antibodies in infectious diseases are subject to variation depending on the nature of the infection and other factors. While there is still room for much systematic study in this part of the physiology of infectious diseases the facts at hand appear to warrant certain statements of a more or less general character. In several acute bacterial diseases the course of specific anti- bodies for the infecting agent in the typical attack terminating promptly in recovery without complications resembles alto- gether that of the antibody curve obtained after a single injec- tion of antigen in a normal animal. This has been found to hold good for the pneumococco-opsonin in pneumonia,** for the streptococco-opsonin in erysipeias,*® for the diphtherio-opsonin in diphtheria,*® and also for the diplococcus of Laveran in mumps.** It is of interest to note that a rise in diphtheria antitoxin occurs in spontaneous recovery from diphtheria.** The streptococcus opsonin presents characteristic variations in searlet fever and we have in this fact an additional direct indication that streptococcus from the beginning plays an important réle in scarlet fever whether it be regarded as the primary cause or not. I say from the beginning because the curve presents a degree of parallelism with the phenomena of the acute stage of the attack,—showing an early fall and rising FORMATION AND FATE OF ANTIBODIES 169 as the symptoms abate,—so that streptococcus infection must be the cause of at least some of these phenomena.** In acute articular rheumatism the opsonic index for Micro- coccus rhewmaticus and Streptococcus pyogenes follows the same course.** As new joints are involved and fever develops the index falls below normal, and on improvement of joint and symptoms the index rises above normal. In certain forms of acute otitis in which pseudodiphtheria bacilli are dominant the CuHart IV. [Ete = ine NEE SC STINE ES SIV ER UO Li ne AR ia eT Sry a [__ 2s ats soa arse eee ap RE RES TTT eA | eA cn HH BESEoe ae pes CECE EEE RESSEae Seer ee testestestetestoresstaitets HH HEBaees Wirgin,*® and others “ indicate that the giving of alcohol in mildly intoxicating quantities for several days after the injection of the antigen restrains the formation of antibodies. Wirgin found that the longer after the injection before he gave the alcohol the less its depressive effect. Fried- berger and Trommsdortt’s results point to a favorable influence on antibody formation by alcohol in a single mildly toxie dose at or near the time the antigen is introduced, but Wirgin’s experiments go rather to the contrary effect. Benjamin and Sluka‘** and Liawen *® find that the produc- tion of antibodies may be unfavorably affected by X-rays. In rabbits previously treated with X-rays and injected with beef serum two to four days later there was produced little or no precipitin and the antigen disappeared slowly from the blood; but the rays had no effect if appled four days after the injec- tion or on animals whose serum was rich in precipitin (Benja- min and Sluka). Febrile processes are associated intimately with the pro- duction of antibodies and it hes near at hand to wonder whether this production proceeds in the same way at heightened and normal temperature. The influence of experimental hyper- thermia on the formation of antibodies has been studied by Rolly and Meltzer,*® Liidke,*! and others. Rolly and Meltzer find that typhoid agglutinins and bacteriolysins are produced more rapidly and abundantly in rabbits that are kept overheated than in those which are kept cool. Ltidke reports similar results; he finds stimulation of the heat centre by puncture to cause not merely an increase in the output of agglutinin but also to so modify the agglutinin that an unusually firm sort of agglutina- tion results. Torri, ** on the other hand, who also studied the effect of puncture of the thermic centre on the development of typhoid antibodies in rabbits, was unable to determine whether the hyperthermia had any effect one way or the other. Graziani ** found that of rabbits injected in the same way with filtrates of typhoid cultures but kept at different tempera- tures, viz., + 32°, + 38°, and + 2-4° C., those kept at the low 180 HARVEY SOCIETY temperature developed the most agglutinin. In another experi- ment he kept all the animals at +32° C., bathing one-half of them in water at +20° for 30 minutes morning and evening, and in this case the bathed animals produced more agglutinin. These experiments as well as those of Agazzi,** who attempts to show that arsenical substances promote the formation of typhoid agglutinins in properly immunized animals, might have given more convincing results if the agglutinin content had been measured at more frequent intervals. The experiments I have mentioned deal mostly with the earlier phases of antibody production. The course may be influenced in the later stages also. Ltdke, being immunized with typhoid and with dysentery bacilli, found that hot baths during the stage of decline were followed by a distinct rise in the agglutinins. And in rabbits in the declining phase after being injected with typhoid bacilli, Fukuhara ** found various influences to cause a temporary rise in the agglutinin and lysin, such as chilling and warming the surface of the body, the giving of a single dose of alcohol, and the introduction of cer- — tain organ extracts. And reference has been made to the observation by Madsen and Tallquist that pyrodin and pyro- gallol,—poisons which destroy red corpuscles,—cause a distinet rise in certain antibodies if given in the third phase of the immunity curve. ANTIBODIES IN PASSIVE IMMUNIZATION Loos in 1896 ** was the first to demonstrate that antitoxin enters the blood of children injected with antidiphtherie serum. EK. Mueller ** and others found that the antitoxin disappears from the blood quite early, none being demonstrable after three weeks. Since then the time of appearance, the concentration and the fate of the antibodies in the blood in passive immuniza- tion have been subjected to special study. The results as to the influence of the place of introduction on the time when the maximum concentration of antibodies in the blood in passive immunization is reached may be sum- marized to this effect: On intravenous injection the maximum FORMATION AND FATE OF ANTIBODIES 18i concentration is reached at once while on subeutaneous, intra- muscular, and intraperitoneal injection it is reached only after an interval variously stated at 24 to 48 or 72 hours.** In man, J. Henderson Smith ** found absorption of diphtheria anti- toxin from the subcutaneous tissue complete only after 2 to 3 days. In the case of certain agglutinins and of diphtheria antitoxin Levin *® determined that the maximum concentration in the blood in animals is reached in about three days after subeutaneous and intramuscular injections, but during the first 24 hours the absorption is greater from the muscles than from the subeutaneous tissues—at the end of 10 hours, 14 times greater. Levin also found that the introduction in animals of immune serum, no matter whether from the same or different species and whether by intravenous, intramuscular, or subcutaneous injec- tion, appears to be followed by an immediate and marked loss in antibodies for which he could offer no explanation. That is to say, the demonstrable content of antibody in the serum of the animal in all cases falls far short of the amount calculated on the basis that the antibodies are simply diluted in the blood. This deficit was greatest when the antiserum was introduced subcutaneously, less when introduced intramuscularly and least when introduced intravenously, but even here it amounted to 40 to 60 per cent. and in some cases more. Marked individual differences occurred. It was less in the case of kindred than of alien antiserum; and especially marked when several differ- ent antibodies were introduced at the same time. On subcu- taneous injection of antivibriolytic serum Tallquist noted that only about one-half as much antilysin appears to reach the blood as when it is injected into the blood. Dr. Carlson and I find that in dogs first thoroughly exsanguinated and then transfused with blood from dogs injected with goat blood anti- bodies almost immediately begin to pass into the thoracie and neck lymph, and that they soon reach the same proportion in these fluids relative to that in the blood as in normal and in actively immunized animals; hence it seems to me that the amount of antibody passing into the lymph, notably after intra- 182 HARVEY SOCIETY venous injection, may go a long way towards making up the deficiency between the amount injected and the amount found in the blood. Ehrlich observed that after passive immunization antiricin and antiabrin may remain in the blood for 30 to 60 days, depending for one thing on the quantity introduced. In an ass injected subcutaneously with antidiphtherie horse serum Bulloch ** was able to demonstrate minute quantities of anti- toxin even at the end of 100 days. That definite traces of anti- bodies could be detected so long after their introduction was received with considerable astonishment. It has since been learned that introduced into the blood antibodies at first are lost rapidly and then more and more slowly. Madsen ** has shown that, at least in certain eases, the loss of antibody is expressible by the same formula in both active and passive immunization, kindred serum being used in the latter case. When it was found ** that in horses injected with antitetanie - serum the tetanus antitoxin is lost at about the same rate as in active immunization, Von Behring concluded that there is no essential difference in the immunity of the blood after active and passive immunization. Famulener °* and Levin determined that after successive intravenous injections of antibodies (antivibriolysin, typhoid and colon agglutinins) at intervals of seven days or so, there was no difference in the rapidity of the loss—the antibodies disappeared at the same rate after each injection. This proved to be the case also when a mixture of antibodies was injected at the same time. In rabbits injected by Levin with serum of goat immunized with colon bacilli all agglutinin disappeared at the end of the same time,—4 to 6 days,—even if the serum (of the same antibody strength) was injected in quantities ranging from 10 to 40 ¢.e. Consequently, it may be advisable, if we wish to maintain the concentration of allen antibodies in the blood at a certain level for a longer period to give a series of relatively small doses rather than a single large dose. At this point brief reference may be made to the fact that in passive immunization antibodies are retained longer if the FORMATION AND FATE OF ANTIBODIES 183 animal is injected with antiserum obtained from its own species than if serum from alien species is used. Tizzoni and Catani, who were the first to ascribe to the origin of the serum injected importance with respect to the rate of disappearance of anti- bodies, found that rabbits injected with antidiphtheric serum from different species retained the antitoxin longest when it was introduced in rabbit serum. Knorr and Ransom observed that tetanus antitoxin is retained longer in passive immuniza- tion if the serum used is derived from the same species as the passively immunized animal. At the same time as kindred antitoxin is retained the longest Ransom found that not all alien antitoxins are lost with the same rapidity. Consequently it is not possible to lay down a general law in regard to the fate of the antibodies of one species in the fluids of another.®® Alien agglutinins and alien bacteriolysins also disappear more rap- idly than the kindred,** in some eases three times as rapidly. To account for the disappearance of antibodies from the blood of healthy animals at least three possible mechanisms have been considered, namely, elimination in the urine and other excretions, deposition in the organs, and chemical transforma- tion. The failure to find antibodies except in exceedingly minute quantities in the urine, saliva, and other secretions (Bomstein,®* Bulloch,°* Staubli °°) and in the organs of immu- nized animals (Bomstein) lends favor to the view that chemical transformation plays an essential réle in the gradual disap- pearance of antibodies from the blood. Certain antibodies, e.g., tetanus antitoxin, pass into the milk in passively immu- nized animals. The suggestion has been made that antitoxin and other antibodies may induce the formation of antibodies and thus lead to the eventual defeat of the purpose of passive immunization. The possibility of formation of anti-antibodies cannot be discussed now, but it may be noted that Kraus and his co-workers *°° failed to obtain any antibodies for diphtheria antitoxin and typhoid agglutinins, and quite recently this is confirmed so far as antitoxin and typhoid agglutinins are concerned.**! The fact that intravenous injection immediately gives a far 184 HARVEY SOCIETY greater concentration of antibody in the blood, and hence in the lymph also, than is ever attained by the gradual absorption from the subcutaneous tissue is a strong point in favor of the direct injection into the blood of diphtherie antitoxin in severe cases as urged by Behring, Madsen, and others. The principle is, of course, equally applicable to other conditions, notably tetanus. According to Berghaus,’ the curative value of anti- diphtherie serum for guinea-pigs on direct injection into the blood is 500 times greater than on subcutaneous injection and 80 to 90 times greater than on intraperitoneal injection. The one objection of consequence that might be urged against the intravenous method is the possible greater danger of anaphy- lactic shock in susceptible persons. Note.—Schreiber * has injected antidiphtheric serum intraven- ously in 20 eases. He states that general improvement follows sooner than on subeutaneous injection while the loeal process in the throat follows about the same course in both cases. In his series no danger- ous symptoms developed. In case of difficulty in entering the vein - he advises that the injections be made into the buttock, which, so far as concerns the rapidity and degree of absorption of antitoxin into the blood, is a more favorable place for injection than the subeu- taneous tissue. According to W. H. Park (Jour. Am. Med. Assoce., 1910, liv, p. 258) large doses of antitetanie serum injected intra- venously within a few hours after the onset of the symptoms of tetanus have given good results. BIBLIOGRAPHY *Brieger u. Ehrlich: Beitrige zur Kenntniss der Milech immuni- sierter Tiere, Zeitschr. f. Hyg. u. Infektionskr., 1893, xiii, p. 336. *Salomonsen and Madsen: Recherches sur la marche de )’immun- isation active contre la diphtherie, Ann. de l’Inst. Past., 1897, xi, p. 315 and 1899, xiii, p. 262. ‘Dean, G.: Problems of Diphtheria Immunity, Lond. ‘Path. Soe. Trans. 1900, Ji, psd: “Pfeiffer and Mega Die Bildungsstiitte der Cholera Schiitzstoffe, Zeitschr. f. Hyg. u. Infektionskr., 1898, xxvii, p. 272. *Morgenroth: Ueber den Antikérper des Labenzyms, Centrlbl. f. Bakt., I, 1899, xxvi, p. 349. *Von Dungern: Die Antikérper, Jena, 1902. "Forssman and Lundstrom: Sur la marche de la courbe d’antitoxine dans l’immunisation active contre le botulismus, Ann. de I’Inst. Pasteur, 1902, xvi, p. 294. FORMATION AND FATE OF ANTIBODIES 185 *Madsen and Walbum: De la Ricine et de la Antiricine, Centrlbl. f. Bakt., I, O., 1904, xxxvi, p. 242. *Famulener, L. W.: A Report of Immunization Curves from Goats Treated with Certain Hemolytic Bacterial Toxins (Vibriolysin, Staphylolysin), Centrlbl. f. Bakt., I, O., 1907, xliv, p. 58. “Bulloch, Wm.: On the Nature of Hemolysis and its Relations to Bacteriolysis, Trans. Lond. Path. Soc., 1901, li, p. 208. “Nicolle, Ch.: Suite d’expériences relatives au phénoméne de Vageglutination des microbes, Ann. de |’Inst. Pasteur, 1904, xviii. p- 209. Liidke, H.: Untersuchungen iiber die bacillare Dysentery, Centrlbl. f. Bakt., I, O., 1906, xl, p. 69. Jorgensen and Madsen: The Fate of Typhoid and Cholera Agglutinins during Active and Passive Immunisation, Festskrift, Statens Serum Institut, Copenhagen, 1902. Deutsch: Contribution a l’étude de l’origine des anticorps typhiques, Ann. de l’Inst. Pasteur, 1899, xiii, p. 689. Wright, A. E.: On the Change Effected by Antityphoid Inoculation in the Bactericidal Power of the Blood, Lancet, 1901, ii, p. 715. Levin, E. J.: Coliagglutinins and Their Course of Formation, Festskrift, Statens Serum Institut, Copenhagen, 1902. Goldberg, S. J.: Die Agglutinationsreak- tion bei Infektionen verschiedenen Grades, Centrlbl. f. Bakt., POU. xxx, p. 605. * Simonds, J. P. and Baldauf, L. K.: The Relation of the Opsonie Index to the Leucopenia and Leucocytosis Following Injections of Heated Bacterial Cultures, Jour. Infect. Dis. 1909, vi, p. 38. *Miessner: Versuche iiber den Einfluss des Malleins auf den Agglutinationswert des Blutes gesunder und rotzkranker Pferde, Arch. f. Wissensch. u. pr. Tierheilk., 1908, xxxiv, p. 539. “Schroeder, K.: Om _ Aareladningens Indflydelse paa_ Blodets Agelutinin-Holdighed, Copenhagen, 1909. * Wright, A. E.: Studies on Immunization, London, 1909. “Friedberger and Dorner: Ueber die Haemolysinbildung durch Injektion kleinster Mengen, ete., Centrlbl. f. Bakt., I, O., 1905, Xxxvill, p. 544. “Madsen, Th.: Communications de l'Institut Sérotherapeutique de l’Etat Danois, 1909, iii, p. 1. “ Jorgensen, A.: Schwankungen des Agelutinationsvermégens des Blutes im Verlaufe des Typhus abdominalis, Centrlbl. f. Bakt., TO: 1905, xxxviii, p. 475. “ Nicolle: Suite d’expériences relatives au phénoméne de l’agglutina- tion des microbes, Ann. de |’Inst. Pasteur, 1904, xviii, p. 209. *Famulener, L. W.: A Report of Immunisation Curves ete. Centrlbl. f. Bakt., I, O., 1907, xliv, p. 58. "Klien, H.: The Opsonins in Typhoid Immunity, Bull. Johns Hopkins Hosp., 1907, xviii, p. 223. 186 HARVEY SOCIETY * Meakins, J. C.: Phagocytic Immunity, Jour. Exp. Med., 1909, xi, p- 100. “ McClintock, C. T., and King, W. F.: The Oral Administration of Antitoxins, Jour. Infect. Dis., 1909, vi, p. 46. “Smith, T.: Active Immunity Produced by So-called Balanced or Neutral Mixtures of Diphtheria Toxin and Antitoxin, Jour. Exp. Med., 1909, ix, p. 241. ” Jorgensen and Madsen: The Fate of Typhoid and Cholera Agglutinins During Active and Passive Immunization, Fest- skrift, Statens Serum Institut, Copenhagen, 1902. “ Tallquist, T. W.: Untersuchungen iiber aktive und passive Immunisierung mit Vibriolysin, Zeitsch. f. Hyg. u. Infektionskr., 1908, lviil, p. 165. * Klien, H.: The Opsonins in Typhoid Immunity, Bull. Johns Hopkins Hosp., 1907, xviii, p. 223. “Meakins, J. C.: Phagoeytie Immunity, Jour. Exp. Med., 1909, xi, p. 180: “Wolf, S.: Beitriige zur Lehre der Agglutination mit besonderer Bezugnahme die Differenzierung der Coli- und Proteus-Gruppe und auf die Mischinfektionen, Centrlbl. f. Bakt., 1899, xxv, p. 311. Castellani, A.: Die Agglutination bei Gemischter Infektion ete., Zeitschr. f. Hyg. u. Infektionskr., 1902, xl, p. 1. Kentzler, J. u. Benezur, J.: Agglutination bei Mischinfektion, Centribl.. f. Bakt., I, O.,.1908, xlvu; p. 263, “Tversen, Jul.: Ueber die Schwankungen der Agglutinations- vermogen des Serums im Verlaufe des Typhus abdominalis, Zeitschr. f. Hyg. u. Infektionskr., 1905, xlix, p. 1. " Friedberger, E.: Ueber der Einfluss des Alkohols und der Misch- vaccination auf die Intensitaét der Choleraambozeptorbildung beim Kaninchen, Congrés intern. d’hygiene et de démographie, Bruxelles, 1903. * See reference No. 18. * Verlauf der tuberkulésen Allergie bei einem Falle von Masern u. Miliartuberkulose, Wien. kl. Wochenschr., 1908, xxi, p. 861. * Ehrlich, P.: Ueber Immunitat durch Vererbung und Siéiugung, Zeit- sehr. f. Hyg. u. Infektionskr., 1892, xii, p. 183. “ Brieger -u. Ehrlich: Beitriige zur Kenntniss der Milch immunis- ierter Tiere, Ibid., 1893, xiii, p. 336. * Meltzer, S. J., and Norris, C.: The Bactericidal Action of Lymph Taken from the Thoracie Duct of the Dog, Jour. Exp. Med., 1897, ii, p. 701. * Baude and Carlson: The Influence of Various Lymphagogues on the Relative Concentration of Bacterio-agglutinin in Serum and Lymph, Am. Jour. Phys., 1908, xxi, p. 221. Hughes and Carlson: The Relative Hemolytic Power of Serum and Lymph FORMATION AND FATE OF ANTIBODIES 187 under Varying Conditions of Lymph Formation, Ibid., p. 237. Becht and Greer: A Study of the Concentration of the Anti- bodies in the Body Fluids of Normal and Immune Animals, Jour. Infect. Dis., 1910, vii, p. 127. Hektoen and Carlson: On the Distribution of Antibodies and their Formation by the Blood, Jour. Infect. Dis., 1910, vu, p. 319. * Wolf, H. E.: Observations on the Opsonie Index and the Anti- pneumococcal Power of the Blood in Pneumonia, Jour. Infect. Dis., 1906, iii, p. 731. MacDonald, C. G.: Immunity im Pneumococeal Infections, Aberdeen University Studies, No. 21, 1906, p. 367. De Marchis, F.: Le modificazioni del potere opsonico del siero di sangue nella polmonite fibrinosa, Lo Sper- imentale, 1909, Ixu, p. 681. ” Tunnicliff, R.: The Opsonic Index in Erysipelas, Jour. Infect. Dis., 1898, v, p. 268. “ Tunnicliff, R.: The Opsonie Index in Diphtheria, Jour. Infect. Dis., 1908, v. p. 14. Menabuoni: Abstr., Zeitsch. f. Immu- nitdtsf. u. Exp. Therap., Ref., 1909, i. p. 52. “Herb, I. C.: Experimental Parotitis, Arch, Int. Med., 1904, iv, p. 201. “Loos: Ueber das Verhalten des Blutserum gesunden u. Diphtheriekranken Kinder zum _ Diphtherietoxin, Jahrb. f. Kinderheilk., 1896, xl, p. 360. “Tunnicliff, R.: The Streptoeocco-opsonie Index in Searlatina. Jour, Infect. Dis., 1907, iv, p. 304. Banks, A. G.: The Variations in Searlet Fever of the Opsonie Power for Strep- tococci, Jour. Path. a. Bact., 1907, xii, p. 113. “Tunnicliff, R.: The Opsonie Index in Aeute Articular Rheuma- tism, Jour. Infect. Dis., 1909, vi, p. 346. “Hamilton, A.: The Opsonie Index and Vaccine Therapy of Pseudodiphtherie Otitis, Jour. Infect. Dis., 1907, iv, p. 313. “Tunnicliff, R.: Jour. Infect. Dis., 1909, vi, p. 346. * Amako, T.: Ueber die Schwankungen der opsonischen, agglutinieren- den und bakteriolytischen Kraft des Serums im Verlaufe der Cholera, Centrlbl. f. Bakt., I, O., 1909, xlviii, p. 602. Swan- schenzow (Zeitschr. f. Immunititsf. u. Exp. Therap., Ref., 1909, i, p. 53) found the opsonin to describe a simple curve in cholera. * Jorgensen, A.: Sechwankungen des Agglutinationsvermogens des Blutes im Verlaufe des Typhus abdominalis, Centrlbl. f. Bakt., I, O., 1905, xxxviii, p. 475. “Iversen, Jul.: Ueber die Schwankungen des Agglutinationsver- mogens des Serums im Verlaufe des Typhus abdominalis, Ztschr. f. Hyg. u. Infektionskr., 1905, xlix, p. 1. See also: Courmont: Courbe agglutinante chez les typhiques, Rev. de 188 HARVEY SOCIETY med., 1900, xx, p. 317. Forster, O.: Quantitative Untersuch- ungen iiber die agglutinierende und bakterizide Wirkung des Blutserums von Typhuskranken ete. Ztschr. f. Hyg. u. Infektionskr., 1897, xxiv, p. 500. © Richardson, M. W.: Studies upon Bacteriolysis and Typhoid Immunity, Jour. Med. Research, 1901, vi, p. 187. ™ Berl. kl. Wochensch., 1904, xli, p. 213. * Aaser, P.: Om Fagocytose og Opsonin, Tidskr. -f. d. norske Lagefor., 1907, xxvu, p. 931. * Clarke, C. P. and Simonds, J. P.: A Study of Typhoid Opsonins, Jour. Infect. Dis., 1908, v, p. 1. “Bohme: Untersuchungen iiber Opsonine, Miineh. med. Woch- ensch., 1908, lv, p. 1475. See also Milhit: Spécificité des opsonins—Diagnostie opsonique dans la Fiévre Typhoide, Arch. d’Anat. path. et de Méd. exp., 1908, xx, p. 402. * Gaethgens, W.: Opsoninuntersuchungen bei Typhusbazillentragern, Deut. med. Wochenschr., 1909, xxxv, p. 1337. * Hamilton, A.: The Opsonie Index of Bacillus Carriers, Trans. Chicago Path. Soce., 1910, viii, p. 28. * Bert and Lamb: Mediteccitedn or Malta Fever with Special Reference to the Agglutinating Substances, Lancet, 1899, Sep. 9. “ Pfeiffer and Marx: Die Bildungsstiitte der cholera Scehiitzstoffe, Zeitschr. f. Hyg. u. Infektionskr., 1898, xxvii, p. 272. “ Deutsch, L.: Contribution 4 V’etude de lorigine des anticorps typhiques, Ann. de l’Inst. Pasteur, 1899, xiii, p. 689. “ Liidke (Miineh. med. Wochenschr., 1909, 56, 1469) appears to have obtained similar results. * Romer, P.: . Experimentelle Untersuchungen uber Abrin (Jequiri- tol)- Immunitit als Grundlagen einer rationellen Jequirity- Therapie, Arch. f. Ophthalm., 1901, lu, p. 72. “Von Dungern: Die Antikérper, Jena, 1905. Also Specificitat der Antikérperbildung, Koch’s Festschrift, Jena, 1903. Leber (Experimentelle Beitriige zur Kenntniss der biologischen Vorgiinge bei Tuberkulose, Zeitschr. f. Hyg. u. Infektionskr., 1908, Ixi, p. 465) by means of complement fixation obtained results indieating that in tuberculosis of the eye antituberculin appears in the humor of the infected eye long before it appears in the blood or in the humor of the normal eye. “Forssman, J.: Studien iiber Antitoxinbildung bei aktiver Immuni- sierung gegen Botulismus, Centrlbl. f. Bakt., I, O., 1905, XXXVIl, p. 463. ” Dreyer, G. and Ainley Walker, E. W.: Observations on the Pro- duction of Immune Substances, Jour. Path. and Bact., 1909, xiv, p. 28. FORMATION AND FATE OF ANTIBODIES 189 " Pettersson, A.: Studien tiber Endolysine, Centrlbl. f. Bakt., I, O., 1908, xlvi, p. 405. “Schneider: Die bakterizide und haemolytische Wirkung der tierischen Gewebsfliissigkeiten und ihre Beziehungen zu _ der Leukozyten, Arch. f. Hyg., 1909, xe, p. 40. “Sachs: Ueber die Vorgiinge im Organismus bei der Transfusion fremdartigen Blutes, Arch. f. Anat. u. Physiol., 1903, p. 44. “Roux, E. and Vaillard, L., Contribution a V’etude du Tetanus, Ann. de l’Inst. Pasteur, 1893, vii, p. 65. “ Salomonsen, C. J., and Madsen, Th.: Sur la Reproduction de la Substance Antitoxique aprés des fortes Saignées, Ann. de |’Inst. Pasteur, 1898, xii, p. 763. "Schroeder, K.: Om Aareladningens Indflydelse paa_ Blodets Agelutinin-Holdighed, Copenhagen, 1909. ® Madsen and Tallquist: Ueber die Wirkung einige Gifte auf die Antikorperbildung (Pyrodin, Pyrogallol), Zeitschr. f. Immuni- tatsf. u. Exp. Therapie, 1909, un, p. 469. ®Friedberger and Dorner: Ueber die Himolysinbildung durch Injektion Kleimster Mengen von Blutkorperchern und tiber den Hinfluss des Aderlasses auf die Intensitiit der Bildung himolyti- secher Ambozeptoren bein Kaninchen, Centrlbl. f Bakt., I, O., 1905, ceelxxxi, p. 544. “Trommsdorff, E.: Arch. f. Hyg., 1908, lviii, p. 1. ™See reference No. 31. ® Wirgin: Ueber den Einfluss des Aethylalkohols auf die Bildung von agglutinierenden Stoffen bei Kaninchen, Centralbl. f. Bakt., fe Os 1905, xxxviil,-p. 200. “Trommsdorff: Arch. f. Hyg., 1908, lviii, p. 1. Miller, P. Th.: Ueber den Einfluss Kiinstlicher Stoffwechselalterationen auf die Production der Antikorper, Arch. f. Hyg., 1904, li, p. 365. Fraenkel, C.: Ueber den Einfluss das Alkohols auf die Empfindlichkeit des Kaninchens fiir die Erzeugnisse von Bakterien, Berl. kl. Wochenschr., 1905, xli, p. 53. “Benjamin, E. and Sluka, E.: Antikérperbildung und experimen- teller Schidigung des haematopoetischen Systems dureh Ront- genstrahlen, Wien. kl. Wochenschr., 1908, xxi, p. 311. “Lawen, A.: Exp. Untersuchungen iiber das Verhalten réntgeni- sierter Tiere gegen bakterielle Infektionen mit besonderer Beriicksichtigung der Bildung specifischer Antik6rper, Mitt. a.d. Grenzgeb. d. Med. u. Chir., 1909, xix, p. 141. “Rolly and Meltzer: Experimentelle Untersuchungen iiber die Bedeutung der Hyperthermie, Deut. Arch. f. kl. Med., 1909, xciv, p. 385. “Liidke, H.: Bedeutung der Temperatursteigerung fiir die Anti- korperproduktions, Deut. Arch. f. kl. Med., 1909, xev, p. 424. 190 HARVEY SOCIETY “Torri, G. S.: La clin. med. Ital., 1909, xlvii, p. 609. “Graziani, A.: Einfluss der umgebende Temperatur u. des kalten Bades auf die Hervorbringung von agglutinierender Substanz bei den fiir Typhus immunisierten Tieren, Centrlbl. f. Bakt., I, O., 1907, xlii, p. 633. “ Agazzi: Ueber den Einfluss einige Arsenpraparate auf die Bildung von Antikérpern (agg.) bei Kaninchen, Zeitschr. f. Immuni- tatsf. u. exp. Therapie, O., 1909, 1, p. 736. * Fukuhara: Experimentelle Beitrige zu Antik6rperbildung bei immunen Tieren, Arch. f. Hyg., 1908, lxv, p. 275. “Loos: Ueber das Verhalten des Blutserums gesunden und Diphtheriekranken Kinder zum _ Diphtherietoxin, Jahrb. f. Kinderkr., 1896, xlii, p. 360. * Mueller, E.: Ueber die Aufnahme von Schiitzkorpern nach Einver- leibung von Diphtherieantitoxin, Jahrb. f. Kinderkr., 1897, xliv, p. 394. * McClintock and King (The Oral Administration of Antitoxins, Jour. Infect. Dis., 1909, vi, p. 46) show that if digestion is inhibited diphtheria antitoxin is absorbed from the gastro-intes- tinal tract (guinea-pigs) almost as completely as after subcu- taneous injection. See also Romer and Sames: Beitr. z. antitoxischen Immunisierung auf intestinalem Wege, Zeitschr. f. Immunititsf. u. Exp. Therapie, 1909, iv, p. 270. * Smith, J. Henderson: On the Absorption of Antitoxin from the Subeutaneous Tissues and Peritoneal Cavity, Jour. Hyg., 1907, vii, p. 205. * Levin, E. J.: Ueber passive Immunitit, Zeitschr. f. Immunitatsf. u. exp. Therapie, 1909, 1, p. 3. * Bulloch, Wm.: Duration of Passive Diphtheric Immunity, Jour. Path. and Bact., 1898, v, p. 274. * Madsen, Th.: The Menene of Antibodies in the Organism isa eated by a Formula, Festskrift, Statens Serum Institut, Copen- hagen, 1902. * Ransom, F.: The Conditions which Influence the Duration of Passive Immunity, Jour. Path. and Bact., 1899, vi, p. 180. “Famulener, L. W.: A Report of Immunization Curves from Goats Treated with Certain Hemolytic Bacterial Toxins (Vibriolysin, Staphylolysin), Centrlbl. f. Bakt., I, O., 1907, xliv, p. 58. “Romer and Sames: Ueber die Haltbarkeit heterologen Anti- toxins im Organismus, Zeitschr. f. Immunititsf. u. Exp. Therapie, 1909, iv, p. 270. * Jorgensen and Madsen: The Fate of Typhoid and Cholera Agglutinins During Active and Passive Immunization, Fest- FORMATION AND FATE OF ANTIBODIES 191 skrift, Statens Serum Institut, Copenhagen, 1902. Schutze: Ueber das Verschwinden Verschiedenartiger Immunsera aus dem tierischen Organismus, Festschrift Koch, Berlin, 1903, p. 657. Levin: Loe. cit., in reference No. 90. “Bomstein: Zur Frage der Passiven Immunitaét bei Diphtherie, Centralbl. f. Bakt., 1897, xxii, p. 587. “Bulloch, Wm.: Durability of Passive Diphtheria Immunity, Jour. Path. and Bact., 1898, v, p. 274. “Staubli, C.: Exp. Untersuchungen iiber die Ausscheidung der Typhusagglutinine, Centrlbl. f. Bakt., I, O., 1903, xxxiii, p. 375. ™ Kraus u. Eisenberg: Ueber Immunisierung mit Immunsubstanzen, Centrlbl. f. Bakt., I, O., 1902, xxxi, p. 208. Kraus u. Joachim: Zur Frage der passiven Immunisierung, Wien. kl. Wochenschr. 1903, 1, p. 1389. ™ Fonteyne: Agglutinine et Anti-agglutinine, Centrlbl. f. Bakt., I, O., 1909, lu, p. 377; Anti-antitoxine, Ibid., p. 383. Berghaus, W.: Ueber die Beziehungen des Diphtherieserums zu seinem Heilwert, Centrlbl. f. Bakt., I, O., 1909, 1, p. 87. “Schreiber: Intravendse Injektion des Diphtherieserums, Miinch. med. Wochenschr., 1909, lvi, p. 1597. INFLAMMATION * EUGENE L. OPIE, M.D. The Rockefeller Institute for Medical Research HE obvious causal relationship of bacterial infection to inflammation has tended to obscure the broader signifi- cance of the inflammatory reaction. An immense number of sterile substances, both fluid and solid, soluble and insoluble, organic and inorganic, incite a reaction which differs in no essential respect from that which follows the invasion of micro- organisms. Even so-called physiological salt solution intro- duced into the body may cause acute inflammation; absorption of a protein such as egg albumen or of a fatty substance such as sterile olive-oil is in part dependent on the same process. Views concerning the nature of inflammation are widely diverse, but all are agreed that inflammation accomplishes the destruc- tion and solution of a variety of substances, and notably of those proteins which form the bodies of parasitic invaders. Although absorption from the tissue, so-called parenteral resorption, is made possible by processes which resemble those occurring within the digestive tract, recent compendiums of biochemistry are almost silent concerning the nature of such processes and limit their discussion to a consideration of the part of filtration, osmosis, and the secreting activity of lhning membranes. The pathological problems are unfamiliar to the physiological chemist, and the pathologist is poorly prepared to solve them. It is well known that there is no agreement on what shall be regarded as inflammation, and some have wished to discard the word. I shall cite historical data with the sole purpose of showing that its historical associations offer little aid in deter- * Presented before the Harvey Society, New York, Feb. 19, 1910. 192 INFLAMMATION 193 mining its application; that accepted usage furnishes no more definite criterion. The cardinal symptoms of inflammation—heat, pain, red- ness, and swelling—described by the classical writers, have reference to inflammatory conditions affecting the surfaces of the body; perhaps well illustrated by erysipelas or by a boil. By a series of analogies the term has been applied to changes in the internal organs which exhibit, in some instances, none of these symptoms. Virchow, in the ‘‘Cellular Pathology,’’ shows that each one of the cardinal symptoms at some period has been used as a test of the true nature of inflammation. The name, which implies taking fire, shows that the early writers attached greatest significance to the increased heat of the in- flamed part. At a later period, the condition of the blood- vessels indicated by congestion and redness, attracted more attention, and Boerhaave taught that inflammation was the result of stasis caused by obstruction of blood-vessels. This view prevailed during the period when, in France, pathological anatomy was studied with greatest industry. Ponfick cites the aphorism of Cruveilhier: ‘‘Phlebitis dominates pathology.”’ Yet Cruveilhier defines inflammation as a blood-stasis in the capillaries which is associated with exudation at times of coagulable lymph, at times of pus, perhaps finally of caseous or tuberculous substance. As a eriterion of inflammation, accumulation of exudate received increased attention, and the swelling or tumor of inflammation held a predominant place in the views of Rokitansky. The experimental studies of Cohnheim inaugurate modern views on the nature of inflammation. Inflammation is the reaction which follows an injury affecting the wall of blood- vessels; increased permeability facilitates the escape of plasma and corpuscles into the surrounding tissue. Attempts to study the effect of various injurious substances upon a tissue devoid of blood-vessels, such as the cornea, have shown that well-known inflammatory changes occur in the adjacent vascular tissues and hence flood the injured part with exuded fluid and corpuscles. 13 194 HARVEY SOCIETY Most of the substances which act as inflammatory irritants cause obvious injury to tissues with which they come into con- tact. At first sight it may appear unimportant to decide whether injury to tissue, including its blood-vessel, is the stimulus which puts in motion the numerous processes grouped as inflammation; or if the irritant itself acts directly on the structures with which it is in contact, and attracts to itself elements of the blood or of the tissues capable of neutralizing or destroying its toxicity. The decision will modify any inter- pretation of the phenomena of inflammation. One group of writers who have regarded injury to tissues as the inciting cause of inflammation, have included within its domain all those phenomena which tend to restore to normal the injured part; formation of fibrous tissue replacing elements which have been destroyed becomes a part of the inflammatory reaction. In- flammation is regarded as a process adapted to diminish the harmful consequences of an injury. This is the view expressed by the well-known definition of Burdon Sanderson; it repre- sents the opinion maintained by Cohnheim, Weigert, Ziegler, Neumann, Letulle, Adami. Another group of writers, includ- ing Leber, Metchnikoff, Marchand, Ribbert, Councilman, Klemensiewiez, regard inflammation as a reaction excited by the presence of something injurious to the tissues; inflammation is adapted to counteract and destroy the injurious substance. Study of the phenomena by which bacteria are destroyed and dissolved has given this view a predominant place. All inflammatory irritants produce some form of injury, and moreover, tissue which has been destroyed may act as an inflammatory irritant ; nevertheless, there is a fundamental dis- tinction between a reaction which repairs an injury, and reac- tion which renders harmless an injurious substance. Certain invertebrates with simple structure (hydra, planaria) repair an injury by rapid regeneration of a part removed; phenomena suggesting an inflammatory reaction are wholly lacking. Those who believe that inflammation is adapted to neutralize and destroy the injurious body usually exclude those regenerative changes which replace with fibrous tissue structures which have INFLAMMATION 195 been destroyed, for all writers agree in excluding the regenera- tion which affects the surviving parenchyma when part of an organ has been removed or destroyed. To determine if inflammation is dependent on changes in the blood-vessels attempts were long made to study the process in tissues such as the cornea or eartilage, which contain no vessels. The nearest vascular tissue became inflamed and the attempt failed. Directing his attention from the vertebrates which had heretofore served as objects of experiment to the lowest invertebrates, Metchnikoff has found the long-sought opportunity to study inflammation in tissues containing no blood-vessels. His well-known treatise on the eomparative pathology of inflammation defines the relatively simple reaction which follows application of injurious agents to such animals. Throughout the animal kingdom methods used to obtain food are often employed to destroy enemies. The amceba sur- vives because it can destroy and digest the bacteria which it takes into its substance. In certain sponges, phagocytic cells, which digest the food of the animal, accumulate about a foreign body thrust into its substance. The lower orders of inverte- brates, such as the medusa, the starfish, and certain worms possess no vascular system; situated between the outer covering and the digestive cavity are mesodermie cells which, having no part in the digestion of food, approach, engulf and often digest foreign particles, bacteria, and other organisms which have found their way into the tissues of the animal. By means of amceboid movement they accumulate about any substance capable of exciting their activity. Shall this reactive aceumu- lation of phagocytic cells be designated ‘‘inflammation’’? Those who believe that inflammation is a response of blood- vessels to injurious agencies are unwilling to inelude it. With a broader view, those processes by which protective elements are drawn from adjacent tissues cannot be separated from those changes by which similar cells are drawn from adjacent blood- vessels. Nomenclature of the process is relatively unimportant.’ Yet study of what is universally designated ‘‘inflammation’”’ in animals with fully developed blood-vessels shows that phago- 196 HARVEY SOCIETY eytic cells which react in response to the inflammatory irritant are not necessarily derived from the blood-vessels. To illustrate the chaotic state of prevailing views concerning inflammation, the status of tuberculosis may be cited. Cohn- heim, who attached prime importance to vascular changes, excluded the infectious granulomata; yet many of those who believe that inflammation occurs only in vascular tissue regard tuberculosis as inflammation. Marchand, on the contrary, separates such processes from inflammation, because he believes that they are characterized by multiplication of fixed cells of the tissue. Inconsistencies of accepted nomenclature are readily found. The term parenchymatous nephritis, a survival of Virchow’s conception of inflammation now long abandoned, is applied to a lesion which exhibits none of the vascular and cellular changes which are associated with inflammation of other organs. Injury to the spinal cord is designated as inflammation when it is called traumatic myelitis; yet the secondary occurrence of inflamma- tory changes common to all forms of injury merely serves to emphasize the confusion of two distinguishable conditions (Marchand?). The name ‘‘acute hemorrhagic pancreatitis” has been applied to a lesion which is essentially necrosis, and not inflammation of the pancreas, and its use has hindered a rational classification of pancreatic disease. Should we assume that inflammation occurs in order that injurious substances may be destroyed or removed, the nature and action of the fluid and cells which accumulate acquire predominant importance. The swelling of inflammation is in great part referable to accumulation of fluid derived from the plasma of the blood; yet the wall of the vessel controls this transit, for the protein content of the fluid which passes through the wall of the blood-vessel into the tissue is constantly less than that of the blood-plasma. The proteins of the plasma do not enter the spinal fiuid nor the aqueous humor, yet with inflammation they are found in both fluids. Studies of Klemensiewicz? have shown the effect of in- creased pressure exerted by exudate within the tissue on local INFLAMMATION 197 vascular tension. By an ingenious device he has been able to measure directly under the microscope the pressure capable of producing stasis within the capillaries. When an inflammatory irritant is applied to the tissue under examination, accumula- tion of exudate increases extravascular tension, and a smaller pressure is now capable of causing eapillary stasis. This observation may help to explain the obvious truth that accu- mulation of fiuid in the subcutaneous tissue in response to an irritant, is quickly self-limited ; whereas the same irritant causes an immense serous exudate when introduced into a serous cavity. Later it will be shown that this difference has an important influence on the outcome of the inflammatory reaction and may determine whether suppuration or resolution occurs. During the last ten years an immense amount of laborious study has been devoted to the character and origin of the various cells which accumulate at the site of inflammation. The studies of Cohnheim and of Von Recklinghausen have afforded convincing evidence that the common pus corpuscle is the polynuclear leucocyte of the blood which, under the stimulus of the inflammatory irritant, passes through the walls of blood-vessels. Some of the earlier observers’ have believed that such polynuclear leucocytes may become cells of the fixed tissue, colonize the part, as it were, but there is now universal agreement in the view that they may degenerate but undergo no progressive transformation after they have left the blood- stream. The origin and fate of the numerous mononuclear cells which accumulate in the inflamed tissue, on the contrary, is doubtful. The subject, repeatedly investigated by histological methods, often uninteresting because they are inconclusive, has great biological importance, for it deals with the signifi- cance of lymphatic tissue and the normal and pathological relationship between lymphatic and other tissues of the body. It seeks to determine if a cell formed in one part of the body may establish itself in a distant part and there form an integra! constituent of the tissue. Insight into the changes associated with inflammation assumes an accurate knowledge concerning the tissue in which 198 HARVEY SOCIETY the inflammatory reaction occurs. All of these changes have their origin within the connective tissues of the body whence inflammatory exudates may find their way into other situations. There are yet many defects in knowledge of the connective tissues of the body. In early stages of embryonic life this tissue is represented by a network of cells with branching processes which are continuous with one another. Within the substance of this protoplasmic syncytium, and hence within the cells, according to observations of Fleming, and in recent years of Mall,* the white fibres are laid down. At first all the cells which compose this tissue are fixed, but later cells make their appearance within the meshes of the network. Since these unattached cells exhibit irregular projections which sug- gest that they are capable of amceboid movement, and since they resemble amceboid cells of the circulating blood, they are regarded as wandering cells. Part of them have all the char- acters of lymphocytes and in many situations form small col- lections about the blood-vessels. Part of them are larger than lymphocytes and resemble the large mononuclear cells of the blood; they are frequently collected about blood-vessels. Von Recklinghausen has maintained the opinion that the spaces which, filled with fluid, exist in the meshes of the net- work formed by the fixed elements of the tissue, are in direct communication with lymphatic capillaries and constitute the origin of the lymphatics within the tissue. Nearly half a cen- tury ago (according to Sabin) Langer showed that these lymphaties grow as blind sprouts of endothelial cells. Ranvier has confirmed this almost forgotten observation in recent years, and Sabin‘ and others have shown that the entire lymphatic system sprouts from the endothelial lining of veins and grad- ually pushes its way into various tissues and organs to form a closed system everywhere lined by endothelial cells. Endo- thelium separates the lymph within the lymphatic capillaries from fixed cells of a part. This well-known relationship, usually little considered, has much pathological significance; indeed early observers (Hering, Heller, Thoma*) of the movements of amceboid cells within the tissues, have noted the important INFLAMMATION 199 truth that leucocytes which have wandered from the wall of the blood-vessels and have passed through the spaces within the fixed tissue, may penetrate the endothelial wall of a lymphatic vessel. Embryological study of the lymphatic nodes has explained the relationship of lymphatic tissue to lymphatic vessels. Gulland,* Sabin,” and others have shown that lymphoid tissue makes its appearance in the walls of lymphatic channels which have already been formed; and consequently a layer of endo- Lymphatic _—. a ery = - ) Zs » . = ae X SO eA \-Lunphocyte _---- =~ ban 4 me ; ‘Nee ~~ A Macrobhage Fic. 1.—Diagram showing the relation of the site of inflammation to the regional lym- phatic node. An artery is shown dividing to form a capillary which in turn enters a vein; within the capillary loop is a lymphatic vessel which becomes the sinus of the adjacent node and finally discharges its contents into the venous system. Within the capillary loop to the left of the dotted line is shown the normal position of wandering cells. To the right of the dotted line are shown cells having part in the inflammatory reaction. Poly- nuclear leucocytes which migrate from the blood-vessels may be ingested by macrophages or may enter the lymphatic, pass to the adjacent lymphatic node and perhaps undergo in- gestion by a macrophage within the sinus of tne node. The data on which the diagram is based are described in the text. thelial cells separates the lymphatic tissue from the lumen of the lymphatic vessel, and later from the tortuous sinus to which the primitive channel gives place. The lymphocytes of the lymph-node appear within the meshes of a fibrillated network and in their relation to lymphatics are analogous to the lympho- cytes in the meshes of connective tissue elsewhere. The local changes which with inflammation occur in the lymphatic vessels of the affected part and in the tributary lymphatic nodes (see Fig. 1) are not separable from the changes which have their seat in the blood-vessels and in the interstitial 200 HARVEY SOCIETY tissue. Muscatello® has shown that finely granular material, such as carmine powder, introduced into the peritoneal cavity of a dog, appears within ten minutes in the retrosternal lymphatic nodes; the two retrosternal lymphatic channels which follow the internal mammary arteries are quickly rendered conspicuous by the injected material. Within these lymphatic vessels some of the granules are free in the lymph, whereas others are contained in wandering phagocytic cells which, as MacCallum ® has shown, penetrate the endothelial lining of the diaphragm. Within three-quarters of an hour after injec- tion of Staphylococcus aureus into the subcutaneous tissue of the leg of a guinea-pig, Bezancon and Labbé?® found that the afferent lymphatic vessels of the adjacent lymphatic node were dilated and contained many polynuclear leucocytes which were entering the sinuses of the node. The subsequent changes within the node are well known. The well-known studies of Maximow™ have defined the changes which occur in and about a sterile foreign body, intro- duced into the subcutaneous tissue of various species of animals. In later experiments he has impregnated the body with an inflammatory irritant such as turpentine, or has infected it with pyogenic bacteria, namely, with Staphylococcus aureus and with streptococcus. He has pictured with great clearness the changes observable at intervals varying from a few minutes to many days after onset on the inflammatory reaction. The reaction caused by a sterile body differs from that produced by bacteria in its intensity and in the rapidity with which cor- responding phenomena occur, but the character and sequence of events are identical. Serous fluid quickly accumulates about the infected body and the surrounding tissues become cedematous. Within the first four hours polynuclear leucocytes emigrate from the blood- vessels in large numbers, and properly prepared tissue exhibits many leucocytes making their way through the endothelial lining of vessels. Early emigration of lymphocytes as well has so frequently been observed that its occurrence has been placed beyond doubt. The small round cells which migrate from the INFLAMMATION 201 blood-vessels quickly give place to larger cells with paler, larger nucleus and fairly abundant cell substance. Those cells which have a predominant part in the late stages of inflammation are known by no familiar name, and it is difficult to designate them conveniently. The term ‘‘macrophage,’’ used by Metchnikoff, is applicable, for these cells exhibit phagocytic activity, but the name has a wide significance and may be applied to all large cells capable of ingesting solid particles. The attack on living virulent micrococci is apparently conducted wholly by poly- nuclear leucocytes. With the disappearance of micrococci, mononuclear cells increase in number and in size and begin to exhibit ability to ingest cells and cellular débris. Such phago- eytic cells or macrophages may contain six, a dozen or more leucocytes in various stages of disintegration, together with a variety of inclusions whose origin is no longer recognizable. On the activity of these cells is in large part dependent the solution and removal of the leucocytes which have previously attacked the invading bacteria. The serous cavities, particularly the peritoneal and pleural cavities, offer a convenient opportunity for study of the cellular phenomena of inflammation. The early changes, whether pro- duced by various bacteria or by sterile irritants, do not differ materially. A noteworthy peculiarity of inflammation within serous cavities is the unobstructed and rapid accumulation of serum; the cells which accumulate are in part suspended in this fluid but a greater part adhere to the membranes, such as the omentum or mediastinum which are contiguous with the cavity. Numerous observations have shown that the changes which occur in the serous cavity during the first few hours after inoculation are identical with those which are demon- strable under similar conditions in the subcutaneous tissue. The importance of vascular changes in inflammation has long been recognized; less has been written concerning the significance of the lymphatic system. The studies which have been cited show that the lymphocytes which are in great part at least derived from the lymphatic glands migrate from the blood-vessels and are perhaps transformed into macrophages. 202 HARVEY SOCIETY At the same time lymphocytes and similar larger cells which are scattered in the normal tissue outside of the blood-vessels and often according to Ribbert form rudimentary lymphatic nodes mingle with the cells of the exudate and perhaps take part in the formation of macrophages. The intimate relation- ship of the local focus of inflammation to the adjacent lym- phatie glands is well illustrated by the experimental pleurisy produced by injection of a sterile irritant such as the vegetable protein, aleuronat, into the pleural cavity. The lymphatic glands which are situated in the anterior mediastinum become greatly swollen and microscopic examination shows that changes which occur in the sinuses of these glands are identical with those in progress within the pleural cavity itself. At the end of four or five days the serous cavity contains abundant fluid in which polynuclear leucocytes are abundant; at this time mononuclear phagocytic cells are large and numerous and are engaged in ingesting and dissolving polynuclear leucocytes. The sinuses to the adjacent mediastinal lymphatic glands are much distended and closely packed with the same large phago- cytic cells whose protoplasm often contains many polynuclear leucocytes in various stages of disintegration. In some in- stances almost the entire lymphatic gland is replaced by these cells. Ingestion of polynuclear leucocytes and other cells, essential to complete resolution of the exudate, is begun in the serous cavity and is completed in the regional lymphatic node. By the method previously described cells make their way along lymphatie channels from the primary site of inflammation to the adjacent node. Studies of the fate of bacteria injected into the body have demonstrated the rapidity with which micro-organisms enter thé regional lymphatic nodes, and the partial efficiency of these nodes as filters. Buxton and Torrey '? have injected typhoid bacilli in considerable quantity into the peritoneal cavity of small animals and have estimated by the enumeration of colonies in agar plates the relative abundance of bacteria in the sub- sternal lymphatic nodes, in the blood and in various organs such as the liver, spleen, lungs, bone-marrow, and kidney. INFLAMMATION 203 Within ten minutes after inoculation, they found an enormous number of bacteria in plates prepared from the regional lym- phatie node, and in sections prepared for microscopic examina- tion bacilli are found in the afferent sinus, in part free, in part within phagocytic cells. Notwithstanding this regional fixation of those bacteria which had escaped from the site of inocula- tion, a not inconsiderable number had entered the blood and were scattered throughout the body. Within the interval from five to thirty minutes after inoculation, from twenty to thirty thousand bacteria per cubic centimetre were recovered from the blood. Nevertheless at the end of an hour, the number had fallen to several hundred. Likewise within the first half hour after inoculation the number of bacteria in the liver, spleen, lungs, and kidney was very great; but it fell suddenly and soon became relatively small. This initial rush of bacteria from the peritoneal cavity to the blood has been found to occur with equal readiness in normal and in immunized animals. Experiments of Muscatello have shown that inanimate par- ticles such as powdered carmine pass through the diaphragm into the lymphatic vessels of the mediastinum and reach the circulating blood only through the lymphatic system. Wells and Johnstone** have successfully attempted to show that bacteria do not pass into the blood-vessels of the peritoneum but reach the blood wholly by way of the lymphatic vessels. They have prevented the initial rush of bacteria from the peri- toneal cavity into the blood by ligation of the thoracic duct. By estimation of the number of bacteria in the lymph they have shown that the thoracic duct, during the first hour after inocu- lation of the peritoneal cavity with Bacillus coli discharges an immense number of bacteria into the blood. The foregoing observations show that the lymphatic nodes, during the first hour after inoculation, are not efficient filters for bacteria. Although two lining membranes are interposed between the peritoneal cavity and the interior of lymphatie vessels, solid particles pass with the utmost rapidity from one to the other; the greater part of these particles are not con- tained within phagocytic cells. The membranes separating the 204 HARVEY SOCIETY cavity and the lumen of the vessel are uninterrupted but solid particles pass as if there were direct communication. Further- more, both bacteria and inanimate particles at first pass the lymphatic nodes, but later at the end of the first half hour or hour after inoculation, although the peritoneal cavity and the regional lymphatic nodes contain an immense number of bac- teria, their escape is obstructed and they have almost com- pletely ceased to enter the circulating blood. At this time an inflammatory reaction has begun both at the site of infection and within the lymphatic node. There is little doubt that the quiescent lymphatic node is an inefficient filter whereas the inflamed node, containing even at this early period many phago- cytic cells, is effective in restraining the dissemination of bacteria. Noetzel 1+ injected Bacillus pyocyaneus into the knee-joint of rabbits, and from five to ten minutes later found the organ- ism both in the inguinal, lumbar, and crural lymphatie nodes and in the circulating blood. Pawlowsky *® has demonstrated the presence of staphylococci in the blood and organs of guinea- pigs from twenty-four to forty-eight hours after inoculation of the knee-joint, but has been able to show that this dissemina- tion is inhibited or wholly prevented if before inoculation acute inflammation of the joint has been produced by the injection of some sterile irritant such as turpentine, alcohol or solution of quinine. His observation recalls the studies of Issayeff, who showed that the peritonitis induced by a variety of sterile irritants such as foreign blood-serum, bouillon or normal salt solution, temporarily increases resistance to subsequent intra- peritoneal inoculation of bacteria. Such observations help to explain the well-known resistance to infection exhibited by a granulating wound. A great variety of substances which are either non-dialyzable or insoluble in water are dissolved and removed when intro- duced into the tissues of an animal. It is difficult, perhaps impossible, to cite any substance which introduced from outside of the body into the tissues of an animal fails to excite an inflammatory reaction; physiological salt solution introduced INFLAMMATION 205 into the peritoneal cavity produces active emigration of leuco- eytes. Comparatively little systematic observation has been made on the pharmacology of inflammation and we are as yet ignorant of the factors on which depend peculiarities in the intensity of the reaction and in the character of the exudate which is produced. The reaction is in all instances character- ized (a) by a stage of leucocytic emigration followed when resorption begins, (b) by accumulation of macrophages. It is noteworthy that tubercle bacilli and typhoid bacilli, whose presence in man is usually associated with peculiar lesions exhibiting little resemblance to acute inflammation, produce the same changes during the first twenty-four hours after intro- duction as Staphylococcus aureus (Helly) and other pyogenic cocci. Nevertheless one large group of substances, unlike bacteria, excite the large mononuclear phagocytes with much greater activity than polynuclear leucocytes. The cells of one animal introduced into the body of another of the same or of a differ- ent species are attacked by large mononuclear cells and are gradually dissolved within their substance. This experiment has been repeated under a great variety of conditions by Metehnikoff and his pupils. The same process occurs under physiological conditions, for in the spleen red blood-corpuscles, perhaps those which have undergone some degenerative change and are no longer useful to the body, are ingested and destroyed by large mononuclear phagocytes. When hemorrhage occurs into the tissues, phagocytic cells of similar character, by taking red corpuscles into their substance, aid in the process of absorp- tion. Necrotic tissue in the liver or in other organs is absorbed by aid of the same cells. A similar process occurs when degen- erative changes affect the central nervous system. Absorption of tissues no longer useful to the body, and perhaps already the seat of degenerative change, is accomplished by the aid of mononuclear phagocytes and has many analogies throughout the animal kingdom. Metchnikoff, studying the progress of the metamorphosis of insects, has lately found evidence that the organs and tissues first undergo degenerative changes, and 206 HARVEY SOCIETY later become the prey of phagocytes. Furthermore, one large group of parasitic invaders, including protozoan micro-organ- isms such as malarial parasites and trypanosomes, excite almost exclusively the activity of the mononuclear phagocytes. The observations which have been cited show what cells accumulate about a foreign substance introduced into the body. The more important of these cells are capable of engulfing solid protein particles, and of dissolving them. By what means is this absorption accomplished ? The occurrence of products of protein digestion in inflam- matory exudates was recognized almost fifty years ago; Eich- wald in 1864 found in pus what was then called peptone; and later, Maixner found peptone in the urine in association with a considerable variety of suppurative conditions such as empyema, peritonitis, cerebrospinal meningitis, pyelitis, ete. An observation of Friedrich Miiller has explained the constant presence of so-called peptone in purulent phthisical sputum; a glycerin extract of such sputum is capable of digesting fibrin or coagulated albumin in a weakly alkaline medium. Other purulent sputum has the same property; the sputum of a patient with pneumonia does not exhibit this digestive action before crisis has occurred, but later when it has assumed a white pus-like appearance, the enzyme may be demonstrated. The pus of an abscess contains the same enzyme, but the pus- like fluid from a tuberculous lesion, a so-called cold abscess, fails to contain it. Various observers have shown that enzyme of pus is capable of digesting a considerable variety of protein substances, such as gelatin, fibrin, coagulated egg albumen, and casein. The well-known studies of Salkowski first showed that animal tissues preserved under conditions which prevent the growth of bacteria undergo changes similar to those which occur during the digestion of protein. Friedrich Miller showed that the pneumonic lung consolidated by the presence of inflam- matory exudate within the alveoli is especially susceptible to such autolysis. By the self-digestion of this inflamed pul- monary tissue at body temperature are formed albumose, leucin, tyrosin, and other products of protein disintegration; nuclei INFLAMMATION 207 of the autolyzed tissue quickly disappear as a result of decom- position of nucleins. These observations have been used to explain the solution of fibrin and the disappearance of leuco- eytes and other cellular elements which occurs with resolution of the exudate. Biondi,’® Hedin and Rowland," and others have found that various normal organs of the body autolyze with greater activity in weakly acid than in alkaline solutions, and in this respect resemble pepsin rather than trypsin. Studying the cells of an inflammatory exudate obtained by injection of aleuronat or other sterile irritant, I have repeatedly confirmed the observation that they digest coagulated protein with greatest activity when they are suspended in an alkaline medium. Digestion may be accurately measured by allowing the cells to act at body temperature on blood-serum coagulated by heat; the amount of protein which goes into solution may be accurately determined. ‘Testing the liver, kidney, spleen, lym- phatic node, and bone marrow, it is noteworthy '* that the bone- marrow alone resembles the cells of an acute inflammatory exudate, and digests with greater activity in alkali than in acid. The cell which is predominant in the inflammatory exudate produced by the injection of aleuronat is the polynuclear leu- eoeyte, and histologists are agreed that this cell has its origin in the bone-marrow. In other words, polynuclear leucocytes which, constituting the greater part of the white corpuscles of the blood, migrate during the early stage of the inflamma- tory reaction, and approach and digest solid particles, contain an enzyme which resembles trypsin of the pancreas. They carry this enzyme from the bone-marrow to the site of inflam- mation. Dochez has shown that this enzyme, unlike trypsin, exists within the e¢ells in an active state, and will, without further change, act on protein in the presence of alkali. Tryp- sin, on the contrary, exists in the pancreatic cells as zymogen, and requires activation by enterokinase or by acid before it is able to attack protein. The enzyme of the polynuclear leucocytes, which may be 208 HARVEY SOCIETY conveniently designated ‘‘leucoprotease,’’ may be purified by precipitation with alcohol, and after drying may be preserved almost indefinitely.1° In the moist state, the enzyme thus pre- pared is destroyed by heating at a temperature between 70° and 75° C. Temperatures between 50° and 65° C. acting on the en- zyme during half an hour increase its activity. It acts in an alkaline or in a neutral medium, but is inhibited by acid. Sodium carbonate in concentration of 0.2t00.5 per cent. favors its action ; greater concentration is destructive. The enzyme is much less active than trypsin, but it is not improbable that its activity, tested outside the body, is less than its activity under the fav- orable conditions which doubtless exist within the leucocyte. Examination of the properties of the enzyme which has been described, demonstrates that it is not identical, as several writers have claimed, with the alexin or complement of the blood- serum, for the latter, it is well known, is destroyed by heating to a temperature of 56° C. Jochmann?*’ has shown that it has no bactericidal power and asserts that it digests bacteria which have been killed by chloroform or by heat, whereas it fails to dissolve living bacteria. It is not difficult to bring proof ?° that the cells which accu- mulate in response to the presence of an inflammatory irritant contain a second enzyme capable of digesting albuminous sub- stances; its properties are different from those peculiar to the enzyme of the polynuclear leucocytes. The enzyme which is obtained by treating the cells with alcohol, it has been men- tioned, acts in both neutral and alkaline solutions, but is inac- tive in acid; the fresh cells, however, digest in acid as well as in alkali. This observation suggests that alcohol destroys a second enzyme, present in the fresh cells. Further study has shown that this second enzyme is more labile than leucopro- tease; for whereas temporary heating to temperatures between 50° and 65° C. increases the activity of leucoprotease, it greatly diminishes the activity of the enzyme which digests in the presence of acid. I have previously cited many observations which show that two types of cells are abundant in all inflammatory exudates INFLAMMATION 209 which exhibit a tendency to resolve. When aleuronat is injected into the pleural cavity of a dog the proportion of large mononu- clear cells, which act as phagocytes, gradually increases and with this increase there is increasing power to digest in the presence of acid. I have already pointed out that the phago- eytosis of micro-organisms, foreign particles, polynuclear leu- cocytes, red blood corpuscles, and cellular débris begun in the pleural cavity is completed in the regional lymphatic nodes. At the end of four or five days after the onset of inflammation incited by aleuronat the retrosternal lymphatic nodes are enormously enlarged beyond their normal size and their sinuses are distended with large cells identical with those in the pleural cavity and actively engaged in the phagocytosis of polynuclear leucocytes and other cellular elements. An emulsion prepared from such a lymphatic node in which mononuclear phagocytes are predominant, fails to digest protein in an alkaline or neutral medium but exhibits active proteolysis in the presence of acid. Moreover, this form of enzymotic activity increases with the duration of the changes in the node. The regional lymphatic node contains in almost pure form that enzyme which in the exudate increases with the increased number of macrophages. I have suggested for this enzyme the name ‘‘lymphoprotease.”’ This enzyme, like pepsin, acts in an acid medium and is inhibited by alkali; but it is not identical with pepsin, for it acts with greatest activity in a very weak concentration of hydrochloric acid and is destroyed by that strength (0.2 per cent.) which is favorable to the action of pepsin. It is more closely related to the autolytic enzyme of various tissues. The factor of essential importance is the inerease of this enzyme which is associated with an increase of large mononuclear phago- eytes in the exudate or with an increase of similar cells in the lvmphatie nodes tributary to the inflamed area. The enzymes which have been found in the cells of the serous inflammatory exudate just described are present as well in fibrinous exudates.2* When a small quantity of turpentine is injected into the pleural cavity, coagulable fluid accumulates and reaches a maximum at the end of two or three days. The 14 210 HARVEY SOCIETY exuded fibrin, which contains polynuclear leucocytes during the first three or four days of inflammation, undergoes solution when suspended in an alkaline medium, whereas at a later period when polynuclear leucocytes have disappeared, this property is lost. On the second or third day after onset of the inflammatory reaction, products of proteolytic digestion appear in the serum; reactions indicating the presence of albumose are readily obtained. Such decomposition products are doubtless absorbed with great rapidity, for large quantities artificially introduced disappear from the exudate within twenty-four hours. Although leucocytes contain active enzymes, serous inflam- matory exudates containing cells in abundance fail to undergo autolysis. Experiments which I made several years ago have explained the absence of such autolysis and have disclosed a mechanism by which the activity of the enzyme is limited to the locality in which it is needed. The cells of the exudate separated from the serum undergo autolysis and are capable of digesting foreign protein; but if to the cells the exuded serum is added, digestion is wholly inhibited. The serum contains some substance capable of restraining the action of the enzyme; it is convenient to designate this sub- stance ‘‘antienzyme,’’ without implying thereby that it is a specific antibody adapted to combine with enzyme in accord- ance with laws of chemical union. The antienzymotie action of the exuded serum is exhibited by the serum of the blood as well; it passes with the serum into the inflammatory exudate. The observation of E. Miiller ** that the antienzyme fails to enter the normal cerebrospinal fluid has a considerable interest. The antienzyme is destroyed by heating to 75° C. It is apparently attached to the albumin fraction of the serum for the globulin exhibits no antienzymotic action, whereas the albumin fraction is active. The antiaction occurs in an alkaline or neutral medium, but is destroyed by acid. The phenom- enon can be accurately studied by adding to weighed quantities of leucoprotease different volumes of serum. Such experi- ments do not afford evidence that enzyme and antienzyme com- INFLAMMATION 211 bine in definite quantities. Nevertheless, if to a fixed quantity of serum, increasing quantities of enzyme are added, a point is reached at which the serum fails to restrain completely the activity of the enzyme. In the study of suppuration this observation has considerable importance. Antienzymes in the blood serum similar to that which restrains the action of leucoprotease have long been known. Hahn in 1897 showed that the blood-serum inhibits the action of trypsin. It is not improbable that the inhibitory effects on trypsin and on leucoprotease are dependent upon some pecu- harity of the same substance, for Jochmann and Kantorowicz ** have found that blood-serum which has abnormally high anti- tryptic action exhibits an increased ability to restrain the action of leucoprotease. Furthermore, there is no specific relationship between the enzyme of one species and the antienzyme of the same species; the serum of the rabbit has greater antienzymotic action on dogs’ enzyme than dogs’ own serum.” Birds’ serum, unlike mammalian serum, fails to inhibit leucoprotease, which is peculiar to mammals. The relationship between leucoprotease and its antienzyme in the serum furnishes a mechanism by which the action of the enzyme is limited to the locality in which it accomplishes its function. The polynuclear leucocyte is suspended in a fluid which neutralizes the effect of its enzyme, should this enzyme be set free by disintegration of the cell or by other means. When the polynuclear leucocyte ingests a solid particle of protein matter, for example, a bacterium, it removes it from contact with the serum and brings it into contact with its enzyme. The mononuclear phagocytes are subject to a similar influ- ence, for numerous experiments have shown that the enzyme which they contain is restrained by the serum of the blood, and similarly by the serum of an inflammatory exudate. In what degree this antienzymotie action depends on the apparent alkalinity of the serum, and in what degree on a thermolabile antibody, has not been established. The relation between leucoprotease of the polynuclear leu- 212 HARVEY SOCIETY eocytes and the antienzyme of the serum has served to explain the essential nature of abscess formation. Ribbert?° defines suppuration as follows: ‘‘It is an intense inflammation with which polynuclear leucocytes wander from the blood-vessels in unusually great quantity; the tissue is softened and the serum between the collected pus cells does not coagulate.’’ It may be added that solution of tissue in some instances has a beneficial result, for softening of the least resistant tissues may result in superficial rupture with healing; without escape of pus, it is well known there is little tendency to heal. The peculiar appearance of pus is in part dependent on the presence of a great quantity of pus cells suspended in a relatively small proportion of fluid. A serous or serofibrinous exudate, on the contrary, contains abundant fluid and a rela- tively small proportion of cellular elements. Whereas the serum of the serous or serofibrinous exudate inhibits the diges- tive action of leucoprotease, the serum obtained from pus not only fails to inhibit leucoprotease, but itself contains unre- strained enzyme.*’ By disintegration of leucocytes, doubtless referable to the inflammatory irritant, increasing quantities of leucoprotease have been set free, so that the antienzymotiec activity of the exuded serum is finally overcome. The pro- teolytic enzyme may now come into contact with tissue and with fibrin, and softening is the result. The following experiment serves to explain why the same irritant in the same quantity may cause two different types of inflammation. If a small quantity of turpentine is injected into the subeutaneous tissue of a dog, a large fluctuating abscess filled with creamy pus is formed within four days; there is wide-spread undermining of the skin. The same quantity of turpentine injected into the pleural cavity causes a serofibrinous inflammation which undergoes resolution so that the pleural eavity is restored to its normal condition after about ten days; there is no destruction of tissue and a sear is not formed. In the subcutaneous tissue only a small amount of cedematous exudate can accumulate; the undiluted irritant causes active migration of leucocytes so that the antibody of the exuded INFLAMMATION 213 serum is soon overbalanced by the enzyme set free by disin- tegrated pus cells. In the pleural cavity, on the contrary, a large quantity of serum quickly accumulates and the exudate is serofibrinous instead of purulent; the antienzyme it contains is capable of holding in check the leucoprotease of the accumu- lated leucocytes. If a bit of the fibrinous exudate is suspended in the exuded serum, it is preserved intact. Nevertheless, by repeated injection of turpentine at short intervals into the pleural cavity, accumulation of leucocytes may be prolonged so that finally a condition is produced in which antienzyme can no longer restrain the enzyme. The softened fibrin of such an exudate quickly disintegrates in the serum of the exudate. The foregoing observation introduces a new factor into the discussion concerning the pyogenic activity of many bacteria. It helps to explain how the typhoid bacillus produces abscesses in certain situations such as the kidney and bone; how the pneumococcus, which rarely causes abscess of the lung, in which conditions are somewhat similar to those within the pleural cavity, may cause suppuration in other localities, such as the middle ear, or in the subdural space; how the tubercle bacillus may, under peculiar conditions, cause true suppuration. It is noteworthy that the normal spinal fiuid, unlike other body fluids, contains neither enzyme nor antienzyme, and for this reason, Dochez *7 has made a special study of the changes which occur in association with inflammation. With epidemic meningitis, antienzyme may enter the spinal fluid and quickly leaves it. With more virulent infection caused by pneumo- coeccus or streptococcus, enzyme derived from disintegrated polynuclear leucocytes gives to the fluid well marked power to digest protein. Such active enzyme itself doubtless acts as an irritant and increases the severity of the disease. A few writers, notably Marchand, exclude the infectious granulomata from the domain of inflammation; they are those who, on the one hand, accept the opinion of Baumgarten that the tubercle is formed from elements of the fixed tissue, and on the other hand, do not apply the term ‘‘inflammation’’ to regenerative changes in the fixed tissue. Nevertheless, the 214 HARVEY SOCIETY greater number of pathologists give weight to the truth that the tubercle is formed by a reaction in response to the presence of an invading parasite, and this reaction, in its early stage, is identical in character with that which follows the entrance of other bacteria into the tissues. Tuberculous tissue, moreover, is composed in large part of so-called epithelioid cells; these cells have the anatomical structure and phagocytic activity of the large mononuclear cells which predominate in the later stages of an acute inflammatory reaction. With present knowl- edge, it is impossible to define clearly the relationship of the tubercle to the later stage of inflammation, for the available evidence has permitted no agreement concerning the origin of the epithelioid cells. Study of acute inflammations produced by a sterile foreign body or by bacteria demonstrates with con- siderable certainty that lymphoid cells leave the blood-vessels and, it is probable, assume the characters of macrophages. In the immense accumulation of cells which follows, the identity of various elements is lost and only the uncertain means of trac- ing transitions from one form to another is available for deter- mining origin of various types. Large mononuclear cells are accumulating in the tuberculous and in the non-tuberculous inflammation after the first twenty-four hours. There is no doubt that small round cells with the character of lymphocytes accumulate in the neighboring blood-vessels and migrate from them during the formation of the tuberculous lesion. Though transitions from this lymphoid cell to epithelioid cells are not wanting, there is no convincing evidence that one is derived from the other. Polynuclear leucocytes occur in secant number in tubercles found at autopsy; yet in man (Benda), as in other animals, they are the first cells to accumulate about tubercle bacilli which are free in the tissues. Within an hour after injection of tuberele bacilli into the blood or into a serous cavity, they are surrounded or ingested by polynuclear leucocytes; mononu- clear cells subsequently appear. In some animals, polynuclear leucocytes are very numerous in tuberculous tissue. In the dog, during the first few weeks after inoculation of the pleural INFLAMMATION 215 cavity, polynuclear leucocytes occur in immense number in the tuberculous tissue which is formed in and on the mediastinum. The relative abundance of these cells is dependent on the char- acter of the bacillus, and in some degree is an index of the activity of resistance upon the part of the host. Virulent tubercle bacilli excite a more active emigration of polynuclear leucocytes than non-virulent organisms. If the lesions which are classed as infectious granulomata are passed in review, various conditions intermediate between the tubercle and a simple abscess are found. The actinomycotic nodule has many of the characters of the tubercle, yet poly- nuclear leucocytes are so abundant that a small abscess is formed in the immediate neighborhood of the micro-organism. Glanders, in man and in lower animals, is usually characterized by abundant accumulation of polynuclear leucocytes with necrosis and suppuration. Duval and White ** have shown that the character of the lesion produced in animals varies with the virulence of the micro-organism. Very virulent strains of the bacillus of glanders rapidly cause necrosis of tissue and forma- tion of small abscesses in the liver, lungs and other organs, whereas less virulent organisms produce nodules which are composed of epithelioid and giant cells and have all.the char- acters of tubercles. The specificity of the tubercle is impaired by the observation that various sterile foreign bodies produce somewhat similar nodular lesions. When, for example, finely powdered meal (Kopee *®) in suspension is introduced into the peritoneal cavity, the particles are collected together in clumps and tubercle-like nodules are formed about the clumps scattered upon the peritoneal surface. In other respects these foreign body tubercles do not accurately reproduce the histological peculiarities of the true tubercle. Similar foreign body tubercles have been found scattered throughout the peritoneal cavity when, under conditions which cannot be accurately defined, food particles have entered the cavity through a per- foration in the wall of the gastro-intestinal tract. It is well known that the tubercle bacillus contains an insol- 216 HARVEY SOCIETY uble wax-like substance on which, in part at least, depends its ability to resist solution in the tissues; it is not improbable that its peculiar staining properties are dependent on the same substance. Such wax may be obtained by extraction from tuberele bacilli and introduced in suspension into the body of an animal first attracts polynuclear leucocytes; later mononu- clear phagocytes accumulate, and among them occur giant cells. At the periphery a fibrous capsule is formed; the wax remains undissolved (Tschistowitsch *°). One form of pseudo-tubercle accurately reproduces the histological characters of the true tubercle. About the eggs of the blood-fluke Schistosoma japonicum deposited in the liver and in the intestinal wall nodules with all the characters of true tubercles are formed. Through the kindness of Dr. Henry J. Nichols,*t I have lately had opportunity to examine tissues from a ease of schistosomiasis occurring in the Philippine Islands. The nodules are composed of epithelioid cells contain- ing giant cells; at the periphery of the nodule lymphoid cells are abundant. Coagulation necrosis with the histological char- acters of caseation occurs in the centre of the nodules in con- tact with the egg, and the epithelioid cells at the margin of the necrotic area assume the arrangement frequently seen in true tubercles, namely, with long diameter at right angles to the margin of necrosis. The observations just described suggest that the tubercle has a close relationship, on the one hand, to the late stage of acute inflammation at a time when absorption is in progress and, on the other hand, to the changes which occur about an insoluble substance. The histological data which are available, fail to furnish conclusive evidence concerning the origin of the macrophage, which has an important part in acute inflam- mation, nor of the epithelioid cell of the tubercle. Both cells are capable of ingesting and dissolving protein bodies, and both contain enzymes with similar properties. The dog offers a favorable opportunity for study of the enzymes of tuberculous tissue and for comparison of these enzymes with those present in the sterile inflammatory exudates INFLAMMATION 217 which are readily obtainable from the same animal.*” When tubercle bacilli are injected into the pleural cavity, an immense mass of tuberculous tissue is formed in the mediastinum and the adjacent lymphatic glands undergo enormous hypertrophy. The power of this tissue to digest protein material exhibits certain noteworthy peculiarities. During the first two or three weeks after its formation polynuclear leucocytes are abundant and it exhibits the ability inherent in the leucoprotease of these cells to digest in the presence of an alkaline medium. At a later period with the disappearance of polynuclear leucocytes, this property diminishes and is finally lost. In the early period of its formation the tuberculous tissue digests in weak acid as well and at a later period when leucoprotease is no longer demonstrable the power of energetic digestion in acid persists. The enzyme which has this property may be extracted from the cells with water and preserved during a limited period of time. There is little doubt that it is contained in the epithelioid cells which digest within their substance tubercle bacilli, polynuclear leucocytes, red blood-corpuscles and other cellular elements ; for such cells constitute almost the entire bulk of the newly formed tuberculous tissue. Moreover, when the tuberculous tissue undergoes caseation and the epithelioid cells undergo necrosis so that a fibrous capsule alone persists, protein-digesting activity disappears from the tissue. Autolysis in the presence of acid is exhibited by the liver, spleen, and kidney, and these organs exert a limited power to digest foreign protein. There are at present no available means of determining if the enzyme of tuberculous tissue is a peculiar enzyme or is identical with the autolytic enzyme of certain other tissues. Of especial interest is the observation that the enzyme of phagocytic cells which are capable of intracellular digestion is more active than the autolytic enzymes. Opportunity for an accurate comparison is afforded by the liver studded with innumerable miliary tubercles. Such tissue contains much more enzyme than normal liver. A peculiarity of the serous effusion which accumulates in the infected pleural cavity in contact with the tuberculous tissue 218 HARVEY SOCIETY previously deseribed emphasizes what has been said concerning the character of the enzymes contained in this tissue. Such serous effusion, like other serous effusions, inhibits the enzyme of the polynuclear leucocytes but unlike the serum of all other inflammatory exudates which have been tested, fails to restrain the enzyme which is abundant in the tuberculous tissue. To complete the study of enzymes produced during the course of an inflammatory reaction, it is necessary to examine the adjacent lymphatic nodes. Such tuberculous nodes show enzymotic action which differs in no respect from that of the tuberculous mediastinum. The sinuses of the node are filled with large moncnuclear phagocytes, many of which contain tuberele bacilh. Before caseation has begun, the histological appearance resembles that of the same node during the late stages of pleurisy produced by a sterile irritant such as aleu- ronat; and in both instances there is active enzymotie power of the same character. Evidence of the existence of lipolytic enzyme in the eells of tuberculous exudates and in similar mononuclear cells from other sources has been obtained first by Bergel.*? On plates of wax small excavations are produced after a period of ineuba- tion by exudates containing lymphocytes and especially by the exudate obtained from so-called tuberculous abscesses ; ordinary pus produces no superficial solution of the wax plate. Tuber- culous pus-lke exudates, moreover, are capable of splitting neutral fat obtained from butter. Lymphatic gland and spleen pulp have similar lipolytie action, but bone-marrow, according to Fiessinger and Marie,** who have confirmed the observations just cited, fails to exhibit it. These authors have injected wax and various fats:into the subeutaneous tissues and peritoneal cavity of animals and have found that polynuclear leucocytes first accumulate; an intense mononuclear reaction follows and effects the absorption of the fat. They think that the wax-like substance of the tubercle bacillus is dissolved by the lipolytic enzyme of the mononuclear cells. The conditions under which in the body the intracellular enzymes act and the factors which bring them into action are INFLAMMATION 219 not clearly understood. Intracellular digestion by amcebas and other protozoa occurs in the presence of an acid medium and granules of litmus, and other indicators ingested by amcebas undergo the usual color changes indicative of an acid reaction. When phagocytic cells of vertebrates are allowed to ingest such indicators in granular form, no such change of color occurs. Whatever change of reaction occurs is not indicated by this cross method. The enzyme of the polynuclear leucocytes is active in a neutral or alkaline medium and its behavior in vitro indicates that the reaction of the normal body fluids is favorable to it. The acids, such as acetic acid, which have usually been employed to demonstrate the activity of the enzyme of the mononuclear phagocytes are not present in the cells or in the serum. Never- theless, other acidifying substances such as carbon dioxide, or lactie acid, are capable of bringing the enzyme into action. It is not improbable that conditions which diminish the oxidation of pathological tissue or inhibit its gaseous interchange increase its acid content and produce conditions favorable to the action of the enzyme. Solution of bacteria, such as pyogenic cocci, is doubtless effected by the proteolytic enzymes contained within the poly- nuclear leucocytes. Metchnikoff has brought abundant proof that living bacteria are ingested by the leucocytes, but it is uncertain what part enzymes have in destroying bacteria. The proteolytic enzyme of the leucocytes and the bactericidal com- plement of the serum are not identical. Abundant histological evidence previously cited has shown that the mononuclear cells which accumulate at the primary site of inflammation dissolve within their substance polynuclear leucocytes, many of which have probably undergone degenerative changes before they have been ingested; this process is continued and completed in the adjacent lymphatic nodes. Indeed, it is not improbable that polynuclear leucocytes, together with other products of tissue degeneration, serve as the principal stimulus to the activity of the mononuclear cells. Such intracellular digestion of polynuclear leucocytes is the first step in the resolution of an 220 HARVEY SOCIETY inflammatory exudate. There is secant evidence that polynu- clear leucocytes disappear by autolysis unless suppuration oceurs. Absorption of fluid constitutes a second factor in the reso- lution of an exudate. When, with diminishing activity of the inflammatory irritant, exudation from the blood-vessels ceases, the physiological factors which favor absorption of tissue juices rapidly diminish the accumulated fluid unless the inflammatory irritant or inflammation itself has produced changes which alter the adjacent vascular and lymphatic structures; necrosis, sup- puration, which is always accompanied by necrosis, and new formation of fibrous tissue, three conditions which are usually associated, produce such structural changes. The large mononuclear cells which act as phagocytes are at first only slightly larger than the cells which they ingest, but those which are engaged in digesting many cells attain great size. The fate of these large cells after they have accomplished their function is probably not always the same. Some may enter lymphaties and reach adjacent lymphatic nodes. According to Maximow, some undergo degenerative changes, whereas others remain in the tissue. It is not improbable that disap- pearance of exuded fluid produces conditions unfavorable to their prolonged existence and many probably undergo auto- lysis. Diminished blood-supply and other factors which might impair oxygenation doubtless increase the acidity of their pro- toplasm and favor self-digestion. Human pathology affords numerous instances in which inflammation pursues its course without noteworthy destruc- tion of tissue and, followed by complete restoration to normal, is unaccompanied by any fibrous induration of the part. Lobar pneumonia, acute serofibrinous pleurisy and erysipelas may be cited. Such inflammatory reactions are well represented by the serofibrinous inflammation which follows the introduction of turpentine into the pleural cavity of an animal. The fibrin of such an exudate undergoes autolysis in vitro under conditions which indicate the presence of leucoprotease only during the first three days after onset of the reaction. During this early INFLAMMATION 221 stage autolysis occurs when the fibrin is suspended in weak acid and this ability to undergo self-digestion in acid persists at a later stage when fluid has completely disappeared from the chest. Fibrin obtained by whipping freshly drawn blood exhibits the same property. Since the blood-serum contains an enzyme exhibiting similar proteolytic activity it is probable that fibrin carries with it some of this enzyme when it is precipitated during coagulation. Autolysis referable to the presence of this enzyme may explain the disappearance of fibrin which persists after the fluid of an exudate has been absorbed. In some instances under conditions which are not understood, fibrin fails to undergo absorption and organization with new formation of fibrous tissue follows; fibrin is then slowly absorbed and replaced. Further evidence that formation of scar tissue is not a necessary result of inflammation even when the reaction is inaugurated by extensive destruction is afforded by recent experiments of Whipple and Sperry *° on the necrosis of the liver after poisoning by chloroform. The hepatic cells con- stituting a large part of the liver lobule undergo coagulation necrosis; a considerable number of large mononuclear phago- cytes collect at the site of injury and accomplish the absorption of the dead liver cells. By active multiplication of adjacent liver cells, the parenchyma which has been destroyed is replaced and no new formation of fibrous tissue follows. The liver is restored to normal and there is complete absence of cirrhosis, though a bit of tissue removed three weeks before has demon- strated necrosis of three-fifths of each hepatic lobule. Human pathology affords little evidence that tuberculous exudates may undergo resolution with restoration to normal; yet such resolution is doubtless possible and is probably accom- plished by the same enzymotie action, which brings about the disappearance of an acutely formed exudate. Experiments of J. L. Nichols ** have shown that the exudate of tuberculous pneumonia in immune rabbits undergoes complete resolution. After suppuration has occurred, restoration to normal by the processes which have been described is no longer possible. 222 HARVEY SOCIETY The inflammatory reaction pursues the course which brings it to an end only when enzymes set free by disintegration of polynuclear leucocytes are fully held in check by the serum which accumulates. When intensity of the irritant calls forth increasing numbers of leucocytes, and the density of the tissue affords restricted opportunity for accumulation of fluid, free enzyme overbalances antienzyme and fibrin, necrotic tissues, and perhaps to a limited extent adjacent living tissues undergo solution; in the wall of the abscess fibrous tissue is formed; what is the immediate stimulus to the new formation of fibrous tissue has not been determined. Since long-continued inflammation is associated with new formation of fibrous tissue, such sclerosis has been commonly used as an index of chronic inflammation. Increase of inter- stitial tissue may furnish evidence of pre-existing inflammation even though the regenerative changes in the connective tissue are not included in the conception of inflammation. Never- theless, the resulting confusion has introduced many incon- sistencies into the nomenclature of disease. In many instances of hepatic cirrhosis, the increased inter- stitial tissue is sclerotic and sear-like and all evidence of inflam- mation is wanting; the lesion, indeed, has all the characters of a scar and chronic hepatitis is not more applicable than is chronic inflammation to the sear from a burn of the skin (Marchand). The same remark is applicable to certain in- stances of granular atrophy of the kidney and to chronic lesions of other organs. Such diseases are a combination of degenera- tive change, notably necrosis, inflammatory reaction, regenera- tion of parenchymatous elements, and regenerative changes affecting the interstitial tissue. The relationship of these processes has not been sufficiently analyzed. In most instances of so-called chronic endocarditis the exist- ing lesion, perhaps preceded by inflammatory changes, is sclerosis of the valvular segments, and functional derangement of the valve is referable to peculiarities of scar tissue found in any part of the body. The same objection is applicable to fibrous myocarditis, applied to the lesion which oceurs in asso- INFLAMMATION 223 ciation with arterial disease, impairing the vascular supply of the cardiac muscle. The common designation of chronic arterial disease does not have the affix ‘‘itis’’ indicating its inflamma- tory origin, but arteriosclerosis is used almost synonymously with endarteritis and mesarteritis, lesions in which degenerative and regenerative changes are conspicuous, whereas true inflam- matory reaction is in most instances wholly absent. Thoma has pointed to the truth that the present use of the term ‘‘chronic inflammation,’’ applied to the liver, kidney, heart, blood-vessels, and other organs, means nothing more than chronic disease. Study of pathological structure, eagerly pur- sued during the last two centuries, is not infrequently regarded as an unprofitable field for investigation and perhaps this view is correct should its scope be limited to the observation and description of pathological lesions; but examination of present knowledge concerning the nature and classification of various forms of inflammation shows how meagre is our know tele con- cerning the significance of altered structure. If it were possible to define the origin of the mononuclear cells concerned in the inflammatory reaction of all vertebrate animals as well as it is possible to define the character and source of the common polynuclear leucocytes concerned in the same phenomenon, it might be possible to describe with an accurate generalization the essential nature of the cellular accumulation which follows the action of substances foreign to a tissue. The possibility that the various mononuclear cells which accumulate are derived from the lymphocytes of the blood, offers attractive solution of the matter; but proof is wanting. Bulbi ®% und unten. j¢Nach oben gestellte Bulbi in die Mitte OGesicht x { ) Auris. Apertura oculorum. ©Hochziehen der Augenbrauen. Die iibrigen gelben Zeichen sind aus den eben erklarten zusammengesetzt. Fic. 37.—Vogt’s record of electrical stimulation in monkeys. Fic. 38.— Integrity of the knee of the internal capsule in the brain (Fig. 11). Fic. 39.—Brodmann’s map of the lateral view of the hemisphere. OO APHASIA AND APRAXIA 289 striate zone less so, the lateral field of the occipital lobe passing into the so-called posterior association field of Flechsig, which also has a few as yet unexplored islands of specialized appearance. The typical integrative function of the hemisphere for vision (Fig. 40) is evidently half-vision, and its negative, hemianopsia. The structures with which we can reckon to-day are the striate cortex, probably the exclusive end-station of the geniculo-calearine tract or optic radiation; the relation between cortex and the occipital efferent path is probably broader, but the functional réle of this large efferent path is a secret. The association paths studied on a most interesting ease, of limited bullet wound (Fig. 41) from the Willard State Hospital, were limited (Fig. 42) to a connection with the motor zone, and some fibres towards the auditory zone and others through the callosum, while in the deep or sagittal marrows the external fibres degenerated plainly backward, and the internal fibres forward. The functional integrations or losses of integration of this region point to a quadrantie representa- tion, so that the upper lip of the calearine cortex and the more dorsal radiations go with hemianopsia in the opposite lower quadrants of the visual field; and excitation of the upper lip with movements downward; if Wilbrand’s third observation can be accepted, a horizontal strip of half-blindness may cor- respond to lesion of the valley of the cortex (Fig. 43). An efferent motor path attributed to the angular gyrus is not demonstrated. The visual integrations are thus singled out by: 1. Hemianopsia (in quadrants or complete, either only for colors or usually for forms and light as well), usually from lesion of the posterior cerebellar artery (Fig. 44). 2. Mind-blindness (difficulty of interpretation with suffi- cient vision—separation of the visual sphere from the posterior association field, Fig. 45). 3. Simple alexia (mind-blindness limited to words or letters), superseded by 4. Alexia with agraphia—a more intense disorder inter- fering with the planning of letter forms (Fig. 34). 240 HARVEY SOCIETY 5. Isolated disability to recognize distances and depth. These varieties are more easily explained by lesions of the optie radiation and the striate cortex and by the lesion of the connections with or among the other leading parts of the hemi- spheres than by the assumption of special centres for each type. With the auditory field (Fig. 46) we seem to approach ground much more definitely related to the speech-function evidently with some differentiation of the parts. Mott’s case (Fig. 47) shows that the bilateral destruction of the transverse temporal gyri entails complete deafness, and to this we have to add the striking regularity with which circumscribed lesion of the auditory sphere in the leading hemisphere goes parallel with definite disorders of the speech integrations, a finding corroborated independently by Quensel. Spiller published a negative case, destruction of the whole area without any apha- sia; but in a ease in which also tabes was found only post mortem. The general uniformity is surprising in view of the fact that in the dog, tone-differentiation can be obtained even after ablation of both temporal lobes as shown by the note- worthy training experiments of Kalischer. Functionally we have the following steps of integration and disintegration in the material at hand: 1. Complete deafness from bilateral lesion of T tr. (Hemi- acusia from unilateral lesion? Cerebral loss of special tones or islands of the scale unlikely.) 2. Loss of word-identification, without further disturbance (subcortical and partial bilateral lesions). 3. Preservation of mechanical repetition without under- standing. 4. Loss of word-identification and disorders of elaboration (paraphasia, anomia, ete.), lesion of T tr. and of connections. The greatest advance in the general co-ordination of data of the tactile and motor integrations comes from a brilliant pupil of Wernicke and his discovery of unilateral apraxia. Liepmann examined a man committed to an asylum as a case of apoplectic dementia. The patient was aphasic and had >, [>} "SoD > > Dob. *\* > D. 7 keane eat ii Fic. 40.—Brodmann’s map of the mesial view of the hemisphere. The area striata or visual receiving station proper with fine dots, the area peri- striata with dark cross-hatching. Fic. 41. Fic. 42.—W, wound. X, degenerated fibres of the external sagittal marrow marked by dots (behind the lesion). The normal] lower rest of the external sagittal mar- row indicated by a compact dark streak. Y degenerated fibres of the internal sagittal marrow in front of the lesion. Wilbrand’ Sarnges; Neurologic des Auges ll ™ 20, Fic. 43.—Smallest defects of visceral field of supposed origin in small lesions of the calearine cortex (Wilbrand). Fic. 44.—Typical occlusion of the posterior cerebral artery with right homonymous hemianopsia. Fic. 45.—Mind-blindness in bilateral symmetrical occlu- sion of the postparietal artery. APHASIA AND APRAXIA 241 but few words left. He had no hemiplegia, but a slight dis- turbance of sensibility on the right side, as was later seen, especially for position and motion. This patient seemed to be completely demented, and since he merely fumbled very queerly when asked to do things, he was under suspicion, at first, of being word-deaf, and even mind-blind. Liepmann then observed that the patient carried out any request for which the body as a whole was required; he would go to the door as told, sit down, get up, etc.; he further saw that the confusion invariably started with a strange fumbling of the right arm, and by it the whole course of action would be side-tracked. Liepmann found that when he held the patient’s right arm and forced the left arm into initiative, the dementia disappeared: the right cards were picked out, even writing was possible, and the whole case was plainly one of pseudo-dementia and really a ‘one-sided apraxia.’ Instead of being looked upon as merely an ordinary demented individual, the patient was recognized as having mainly a partial disorder. On examination of serial sections, the motor area and the pyramidal tracts were found intact. There was a subcortical softening underneath F', and F’, and a cyst under the supra- marginal and the inferior parietal gyri, and moreover degenera- tion of the corpus callosum with the exception of the splenium, and a shrinkage of the right angular gyrus. What simulated that which in neuropsychological slang is called a loss of memo- ries of movements, proved to be the partial lack of support by the sensory part of the brain, and the inability to use the experience of the other side. A deep disorder of tactile re- sponses, a complete hemianesthesia, usually shows rather differ- ent results from what this patient showed. One of my cases with hemianesthesia and atrophy of the parietal region could move his left hand with some ease and force as long as he looked at it; but the movements were ataxic and when he shut his eyes or kept his hands out of sight, and I asked him to close his hands rhythmically, he ceased doing it with the left and still thought he was keeping it up (mind palsy). He was wholly unable to recognize objects within the hand which had the 16 242 HARVEY SOCIETY receptor difficulty. In Liepmann’s case there was no ataxia, and the hemianesthesia was not profound, nor was the astere- ognosis complete. Liepmann, therefore, concluded that at least part of the apraxia, of the faulty motor responses, was due to improper use of the motor cortex owing to isolation and not owing to destruction of any special part of its intrinsic motor mechanism; and he laid the emphasis to quite an extent on the lesion beneath the supramarginal gyrus. Further studies led Liepmann to the observation that in some cases of right hemiplegia the left side had lost the capacity to indicate from memory very ordinary signs and movements (the patient would be unable to indicate with his non-paralyzed left arm the motions of beckoning, threatening, throwing a kiss, making a fist, saluting, ringing a bell, counting out coins, catching a fly, grinding an organ, playing the fiddle, snapping the finger, or the use of a comb or brush, or lighting a candle or putting on a stamp), while actual handling of objects, the actual use of a match, ete., was less frequently affected. A few of the cases examined anatomically so far corroborate the hypothesis that this loss is not merely one of intelligence in the broad sense of non-localizable general capacity, but due to the interruption of the callosal path and the elimination of the help of the leading hemisphere, a point which can indeed be used for the localization of lesions (Fig. 48, from Wilson). These studies have given a remarkable impetus to the whole problem of synthesis of cerebral activity and to anatomoclinical correlation. Hartmann, who participated in the studies of Anton on the frontal lobes, communicated two eases of lesion of the frontal lobe and one of the corpus callosum, which led him to the con- clusion that the impetus to serial movements roused from the various sensory spheres of the cerebrum requires the co-opera- tion of the frontal lobe for the imparting of their impulses to the motor zone. The elimination of the left frontal lobe leads to motor mind palsy or loss of initiative of the opposite extremity. The right frontal lobe requires the co-operation and guidance of the left or leading frontal lobe beside the connec- Fria. 46.—Bordmann’s subdivision of the island and trans- verse temporal gyrus. Fic. 47.—Mott’s case of bilateral destruction of the transverse temporal and auditory zone. Complete deafness with preservation of reading. 5 1 ' Right sensomotorium sensomotorium {nternal \ 37 Left hand Right hand Fic. 48.—(From Wilson after Liepmann.) 1, cortical lesion; 2,subcor- tical lesion; 3, capsular lesion; 4, lesion of corpus callosum; 5, subcortical lesion. APHASIA AND APRAXIA 243 tion with its own sensory spheres to have its serial movements properly guided. Thus we see again that separation of the left frontal lobe from the right abolishes those serial movements of the left side which depend on memory. We have here another but more definite formulation of the data of Liepmann, and a prospect to bring harmony into the impressions concern- ing the production of agraphia and aphasia through lesions approaching the motor zone from the frontal lobe, as was the case in Gordonier’s case of isolated agraphia. We seem to have gone far away from the consideration of the cortical areas of the frontal lobe furnished by the brain- maps. From what was said before, it is evident that a most painstaking inquiry into the details of anatomical relations must nevertheless pave the way before it pays to speculate about the mode of insertion of the cerebral motor apparatus upon the subcortical and segmental mechanisms or lower nuclei. But we certainly have gained some broad lines of hemisphere function worth summing up. We can distinguish now: Intellectual disorders pure and simple, not specially charged to any of the senses. Focalized intellectual disorders appear either on the recep- tive or the emissive or in the co-ordinative or elaborative func- tion. The common link is covered by the term agnosia, which covers knowledge of things sensed and remembered, as well as knowledge of how to do things. Agnosia (see diagram) has its more essentially receptive aspect in general or ideatory asymbolia or disorientation, and its more essentially emissive aspect in general or ideatory apraxia. Asymbolia, complicated by essential anesthesia of general or special senses, or independent, can be partial; the disorder of primary identification may be kinesthetic, as astereognosis, or visual, as mind-blindness, or auditory, as mind-deafness, or involve the taste and smell sphere. The agnosia may show especially in the effects on the wording (and then probably involves all the senses), so that probably there is no such thing 244 HARVEY SOCIETY as isolated ‘optic aphasia’; or agnosia affects more the direct motor reactions. We may miss the proper kind of plan of action, as in ideatory apraxia; or the plan may be correct, but one whole side or only one limb or motor unit like the arm or the tongue is out of commission for the act, although not paralyzed as such; this covers the limb-kinetic apraxia in the frontal lesions; and as dyspraxia, I should group the losses from severing the corpus callosum. Beneath this we may further find the possibility of Submental hemiplegic Essential anesthesias complexes Disorders of primary Dyspraxia identification Limb-kinetic apraxia Astereognosis One-sided. Mind Palsy Mind blindness Motor apraxia Cerebral Ataxia Mind deafness? General” Ideatory apraxia’ “Ideatory asymbolia ? AGNOSIA INTELLECTUAL DISORDERS Emissive Disorders—Elaboration Disorders—Receptive Disorders APHASIA Gen. loss (motor aph.) Paraphasia Word deafness Limb-kinetic loss (pure Anomia Word blindness motor aphasia and pure agraphia) Disorders of initiative interference with mere submental co-ordination as in the ataxia dependent on anesthesia, and corresponding reduction of the sensory support of the motor apparatus of the hemi- sphere leading to mind palsy. As a counter part of the other tables of integrations we can therefore give the following summary for the tactile and motor sphere: Tactile and muscular sense: APHASIA AND APRAXIA | 245 1. Hemianesthesia with possibility of some circumscribed foci (more or less of the axial type in the limbs) and possibly a splitting off of the muscular sense by itself. 2. Astereognosis alone (elaboration disorder). 3. Astereognosis with hemianesthesia. (Note the special loss of the muscular sense component. ) 4. Hemianesthesia with ataxia, hemianesthesia of tactile sense component (tactile aphasia and dyschiria hysterical). Then the motor integrations—motility : 1. Complete hemiplegia (or possibly monoplegia in lesions above and in the internal capsule): (a) either Wernicke’s type by joints (purely cortical), or (b) the normal arm-flexion and leg-extension type, or flaccid, or with contractures. Complete flaccidity and occasional atrophy depend on the thalamus involvement; athetosis and tremor as complications or alone belong especially to the rubral complex; while lesions of the pyramid in the medulla and decussation give no symp- toms, at least in the monkey; lesions of the lateral columns of the cord again spastic hemiplegia; and uniform diffuse weak- ness of trunk and extremity is said to go with one-sided lesion of the pyramid in the pons. 2. Mere ataxia or inco-ordination of balance of movement through lack of sensory support (cerebral ataxia). 3. Lack of control of execution as in simple mind-palsy. 4. Motor apraxia (through relative isolation) in which the patient has the sensory and intellectual support needed for a movement, and the movements perfectly well planned, but inability to perform such movements as making a fist, showing the tongue, or speaking a word (which the patient can plan and perhaps can write), or at least a wholly senseless miscarriage or substitution, such as raising an ink-well instead of showing the tongue. 5. In ideatory apraxia, crude movement is preserved and also the sensory support; but the intellectual support is lacking in the form of lack of attention or of planning as shown in the failure of the use of a candle, a pistol, a shoe, ete. (Psych. Bull., 11, 279.) 246 HARVEY SOCIETY Aphasia in such a scheme becomes a part of the agnosia- asymbolia and apraxia problem. A certain type of associative reactions involve linguistic signs. According to our survey we distinguish linguistic elaboration or co-ordination, that which has been ealled inter- nal language, or which covers Marie’s linguistic intelligence. In contrast to Marie’s views I am, however, inclined to recognize the close relation of the word-deafness to lesion of the auditory entrance zone and the relative independence of the word-blindness complex and its connection with the visual apparatus, without, however, being able to determine a cortical area or any definition of the connections, except the general rule that the more we encroach on the block under the so-called posterior association field, the more apt do we become to disturb even the mechanism of writing and to get complications of the paraphasia-paralexia type. Evidently few cases work with both hemispheres, to judge from the rarity of recovery by substitution. On the motor side we evidently meet with more ambidex- trous organizations, and an understanding of the mechanism in anatomical terms is as yet difficult. There is a broader zone of vulnerability, though not quite as large as the quadrilatére of Marie, and it seems to take a large area practically involving the motor operculum to abolish speech permanently. The Broca area is an empirical unit not to be ruled out completely. The duration and recoverability in motor aphasia depends on the size of the lesion and possibly the extent of ambidexterity. The broader interruption in the cases of Liepmann and Hart- mann suggest the possibility of a definition of the frontal lobe functions as the co-ordinator of serial movement. To put the present position briefly : The available facts suggest a plan also chosen by Rieger in his very original studies on hemisphere-function, viz., that of following the impressions of the various senses to the responses along the line of speech and symbolic thought (der sprachlich-begriffliche Apparat) on the one hand, and along the line of direct activity (riumlich-sachlicher Apparat) on the APHASIA AND APRAXIA 247 other. Within each field we may have our special and isolated disorders open to interpretation by localization or along the lines of the disorders of serial movement, the analysis and synthesis of reactions, as Rieger depicts the disorders, and the mere disorders of control. The great risk of didactic schemes is excessive condensation and excessive desires for one-word designations for really com- pound disorders. Size up and detail the mechanisms which enter into the picture, call the spade a spade, the functional disorder a functional disorder, and the lesion a lesion, but do not try and mix it all into one confusing word. The only real and lasting solution comes from the utilization of adequately defined anatomical and functional mechanisms (which can more safely replace the rough schemes of Ferrier and the early localizers). We cannot afford to brush aside really command- ing facts, and replace them by vague units such as the lenticular zone and an arbitrarily extended Wernicke zone, without slight- ing the natural lines of analysis and synthesis. Marie’s strong appeal is a recall to general perspectives of great value, viz., (1) the close relation of the intellectual and linguistic differ- entiations; (2) their great dependence upon the posterior field ; (3) the great variability of the lesions and results in connec- tion with the motor utterances; but the simple data of large enough chains of consecutive observations bring out the true points of his claim quite naturally, without any need for arbi- trary negations. Von Monakow has introduced the term diaschisis to account for the fact of transitory interferences with function and rela- tive localization. He wishes to distinguish clearly the mere vascular shock and pressure and distance effects and the fune- tional disorders actually attributable to the effect of cutting fibres which would enter into the other mechanisms which thus become deprived of certain functional influences without being directly injured as such. He assumes that where we deal with transitory symptoms it is due to indirect and remote dynamic effects and that the effects from actual destruction alone would give the permanent symptoms. I cannot befriend myself with 248 HARVEY SOCIETY the term; all lesion is diaschisis and the recoverability or per- manence depends on the question whether balance of function can be maintained with what is left or not. A diaschisis accord- ing to von Monakow would then cover all the losses of function or balance which need not be permanent. Von Monakow assigns all apractic asymbolic and aphasic disorders to such diaschisis. After all we must admit that if you destroy enough, you will finally get permanent loss even with mechanisms which are very diffuse and not strictly localized. Bickel’s experiments on the reappearance of cerebellar defect symptoms through new lesion of the cerebral motor area and many other data of motor regulation show a certain amount of surplus provision and factors of safety in our organization and many transitory disorders are merely a disturbance of adjustment, possibly in an otherwise weakened part. Diaschisis is there- fore the thing to be explained and not an explanation; it is a matter to be sized up in every case and a query as to which part is thus weakened, if it is to be more than an emphasis of the obvious for those of us who have a broader view of localiza- tion. If we cut one leg off a tripod it falls, and yet you would not say that the tripod stood on that leg alone. If there are more than three legs the loss of one is less serious and as we see at times in dogs almost negligible. That many transitory disorders are such disorders of balance is a much safer assump- tion than that of a peculiar agency acting at a distance, as when von Monakow tries to explain the 11 years of pure motor aphasia in his case as a corticobulbar diaschisis. When the physician meets a case of speech-disorder, he does well to pin his attention first on the existing motor and sensory disorders, and secondarily upon the capacity for elabo- rations from the various points of hearing, vision, tactility, ete., (1) into the various types of direct action and (2) into speech- or thought-functions, with attention to the possible independence of the hemispheres and of the individual mechan- isms. The sum of free or blocked connections ultimately indi- cates best the probable extent of lesions. The next considera- tions are the nature and extent or number of lesions. Exten- APHASIA AND APRAXIA 249 sive autopsy experience tempers one’s diagnostic assurance. A symptom or complex may be due to a simple or unitary lesion, or to several foci, or a focal accentuation of a diffuse process, such as senile or paralytic atrophy, or merely the result of one of those as yet insufficiently explained causes of hemiplegia in uremia or other asthenic states (hemiplegia without lesion). Their distinction is an issue of successive developments rather than of final combination of symptoms. Before concluding I must turn briefly to a group of experi- ences more or less unanalyzed, but suggesting distinctly an attempt to explain the condition by some focal process. I remember a patient who suddenly developed a pseudo-Kor- sakow state, really a systematic disorientation lasting till death within several months—no autopsy. Another somewhat pecu- lar Korsakow syndrome was ultimately found to be dependent on traumatic lesions of the frontal and temporal base (State Hospital Bulletin, vol. i, 140, 1908); further I recall certain peculiar conditions in which a patient would speak of fictitious matters and then complained that she was a humbug, who said things she did not want to utter and that were not so; a form of paralogia rather than paraphasia (Kr. and Mi.), twice observed by me in connection with brain- tumor compressing the anterior part of the Sylvian fissure; further the peculiar states of focal hallucinations connected with hemianopsia, and finally the occasional conditions of vitia- tion of character and of a peculiar jocose attitude in tumor of the frontal lobe, or as Taylor in Boston observed it, a condition of psychasthenia; or we meet states simulating hysteria in brain-tumor—these are grounds which sometimes present great difficulty even if we have fairly definite principles with which to work out the less complex problems. In all these doubtful states the only safe road is to analyze first the motor and sensory symptoms, and the symptoms of language in direct connection with them with attention to the several levels or units, viz., (1) word identification, word- finding, word-elaboration, paraphasia, and special attention also to music, for the auditory zone and the lower parts of the 250 HARVEY SOCIETY lateral* parieto-occipital region; (2) the reading or reading and writing disorders for the posterior and upper areas of the parieto-occipital zone, supplemented by the search for defects in the lower, visual quadrant, and evidences of disorientation, or (3) the sensory area with the conditions of astereognosis. In the domain of motor speech and writing, the appearance of the speech symptoms before any motor or sensory symptoms speaks strongly in favor of a disorder in front of the motor area; otherwise it may be mere guessing to decide in what direec- tion from the actual tongue-larynx region the focus may be. I must pass the problem of distribution of the mechanisms according to blood-vessels. While this offers excellent land- marks the problem will always be complicated by some indi- vidual variations, the possibility of multiple lesions, or of diffuse alteration beside the focal one, and the occasional exist- ence of slowly progressing affections with sudden apoplectiform manifestations. The slowness of progress in this complex problem is to a large measure due to the uncertainties as to whether autopsies can be obtained and as to whether the great amount of work required in the repeated examinations will be allowed to receive its final control. The other reason is the great amount of work and expense implied in the examination of the brains. In all these respects physicians should collaborate not only in the State hospitals (where the cases are often greatly complicated by mental disorders, but are studied with growing efficiency and admirable determination, so as to put the scientific and medical world under growing obligations), but especially also in general hospitals, in traumatic states or embolisms, not complicated by additional deterioration of the brain. Another very com- mendable way to help is encouraging fellowships of research. Perhaps when the North Pole and the South Pole shall at last be properly discovered, man will bestow more attention upon the most wonderful creation of nature, the organ of plasticity of behavior. URIC ACID IN GOUT* PROF. A. MAGNUS-LEVY University of Berlin, Germany INTRODUCTION P to the present time the only chemical problem considered in gout has been that of uric acid. As yet no other chemical stand-point has been found. Are we on the right track in doing this? Is the uric acid really the principal toxic agent in gout, or only one of several? From time to time, espe- cially when there appeared to be no progress in the investigation of the subject, protests arose that research was on the wrong path. The relations between uric acid and the occurrences in gout became evident in the acute periods. The attacks are certainly connected in some way (yet to be determined) with the accu- mulation of this substance in the body, and during the attacks the output of uric acid is greatly altered. Experiments made with uric acid injections prove their toxic character ; the inflam- mation and the pains provoked by them resemble in many respects those occurring in attacks of gout. Though conceding this, Von Noorden a number of years ago, following Ebstein’s theory of the formation of uric acid in the bones, expressed the opinion that the deposits of uric acid were quite independent of the general metabolism of uric acid. Although local influences play an important part in gout, this view of Von Noorden is easily to be refuted. The microscope shows clearly that the erystals are first found near the free surface of the cartilage and that they advance slowly towards the deeper layers. Thus they must pass from the * Delivered March 19, 1910. 251 252 HARVEY SOCIETY cavity of the joints into the cartilage. Secondly, when a tophus on the finger reaches the size of a chestnut and even larger, it is impossible to imagine that the enormous masses of sodium urate stored up here could be formed in loco, even in the course of many years. The metabolism of connective tissue is much too torpid, and nuclear material, known to be the sole source of uric acid, is contained only in small amounts in the euticle and neighboring tissues. Moreover, wherever you find deposits of sodium urate in the body, that is, wherever gout is apparent, there occurs at the same time an excess of this substance in the blood. Although this principle is not reversible, it proves cer- tain connections between the local and the circulating uric acid. PRELIMINARY NOTES In reference to uric acid, gentlemen, please remember that in reality we have to deal only with monosodium urate in the fluids within the body. Some authors have pretended the exist- ence of a quadriurate of sodium, or as we would eall it now, of monosodium biurate or hemiurate; others have insisted that in a fluid containing a number of basic molecules like serum, the chemical union of an acid with one of the bases could not be recognized. Physically considered there exists no mono- sodium urate in the serum, no more than does monosodium carbonate. All these salts are dissociated into their ions, and we have to deal only with sodium, potassium, calcium ions and with uric, chloric, carbonic acid ions, existing side by side. But with the same right with which one speaks in the common lan- guage of chemistry of the existence of sodium chloride, of monosodium carbonate and monosodium phosphate in the serum, with just as much right you may also speak of a monosodium urate. Ninety per cent. of the kations in the serum belong to the sodium, only 10 per cent. to potassium, calcium, and so on. In a neutral solution which, in a physical sense, the serum is, the proportion between kations and anions, or between acids and bases, is such that we can only speak of monosodium car- bonate and of monosodium urate. The existence of bisodium URIC ACID IN GOUT 253 urate is just as impossible here as the existence of a bisodium carbonate. It is to Gudzent that we owe this clear exposition. FUNDAMENTAL PRINCIPLES As compared with former times we now stand on a firm foundation in regard to our knowledge of the formation of uric acid. This substance is derived from the nuclear purins, adenin, guanin, hypoxanthin, and xanthin. Their transfor- mation into urie acid is the work of hydrolytic, or desamidizing and oxydizing ferments. The ferment which katabolizes uric acid is called the uricolytic ferment. I need not detail these processes, but will refer you to Prof. Mendel’s lecture delivered on this subject three years ago before this society. This being established, two questions arise: First, whether this intermediate way of metabolism is obliga- tory in the sense that every molecule of the purin bases must reach the stage of uric acid, or whether adenin or hypoxanthin, and so on, can be katabolized without first being converted into uric acid. Thus far the possibility of the latter proposition has never been demonstrated in the organism. Chemical consid- erations and biological analogies lead us to consider the trans- formation of purin bases into urie acid as the only process effective in mammals. The second question, whether in mammals there are still other sources of uric acid, is doubtful. No proofs have yet been brought forward of formation of uric acid from urea and from acids with three-carbon atoms, a process which is of fundamental importance in birds. So we may neglect this kind of synthesis. On the other hand it is beyond question that purins are newly formed in suckling animals, whose food is practically purin-free. It seems improbable that the synthesis effected here starts from urea; we must rather look for a higher amino-acid compound, as a material which can yield purins. It would be very surprising if this synthesis were performed on a larger scale than was needed for the growth and for the renewal of the tissues, 7.e., for the synthesis of nucleoprotein. 254 HARVEY SOCIETY I do not think that in this process an excessive formation of purins takes place, nor that superfluous material for the for- mation of uric acid is left over. We may, with our present knowledge, consider the purins as the only source of uric acid in human metabolism. We distinguish the exogenous purins ingested with the food from the endogenous derived from the nucleins of the body. By dieting we can make ourselves independent of the great variations in the output of uric acid due to the different amounts of ingested purins. By giving a purin-free food, which need not be absolutely the same every day, the urine will contain endogenous uric acid only. The quantity eliminated, although varying in different men from three to six tenths of a gramme, is fairly constant in the same person under normal conditions, thus proving the existence of an equilibrium between forma- tion and destruction of uric acid. The simplest conception of this is, not that formation and destruction vary always in the same sense and proportion, but that both formation and oxidation are constant. The katabolism of uric acid leads to the formation of allan- toin in dogs and rabbits. In men its fate is unknown. In contrast with the results in most animals human organs have failed to show a distinct uricolytic power. Hence Wiechowsky drew the conclusion that uric acid was indestructible in the human body. He based this opinion also on a second fact, namely, that uric acid injected into the muscles of men is found again almost quantitatively in urine. Schittenhelm, although confirming Wiechowsky’s experiments with human organs, was able to show that uric acid undergoes destruction in living men. When a person, who was in a nearly perfect N equilibrium, was fed with 10 grammes of nucleic acid, the output of purin substances was only slightly increased, and instead, an increase of urea (or a substance behaving similarly) was found, and in amount corresponding nearly to that of the N of the ingested purin compounds. It would be of the greatest interest for phys- iological chemistry to determine what are the intermediate and end products of the uric acid break-down; but for the knowl- URIC ACID IN GOUT 255 edge of gout, this is of less importance. It suffices for the present to know, that this toxic substance disappears as such. If Wiechowsky’s opinion were right it would mean a sim- plification of experiments and of theory, for one unknown component in the complex equation of urate metabolism, the destruction of uric acid, would have been eliminated. Recapit- ulating, one may say that the present foundation for investiga- tions of the metabolism in gout is as follows: We have to deal only with one source of uric acid, namely, the purins; and, after excluding the exogenous purins, we anticipate finding a uniform elimination at certain periods. This we accept as a firm basis for comparison, both as regards the metabolism of healthy men as also with regard to the variations which occur in different periods of gout. THE FACTS IN GOUT What are the facts in gout which are established with such accuracy that we can take them as a basis for our considera- tion? What results are sufficiently probable to be taken into serious discussion? The facts and the probable truths are as follows : 1. The presence of uric acid in the blood. 2. The presence of crystalline deposits. 3. The increased output of uric acid in the attacks of gout. The augmentation can reach from three to five tenths of a gramme daily and more, and may sometimes last for a week or even two. 4, A decline in the output often precedes the attack, but this diminution is not as marked as the subsequent increase, and its duration is only one or two days. 5. In the intervals between the attacks the elimination of endogenous uric acid is asserted to be lower than in normal health. There are doubts whether this statement has been proved beyond question. 6. The elimination of exogenous uric acid is often retarded. To elucidate the connections between these data it would 256 HARVEY SOCIETY be necessary to have a complete knowledge on the quantitative side of the following processes in both health and disease: 1. Of the formation of uric acid and 2. Of its destruction. 3. Of its elimination by the kidneys. From a combination of these three processes arises a knowledge 4. Of the accumulation of urates in the body, especially in the blood. 5. Of the conditions of the sodium urate precipitation in the tissues, as well as the conditions of its removal. Herein is included a knowledge of the physical and chemical behavior of sodium urate. FORMATION OF URIC ACID Is the formation of uric acid increased or is it diminished in gout? In almost every case where we see an increased output of uric acid, for example in leukemia and pneumonia, we find an enlarged destruction of nuclear material and vice versa. In gout, at least in the intervals, and when the patient receives a purin-free diet, neither one nor the other is to be remarked. The katabolism of nuclear material therefore is certainly not increased. On the other hand, we have just as little reason to assume a diminished formation. Brugsch and Schittenhelm, it is true, have recently advanced the theory, that the total nuclear metabolism in gout is retarded. But even if this were true, I must insist that retardation of metabolism is not in every case identical with diminution. Perhaps in advanced stages of the disease, a decrease in the formation of nuclear compounds may take place, due to a kind of cachexia or premature aging. In the experiments made hitherto, I miss clearly defined state- ments as to the age and the general state of health. Most of the gouty patients treated in hospitals are in an advanced stage of their illness and evince signs of cachexia, even when showing a good volume of muscles and of subcutaneous fat. The healthy people, whose nuclein metabolism has been com- URIC ACID IN GOUT 257 pared with that of the gouty were generally younger. In order to obtain a more reliable material for the comparison between normal and gouty metabolism, stricter attention ought to be paid to this point regarding age. Bearing this criticism in mind the following exposition may be considered. In any case it seems to be certain, even though in opposition to former opinion, that the occurrences in gout cannot be explained by excessive formation of uric acid. In the older days whenever an increased output of uric acid was noticed, the cases were mostly those of wealthy people overfed with meat. Gouty patients are generally hearty eaters. The high amount of uric acid in these people is due to the ingested purins, not to the diathesis. Analogy with obesity produced by polyphagy is striking. The gouty process, though no doubt aggravated by an excess of exogenous purins, does not depend on an increased uric acid formation. DESTRUCTION OF URIC ACID In many experiments the twenty-four hour output of uric acid in the urine of gouty men is diminished. As far as this is not due to lessened formation, it proves an increase in the absolute quantity of uric acid destroyed. The accumulation in the body does not reach such an extent as to explain the difference of excretion between healthy and gouty men. This difference often reaches 100 mg. in 24 hours. Supposing this quantity were retained in the body day by day for one year only, the total sum would amount to 36 Gm. Deposits of such enormous size are extremely rare, at least in Germany. I doubt whether in the majority of cases 10 Gm. of sodium urate may be extracted from the gouty body. Since the lower quantity of uric acid in the urine is not explained by an accumulation, it must be ascribed to an increased destruction. As Brugsch and Schittenhelm rightly suggest, this does not mean absolutely that the oxydizing power of uricolytic ferments is greater than normally or that their quantity is increased. The reason may be, that each molecule of uric acid circulates a longer A UFf 258 HARVEY SOCIETY time in the body, it is exposed more often to the influence of the oxydizing ferments, and thus there is more chance for its destruction. The investigation of the blood renders it probable that such prolonged circulation of urates really occurs in gout. The conclusion that the endogenous uric acid is katabolized in greater amount is confirmed by the results of experiments in which the fate of exogenous purin was studied. In a great number of these observations the output of uric acid after feeding on sweetbread or other nuclein-containing material was less than in healthy persons. It also is to be remarked here that more time is needed for the elimination of exogenous purins than normally, in spite of the fact that smaller quanti- ties are excreted. Different gouty patients do not behave alike. There are some in whom destruction and elimination are per- formed in the same length of time and to the same extent as in healthy persons. It would be premature to assume a retar- dation of the purin metabolism as being the constant rule. I refer again to the above mentioned objections concerning the influence of age and general state of health on the purin metabolism. EXCRETION OF URIC ACID The excretion of uric acid by the kidneys depends, like that of any other substance, on two factors. The elimination can be diminished when the secretory power of the kidney is les- sened; and it ean be prevented also, when the substance in question is united to another complex and thus bound. In nephritis secretion of uric acid is strongly disturbed. Thus during favorable periods, even when no other substance is retained (in any amount worth mentioning), some milligrammes of uric acid are always present in the blood of the nephritic patient. It seems to me as though the special capacity for the excretion of urates is more limited and can more easily become deficient than any other function of this organ. I will try to explain this. In administering an excess of any substance, the level of its end-products rises in the blood. That is due to the fact that URIC ACID IN GOUT 259 the output by the kidney does not keep pace with the influx into the blood. But though with every other substance the level rises only by a small amount, with wric acid it may rise to ten times its normal height. While scarcely one-half a milligramme is contained in one hundred e.c. of normal blood, Weintraud found 5 milligrammes after feeding with an excess of sweetbread. We may now speak of the second factor, that is, of prevent- ing the elimination of a substance by its chemical combination with another. Such a theory has been offered in order to explain the non-excretion of sugar in the healthy organism. Minkowsky tried to introduce this idea into the theory of gout. He assumed a chemical union of uric acid with a nucleic acid. But the search for such a compound in the blood has not been successful. Gudzent, on applying Michaelis’ method of com- pensation to the dialysis of gouty blood, was able to show that the urates exist only in the free state. I might also mention Garrod’s thread experiment as speaking against any organic union. Acetic acid, which sets. free all uric acid in the serum or at least the greatest part, is an extremely weak acid, its strength being some hundred times less than that of hydro- chlorie acid. Under the conditions of the thread experiment it would not suffice to split any organic union. The hypothesis of Minkowsky having been thus refuted, Gudzent showed another possibility. By physico-chemical methods he pointed out that the sodium urate exists in two modifications, which differ only in their solubility. He ascribes the existence of these two forms to a difference in the chemical structure, to a tautomerism. The lactam form, that is, the substance having the formula which is generally used for the structure of uric acid, is the more soluble one, but it tends to pass into the lactim form, which is less soluble. If the differ- ence in solubility of the sodium urates is really due to a differ- ence in the chemical structure, and not purely to physical reasons, one could imagine that the tautomeric forms would behave differently in regard to their excretion by the kidneys. Experiments in this direction have not been performed. 260 HARVEY SOCIETY URIC ACID IN THE BLOOD, ETC. Normal blood is practically free from uric acid. In gout it ean contain 5-10 mg. per 100 ¢.c. after a mixed diet has been taken; after a purin-free diet the amount.is lower, being from 2-4 mg., and in no case is uri¢ acid absent. Its presence in gout cannot be explained by an excess in the production of uric acid as is the case in leukemia and in pneumonia. This con- clusion, though it is contrary to the opinions of a former time, is beyond doubt. It can be attributed chiefly or partly to a passive retention in those cases where the kidneys are suffering from what has been called kidney gout (Nierengicht), whether this occurs at the beginning of gout or in its later stages. On this point no controversy exists between the different authors. But how is the presence of uric acid to be explained in those cases where the most accurate research does not reveal any sign of nephritis? The answer is only to be given conditionally. If neither the formation of uric acid is increased, nor its destruction greatly diminished,—and these two conditions seem to be realized in gout,—if moreover the uric acid is not pre- vented by a chemical union or by an abnormal structure from passing into the urine, only one conclusion is possible. The retention is due to a deficient and restricted secretory power of the kidneys. The existence of such a renal inadequacy does not mean a real nephritis. I think it possible that a single function of this organ can become almost entirely insufficient. Later on, real damage to the kidney and a nephritis frequently follow. As stated already by Garrod, urie acid is present also in lead poisoning. Here, as in gout, it is present at a time when no sign of nephritis can be noticed. Concerning this phenome- non, resembling that in gout, one may assume the same explana- tion, namely, a deficiency in the secretory power of the kidneys, or a kind of incipient and latent nephritis. If you are not willing to accept the above explanation for the very early stages of gout because of the absence of albu- URIC ACID IN GOUT 261 minuria, you must deny it also for the first periods of lead- poisoning. Whatever interpretation avails in gout must also be admitted for lead intoxication. In any case the mechanism of the accumulation of urates in the gouty blood is not a specific phenomenon of gout. Brugsch opposes this explanation on account of an experi- ment in which uric acid injected into the muscles of a gouty patient was eliminated just as completely and quickly as in a healthy man. I do not consider this objection to be conclusive. The gouty kidney although not able to eliminate one gramme of urates quickly enough on the normal level of urates in blood, could accomplish this task very well at the higher level already existing before the injection of uric acid. That this conception is right, Widal has obviously proved by the example of the behavior of urea in nephritis. Brugsch and Schittenhelm fall to a somewhat mystical explanation. They ascribe the accumulation of urates in the blood to a diminished destruction. But when a substance has once entered the blood, its elimination depends only on the relation between this fluid and the kidneys. I think the follow- ing argument will be convincing. On feeding with sweetbread, or during absorption of pneumonie exudates, the blood contains just as large a quantity of urates as does that of a gouty patient. Now, persons who do not suffer from gout will eliminate the excess of urates within a few days, and, even when adhering to a mixed diet, their blood will soon be free from urates. In gout, on the contrary, notwithstanding the ingestion of a purin- free diet under the influence of which the formation and the influx of uric acid into the blood is much lower, uric acid will by no means disappear from the blood. Brugsch and Schitten- helm ascribe this to a retarded decomposition, though conced- ing that the total amount of uric-acid break-down is above the normal. Something is lacking in their theory and they seem themselves to be aware of it. Although denying that the threshold (“‘Schwellenwert’’) of the kidney for the output of urates is elevated in gout, they speak of a certain torpidity of this organ. 262 HARVEY SOCIETY The deduction which ascribes the retention of uric acid to the kidneys as enunciated by Garrod and adopted since by several authors, is rather repugnant to me as well as to patholo- cists in general. The disturbance which affects the entire organ- ism would, according to this opinion, depend principally on the passive retention of a single metabolic product. This concep- tion is not very satisfactory. I would be very much pleased indeed if any other solution of the problem could be found. However, rebus sic stantibus, and without fixing my opinion for the future, I find no other solution for the problem than that given above. Another interpretation might terminate the unpleasant dualism in the theory of gout. The distinction between a primary metabolic and a primary kidney gout, which has always found supporters from Garrod down to Brugsch and Schittenhelm, would become unnecessary. Against the doctrine of gout being due to deficient kidney function, clinical doubts arise which we are unable entirely to remove. Why does not gout appear in each case of nephritis, why not in leukemia where we always find an excess of uric acid in the blood? Many eases of interstitial nephritis progress so slowly that secondary changes, such as deposition of urates and other abnormal phenomena, might have time to develop. Is perhaps the retention of uric acid present only in the later stages of contracting kidneys? We know nothing certain about this, nor about the supposed deficiency of the kidneys in gout. Does the condition reach back to youth or even to childhood? The deficiency in that case would have lasted much longer than in any case of nephritis. At any rate we must emphasize the fact that in many autopsies of uremies concretions are found in the joints, without any history that the patients had ever suffered from primary gout. There is no doubt that other organs besides the kidneys must play an important part in the pathogenesis of gout. Of all the organs, only the joints and the part they play in the pathology of gout can be discussed in detail. Are the cartilages and the connective tissues implicated only in a passive URIC ACID IN GOUT 263 way in gout, or are they active? Further questions to be answered are the following: By what mechanism and at what period does the deposition occur? What symptoms are pro- duced by the precipitation of crystals and what happens to the erystals during the paroxysm ? Contrary to the opinion of Ebstein we may at present take it for granted that deposition of urates does not occur in necrotic but only in living tissue (Minkowski, Freudweiler). And also in these deposition can only occur if the liquids which circulate in them are saturated with sodium urate. As to this point whether the amount of urate of soda in the blood corre- sponds to the saturation point, the opinions of most authors are or were erroneous. It was generally believed that the serum of the gouty patient was not saturated with urate of soda. This assumption has been put forward on account of the experi- ments of G. Klemperer. He observed that 100 ¢.c. of normal serum can dissolve 150-200 mg. of uric acid; that is ten times more than it ever contains during life. The serum of a gouty patient behaved in the same manner. This observation is correct, —of that there is no question. But Klemperer failed to observe the reaction to its end; if he had done so, he would have seen the abundant precipitation of urates after some time. Serum indeed has the power of dissolving a good deal of uric acid, by transforming it into urate of soda, but only a small amount of the soda salt is kept in solution. The prob- lem has been attacked from a wrong direction. The biological question is not how much uric acid can be transformed by the alkalies of serum into urate of soda, but how much sodium urate the serum is able to take up without precipitation. Roberts in 1890 carried out experiments and, by the use of a correct method, found that serum holds only a few milli- grammes of urate of soda in:solution. The figures found by Gudzent twenty years later were almost identical with those of Roberts, and from them Gudzent has calculated that theoret- ically 100 grammes of serum are saturated when containing 8 mg. of urate of soda. If we apply these figures to the free acid instead of to urate of soda, and to blood instead of to 264 HARVEY SOCIETY serum, we are surprised to see that 100 grammes of blood are saturated when they contain 4 milligrammes of uric acid. Such an amount has indeed been found in the blood of people suffer- ing from severe gout when on a mixed diet. At the point of saturation the conditions for precipitation are nearly always present. In the circulating blood, however, precipitation never takes place; it is prevented by its contin- uous movement and by the perpetual exchange of the urate molecules. While some of them pass to the urine others enter the blood from the organs. PRECIPITATION OF URIC SODA The conditions for precipitation are much more favorable within the lymphatics and within the synovia of the articular cavities than elsewhere, for here the flow of liquids is very slow. I myself found four and six milligrammes of uric acid in 100 e.c. of the fluids aspirated from inflamed knee-joints. But in spite of this saturated state one rarely finds crystals in sus- pension, and when such are encountered they may perhaps have originated from the impregnated cartilages which had undergone destruction, or from tophi which had forced a way into the articular cavity. As a rule, crystallization takes place only in the organized tissue. By many it has been taken for granted that the cartilages possess a specific power to attract dissolved urate of soda and to bring it to crystallization. Moreover, the phenomenon can also be produced by excised cartilaginous tissue. This was shown by Roberts many years ago, and later on by Almagia and Brugsch. Tarsal bones of a pig were suspended in phials charged with a saturated solution of urates; after a few days they were incrusted with needles and presented an aspect which, in intensity and distribution of deposits, resembled that of a gouty joint. The process of precipitation is thus not a vital one, but is purely passive, brought on by certain chemical rela- tions and properties still existing in the dead cartilaginous tissue. But possibly even this chemical reaction is not of a complicated order. Roberts in a purely experimental way URIC ACID IN GOUT 265 e showed that the solubility of sodium urate decreases with increasing concentration of soda salts in the liquid. The theory of solution agreed fully with the actual experiment. An addi- tion of 0.2 per cent. sodium chloride to a saturated solution of urate of soda brought about precipitation. Roberts and Gudzent suggested that cartilage, tendons, ete., are richer in sodium salts than the blood serum. Their sodium content reckoned as sodium chloride, rises to 0.9 per cent., while in serum it is only 0.7 per cent. When the lymph saturated with urate of soda penetrates into the cartilaginous tissue where the concentration of the sodium ions is higher, a state of super- saturation is brought about, and precipitation can take place. In this hypothesis, the supposition is made that the sodium exists in the state of ions in the cartilages. Whether this is true or not is unknown. If they do not, if they be united to an organic compound, the explanation of Roberts and Gudzent would not hold good. In response to the question why precipitation does not occur in all joints,—their chemical composition being the same every- where,—we must refer to mechanical and thermal injuries, the influence of which on the localization is clearly proved by daily experience, even though we do not find such explanation entirely satisfactory. But the nephritic patient is exposed to the same influences. Why does not precipitation occur in every man suffering from contracting kidney? For the second time we are obliged to put this question. If the retention of uric acid in the blood is a passive process, if the deposition within the tissues is a purely physical process, by what properties is gout then to be char- acterized ? There must be an active element in its pathogenesis belong- ing exclusively to gouty diathesis. What then is this active principle? Science as yet gives no answer to this question, which touches the central point of the problem. Therefore I will not attempt to present any hypothesis. Surely a heredi- tary influence exists in gout, and a neuropathological influence is present as well as the chemical and the humoral-pathologieal. 266 HARVEY SOCIETY The clinician must always be aware of this. But as long as we cannot determine the mechanism by which the nerves act on the metabolism in gout, no advantage results from emphasizing such an influence. All organs yielding or destroying uric acid (and they are many) were at one time thought to be implicated in the phe- nomena of gout. Thus a predominant part was formerly attributed to the liver, to the bowels, or to the spleen, ete. But their influence, if indeed any exists, is not a controlling one over the uric acid metabolism. It is for the present worth- less to search for increased formation or for decreased destruc- tion in the single organs, since we know that as a total the uric acid metabolism always remains within the ordinary limits. If muscular work prevents paroxysms or lessens their violence, this need not be referred to an increased destruction of uric acid, but rather to an immediate action on the joints, changing the quantity of synovial liquid or its composition, or the rapidity of its circulation. The action of other organs too might be such an indirect one. Thus the active principle, so much sought for, which causes gout remains a mystery, not to be unravelled by our present means. Leaving the solution of the problem to the future, we return to the fate of urate deposits. We can discuss the essen- tials of this question without regard to the problems of gouty diathesis and pathogenesis. RELATION BETWEEN THE ATTACKS AND THE URATE DEPOSITS At what time does precipitation take place? Is it a chronic process going on continuously, or does it depend upon attacks and in what manner is it connected with them? When con- sidering those tophi whose fate we can trace with the eye, we know that they may appear, grow, and disappear without the bearer knowing anything about it. Especially the largest deposits, those upon the hands, grow with hardly any inflam- matory reaction. But this fact does not absolutely exclude a relation between the precipitating process and the attack. The URIC ACID IN GOUT 267 same process which on slow development causes no strong reaction may induce the most violent symptoms if the onset is one of great intensity. According to Garrod the attack is provoked by a sudden precipitation of crystals in the tissues. The attack consists merely of such deposition of urates and of the inflammation induced by their presence. The deposits, once formed, remain for a lengthened time, often throughout life. ‘‘ When erys- tallization of the salt takes place in any tissue, inflammation is suddenly lit up by its presence and a paroxysm of gout ensues.’’ Roberts is of the same opinion. There is one objection, which renders it difficult to consent in this respect to the authority of the eminent English clinician. One of the chief rules of chemistry, corpora non agunt nisi soluta, is applicable also to the conditions in the organism. If the uric acid salt is precipitated, the chemical influence of the solid material upon the tissues stops instantly and does not reappear until it is dissolved again. In the meantime its action is purely physical. The mechanical irritation may perhaps be stronger at the moment of erystallization and during the growth of the crystals than in periods of absolute rest. But the pressure which the crystals exert upon the tissues is scarcely sufficient to bring about serious reaction. Garrod’s opinion is just as improbable to me as the hypothesis which refers the symptoms of bronchial asthma to an irritation of the air- conduits by Charcot-Leyden crystals. In opposition to the English authors, I am rather inclined to ascribe the irritation to the dissolved urates. According to this conception a sudden increase of concentration will induce the inflammation. This supposed rise of the amount of urates within the synovia could be brought about in a twofold manner. Accord- ing to Garrod’s opinion, it is only a part of the general urate accumulation in the whole body, induced by a deficient excre- tion. The urate molecules would pass from the serum to the synovia and hence to the cartilages. However a contrary con- ception could also be supported. A rapid solution of the deposits might lead to an increased concentration in the syno- 268 HARVEY SOCIETY vial liquid, thus explaining the violence of reaction perhaps better than the hypothesis of Garrod. As to the question why a solution of deposits should occur so rapidly, I own I am not able to reply. The above conception is an elaboration of Pfeiffer’s ideas somewhat changed. In opposition to Garrod Pfeiffer con- sidered the attack as consisting of a solution and removal of the deposits of urate. He based his opinion upon reasons which do not seem to me to be justifiable. Still his idea as a whole seems to me probable and for the following two reasons: 1. The inflammation spreads far beyond the places of the crystalline deposits. One may find an aspect of the skin fully resembling a phlegmon, and even a severe lymphangitis, which though aseptic has sometimes misled the surgeon. This proves that an irritating substance is carried away from the afflicted joints. If the sodium urate is really the inflammatory agent in the joints, it seems likely that the irritation of the more distant tissues is caused by the same substance, carried away through the lymphatics. 2. The second reason in favor of Pfeiffer’s views is the excess of endogenous uric acid excreted during the attack. ' The excess is far greater than the diminution preceding the paroxysm. In one of my observations the surplus eliminated in eight days of a violent attack amounted to more than 3 grammes. Although the origin of these urates cannot be pointed out with accuracy, it may probably be attributed in whole or in part to a solution of the crystalline deposits. I might also lay stress upon the fact that in aspirating the joint exudate as completely as possible, the last portions, origi- nating from the interior parts of the cavity, are always found to be richer in leucocytes than the first portions. And it is well known that leucocytes are the instruments for attacking the urate crystals as well as the vehicles for their removal. Perhaps the theory of Garrod and that of Pfeiffer might be combined in such a way as to assume that a general retention of urates provokes the outbreak of an attack, and that the inflammation which follows leads to a removal of the deposits. URIC ACID IN. GOUT 269 PECULIAR EXCITING CAUSES FOR THE ATTACK The accumulation of uric acid in the fluids of the body supposed by Garrod to precede and to provoke the attack has not yet been demonstrated by analysis of the blood itself. If it were possible to determine the uric acid in the blood day by day before and during the paroxysm, we would be able to judge the part played by the accumulation of urates better than we can now. We are now obliged to rely upon analysis of the urine. The decrease of the uric acid output in the twenty-four hours prior to the attack, which amounts to from one hundred to two hundred milligrammes, may be interpreted as meaning an accumulation in the body; but sometimes this is altogether lacking. The best support of this part of Garrod’s theory is to be found in the more recent observations, that consumption of sweetbread sometimes brings about the paroxysm. It may be asked whether this coincidence is not merely accidental. In spite of such doubts and gaps, this part of Garrod’s doctrine is generally acknowledged and accepted. However, this is not true as concerns the second part of his theory. Garrod assumed a decreased alkalinity of the blood to be the second cause exciting the gouty attack. This part of his doctrine should be dropped entirely. It is refuted by the actual examination of the gouty blood, as well as by the proofs in the test-tube, and it is contradicted also by the doctrine of physical chemistry. In measuring the alkalinity of the blood by means of titration in more than twelve patients before, during, and after the paroxysm and also during the intervals, I failed at any time to find any marked difference. Moreover, all chemical analogies teach us that an addition of hydrochloric acid to the solution of any salt may bring about precipitation of the free acid, but never that of the salt. If we had to deal with bisodium urate, the addition of an acid would produce its conversion into monosodium urate, and this, being less soluble than the original salt, would erystallize out. This would be the same process as the precipitation of monosodium sulphate produced by adding sulphuric acid to a solution of neutral 270 HARVEY SOCIETY sodium sulphate. But please remember, gentlemen, that in the serum only monosodium urate is present, which, according to the above exposition, is not precipitated by acids. Roberts’s test-tube experiments correspond fully with these theoretical deductions. Serum contains about 0.5 per cent. of sodium chloride and 0.2 per cent. of monosodium carbonate. The addition of hydrochloric or acetic acid to serum saturated with urate of soda brings about a precipitation of uric acid. If, on the other hand, sodium carbonate be removed from the serum by dialysis, not the slightest crystallization occurs, although the serum has lost its alkalinity. On the other hand, addition of monosodium carbonate to the serum causes precipi- tation, although this be contrary to Garrod’s opinion, and in spite of the high increase of the so-called alkalinity. The solu- bility and the precipitation of urate of soda have nothing to do at all with the alkalinity, but depend merely on the concen- tration of sodium ions or salts in the solution. Indeed, addi- tion of sodium chloride effects a precipitation just as well as does monosodium carbonate. The question of diminishing alkalinity is of the utmost importance for the precipitation of uric acid in urine, it is pre- dominant in the pathogenesis of gravel, but it must be excluded entirely from the theory of gout. It is replaced by the factor of salt concentration, and especially of the concentration in soda ions. Please remember, gentlemen, in connection with these explanations of the conception of Roberts, that it is the high amount of soda salts in the connective tissue which very likely determines the place of the erystallization of urates. It will be hard for many to drop entirely the doctrine of the importance of a varying alkalinity of the blood in gout. You may feel inclined perhaps to found your conservatism on the valuable experiments of Loghem. This Dutch scientist had found that deposits of uric acid, produced by hypodermic injections in the rabbit, are soon converted into deposits of urate of soda, and that this transformation is followed by a serious inflammation. In dogs this transformation is not observed nor is there any inflammation. The metabolism of the URIC ACID IN GOUT 271 rabbit kept on green food is, roughly speaking, an alkaline one. In the same sense, we might apply the term of an acid metab- olism to a dog fed on meat. According to this difference the fate of uric acid injected varies in the two species. But if the dog receives 20 grammes of sodium carbonate, his metabolism becomes more alkaline, and the conversion of the uric acid into urate of soda takes place exactly as in the rabbit. Vice versa, in the rabbit it is prevented by administering hydro- ehloric acid. The results of the older experiments of Pfeiffer and the later ones of Silbergleit coincide fully with those of Loghem. In these experiments the influence of acids and alkalies upon the chemical occurrences in the body are evident. In many respects these experiments are of high value and they have justly attracted attention. However, the observations have nothing to do with the conditions in gout. In Loghem’s experi- ments an entirely different problem was attacked, namely, the transformation of uric acid into urate of soda. It is obvious that the transformation is favored by an excess of alkalies in the body, as well as in the test-tube, and that it is retarded or prevented by an increase of acid substances alike in the body and in the chemist’s phial. As to the solubility and precipi- tation of urate of soda, Loghem’s experiments prove just as little as Klemperer’s experiments on the degree of saturation of serum with urate of soda. THERAPEUTICS Gentlemen, if I now consider the therapeutics of gout, I must of course confine myself to the discussion of such points as bear relation to the metabolism of uric acid. The two drugs which are the most efficacious in the paroxysm, colchicum and salicylate, behave entirely differently as regards the elimination of uric acid. Salicylates in sufficient doses increase the output of uric acid materially, by half a gramme and more daily. This effect, however, seems to dis- appear within a few days. Colchicum, which, no doubt, is the stronger remedy, produces no change, or if so, it diminishes the 272 HARVEY SOCIETY quantity of uric acid elimination. I do not believe that these two drugs, the chemical effects of which are so different, owe their efficacy to the same pharmacological action. One should try to define the mechanism of their action. Setting aside their soothing capacity I shall discuss only their action upon the uric acid within the body. If, according to Pfeiffer’s opinion, the attack were a process of solution and removal of urates, one could conceive that salicylates favor the elimination, since increased uric acid output follows the administration of these substances. But we do not know whether this increased output originates from the deposits. Even healthy persons when taking salicylates show a slight increase of uric acid elimination. Comparative experiments with this remedy ought to be made on a large scale upon healthy and gouty people. As to colehicum, one might think that it inhibits the process of solution, thus putting an end to the inflammation and to the paroxysm. Its effect is certainly more rapid and more intense than that of sodium salicylate. If this conception were right the colchicum would be only a palliative remedy for the attack, while the salicylates in spite of their slower action would be the preferable remedy. It is well known that colchicum is useless in chronic gout. If the attack be interpreted in the sense of Garrod, the efficacy of both drugs would have to be explained otherwise.