UNIVERSITY OF CALIFORNIA MEDICAL CENTER LIBRARY SAN FRANCISCO FROM THE LIBRARY OF ALBION W. HEWLETT, M.D. PHYSIOLOGY AND BIOCHEMISTRY IN MODERN MEDICINE BY J. J. II. MACLEOD, M.B. PROFESSOR OF PHYSIOLOGY IN THE UNIVERSITY OF TORONTO, TORONTO, CANADA; FORMERLY PROFESSOR OF PHYSIOLOGY IN THE WESTERN RESERVE UNIVERSITY, CLEVELAND, OHIO ASSISTED BY ROY G. PEARCE, B.A., M.D. Director of the Cardiorespiratory Laboratory of Lakeside Hospital, Cleveland, Ohio AND BY OTHERS WITH 233 ILLUSTRATIONS, INCLUDING 11 PLATES IN COLORS ST. LOUIS C. V. MOSBY COMPANY 1918 COPYRIGHT, 1918, BY C. V. MOSBY COMPANY Press of C. V. Mosby Company St. Louis -Ml 6 TO M. W. M. PREFACE The necessity of allotting the various subjects of the medical curric- ulum to different periods, so that the more strictly scientific subjects are completed in the earlier years, has the great disadvantage that the student, being no longer in touch with laboratory work, fails to employ the scientific knowledge with full advantage in the solution of his clin- ical problems. He is apt to regard the first two or three years in the laboratory departments as inconsequential in comparison with the sup- posedly more practical instruction offered during the subsequent clinical years. He is taught by his laboratory instructors to observe accurately, and to correlate the observed facts, so that he may be enabled to draw conclusions as to the manner of working of the various functions of the animal body in health, and before proceeding to his clinical studies, he is required to show a proficiency in scientific knowledge, because it is recognized that this must serve as the basis upon which his knowledge of disease is to be built. When the clinic is reached, however, the meth- ods of the scientist are not infrequently cast aside and an understanding of disease is sought for largely by the empirical method ; namely, by the endeavor to see and examine innumerable patients, to diagnose the case according to the grouping of the signs and symptoms, and to treat it by the prescribed methods of experience. So much has to be learned and so much has to be seen during the clinical years, that the student gives little thought to the natui'e of the functional disturbance which is responsible for the symptoms; he fails to realize that after all, there is no essen- tial difference between the condition brought about in his patient by some pathologic lesion, and that which may be produced in the labora- tory by experimental procedures, by drugs or by toxins. It must of course be recognized that just as the science of medicine originated by the grouping of symptoms into more or less characteristic diseases for which the most favorable method of treatment had to be discovered by experience, so must a certain part of the medical training be more or less empirical but it should at the same time be realized that such a method is only a means to an end, and that the real understanding of disease can be acquired only when every abnormal condition is inter- preted as a primary or secondary consequence of some perverted bodily function, and when the training in observation and the inductive method is carried from the laboratory into the clinic. VI PREFACE It is a constant experience of clinical instructors who would employ scientific methods of instruction, that they find the students not only indifferent to an analysis of their cases from the functional standpoint, but also that they are too inadequately, prepared in fundamental phys- iologic knowledge, to make the analysis possible. The student may have a superficial acquaintance with the main facts of physiologic science but have failed to acquire the enquiring habit of mind which will en- able him, through reflection, comparison, and personal research, to ap- ply the knowledge in practical, medicine and surgery. For this lack of correlation between the laboratory and clinical stud- ies, the clinical instructors are not alone responsible. The laboratory courses are frequently given without any attempt being made to show the student the bearing of the subject in the interpretation of disease, or to train him so that in his later years he may be able to adapt the methods of investigation which he learned in the laboratory, to the study of morbid conditions. It is self-evident that (without any knowledge of disease) the extent to which the student in the earlier years of the course could be expected to appreciate the clinical significance of what he learns in the laboratory is limited, but this should not deter the in- structor from indicating whenever he can, the general application of scientific knowledge in the interpretation of diseased conditions. But the chief remedy of the evil undoubtedly lies partly in the continuance of certain of the laboratory courses into the clinical years, and partly in the study of medical literature in which the application of physiology and biochemistry in the practice of medicine is emphasized. Notwithstanding the sufficient number of excellent textbooks in phys- iology available to the medical student, there is none in which partic- ular emphasis is laid upon the application of the subject in the routine practice of medicine. In the present volume the attempt is made to meet such a want, by reviewing those portions of physiology and bio- chemistry which experience has shown to be of especial value to the clinical investigator. The work is not intended to be a substitute, either for the regular textbooks in physiology, or for those in functional pathology. It is supplementary to such volumes. It does not start like the modern test in functional pathology, with a consideration of the diseased condition, and then proceed to analyze the possible causes and consequences of the disturbances of function which this exhibits; but it deals with the present-day knowledge of human physiology in so far as this can be used in a general way to advance the understanding of disease. In a sense it is therefore an advanced text in physiology for those about to enter upon their clinical instruction, and at the same PREFACE Vll time, a review for those of a maturer clinical experience who may desire to seek the physiological interpretation of diseased conditions. In attempting to fulfil these requirements, it has been deemed essen- tial to go back to the fundamentals of the subject, and to explain as simply as possible the physical and physicochemical principles upon which so large a part of physiological knowledge depends. Physiology may be considered as an application of the known laws and facts of physics and chemistry to explain the functions of living matter, and it is only after the extent to which this application can be made has been appreciated, that the knowledge may be used to serve as the foundation upon which a superstructure of clinical knowledge can be built. In order that the volume might be maintained of reasonable size, it has been necessary to select certain parts of the subject for particular emphasis, the basis of selection being the degree to which our knowledge clearly shows the value of the application of physiological methods both of observation and of thought in the study of diseased conditions. This has not been done to the extent of omitting the apparently less essential parts, for these have been treated in sufficient detail to link the others together so as to preserve a logical continuity, and show the bearing of one field of knowledge on another. There are however certain parts of the science, particularly the physiology of nerve and muscle, of the special senses, and of reproduction, for which application in the general fields of medicine and surgery is limited, and these parts have been omitted entirely. It has been judged that this perhaps somewhat arbi- trary selection is justified on the ground that the ordinary text in physiology covers these subjects sufficiently, except for the specialist, for whom on the other hand, no adequate review would have been pos- sible within the limits of such a volume as this. With reference to bio- chemistry, no attempt is made to review the properties or describe the characteristic tests of the various chemical ingredients of the body tis- sues and fluids. This is already sufficiently done in the textbooks on biochemistry, and in the numerous manuals on clinical methods. Bio- chemical knowledge is treated rather from the physiologist's stand- point, as an integral part of his subject, particular attention, neverthe- less, being paid to the far-reaching applications, of this latest depart- ment of medical science, in the elucidation of many obscure problems of clinical medicine, such as those of diabetes, nephritis, acidosis, goiter and myxedema. To make the volume of value to those who may not have had time or opportunity to familiarize themselves with the techni- cal methods of the physiologist and biochemist as used in the modern clinic, a certain amount of space is devoted to a brief description of the methods that appear at present to be receiving most attention, and to 1)0 of irreatest value. Vlll PREFACE Finally, it should be mentioned that the principles of serum diagnosis and therapy are omitted, since these belong to a highly specialized science requiring an intensive training of its own. In the hope that the volume may be instrumental in arousing sufficient interest to stimulate a more intensive study of the various subjects which it introduces, a brief bi-bliography is given at the end of each section. The references selected are to papers that are more partic- ularly knoAvn to the author; they are not necessarily the most impor- tant publications on the subject, but are often chosen because of the useful reviews of previous Avork contained in them, rather than because of their own originality. Some of the papers, however, are referred to as authority for statements of fact which may arouse in the reader a desire to ponder for himself the evidence upon which these are based. The references are usually divided into two groups, "monographs" and "original papers," and it is only occasionally that specific reference is made to the former in the context. The original papers, on the other hand, are referred to by numbers. With the general field of the subject so well covered by such excellent textbooks as Bayliss' "Principles of General Physiology," Stewart's, HowelPs, Starling's, and Halliburton 's "Human Physiologies," and Leonard Hill's "Recent and Further Ad- vances in Physiology," the author has felt free to pick and choose from the monographs and original papers, topics that are ordinarily passed over cursorily in the textbook, and when this has been done, the refer- ences are somewhat more extensive. Such is the case for example in the chapters relating to the chemistry of respiration, to the metabolism of carbohydrates and fats, to the problems of dietetics and growth, to the physicochemical basis of neutrality regulation in the animal body, and to the action of enzymes. Acknowledgment is gratefully made for the assistance and advice in the preparation of the book, particularly to Doctor R. G. Pearce, for the contribution of several chapters, to which his name is attached, and for which he is entirely responsible ; and to Doctor E. P. Carter, whose criticisms, after patient perusal of the unfinished manuscript, were of inestimable value in its final revision. Acknowledgment is also made to Doctor R. W. Scott and Professor F. E. Lloyd, for valuable criticism and advice, and to the former for a chapter on the "Clinical Applica- tion of Electrocardiographs." To Miss Achsa Parker, M.A., the author owes a great debt of gratitude for the thorough and painstaking way in which she prepared the manuscript for the press, and for her never- tiring endeavors to have the spelling and punctuation in conformity with Webster's Dictionarj'. For assistance in the preparation of the index thanks are due to Miss Marian Armour and Mrs. MacFarlanc, PREFACE IX and for permission to use certain of the figures and illustrations, to the various authors and publishers who granted it. For the excellent man- agement and careful execution of the presswork, the author wishes to thank the publishers, whose courteous and friendly dealings have always made the work easier. J. J. R. MACLEOD. University of Toronto, Toronto, Canada. CONTENTS PART I THE PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES CHAPTER I PAGE GENERAL CONSIDERATIONS 1 The Laws of Solution, 3 ; Gas Laws, 3 ; Osmotic Pressure, 4 ; Biological Methods for Measuring Osmotic Pressure, 6; Hemolysis, 7; Plasmolysis, 8. CHAPTER II OSMOTIC PRESSURE (CONT'D) 10 Measurement by Depression of Freezing Point, 10; The Role of Osmosis Dif- fusion, and Allied Processes in Physiologic Mechanisms, 11. CHAPTER III ELECTRIC CONDUCTIVITY, DISSOCIATION, AND IONIZATION 16 Biological Applications, 19. CHAPTER IV THE PRINCIPLES INVOLVEI> IN THE DETERMINATION OF HYDROGEN-ION CONCENTRATION 22 Titrable Acidity and Alkalinity, 22; Actual Degree of Acidity or Alkalinity, 23; Mass Action, 23; Application to the Measurement of H-ion Concentration, 26; Application in Determining the Real Strength of Acids or Alkalies, 28. CHAPTER V THE PRINCIPLES INVOLVED IN THE MEASUREMENT OP HYDROGEN-ION' CONCENTRATION (CONT'D) 29 The Electric Method, 29 ; The Indicator Method, 32. CHAPTER VI REGULATION OF NEUTRALITY IN THE ANIMAL BODY AND ACIDOSIS 36 Buffer Substances, 36 ; Theory of Acidosis, 38 ; Measurement of the Reserve Alkalinity, 41 ; Titration Methods, 41 ; CO2-combining Power, 42 ; Indirect Methods, 46. CHAPTER VII COLLOIDS 50 Characteristic Properties, 50; Characteristics of True Colloidal Solutions, 51; Tyndall Phenomenon, 51; Relative Indiffusibility, 51; Electric Proper- ties, 55; Brownian Movement, 57; Osmotic Pressure, 57. Xll CONTENTS CHAPTER VIII PAGE COLLOIDS (COXT'D) GO Suspensoids and Emulsoids, GO; Gelatinization, 61; Imbibition, 62; Action of Electrolytes on Colloids, 63; Proteins as Colloids, 63; Surface Tension, 64; Adsorption, 65 ; Everyday Reactions Depending on Adsorption, 66 ; Conditions Influencing or Influenced by Adsorption, 67; Physiologic Processes Depending on Adsorption, 69. FERMENTS, OR ENZYMES ' 71 The Nature of Enzyme Action, 72 ; Properties of Enzymes, 73 ; Reversibility of Enzyme Action; 77; Specificity of Enzyme Action, 79; Peculiarities of Enzymes, 80; Types of Enzyme, 81; Enzyme Preparations, 82; Conditions for Enzymic Activity, 82 PART II THE CIRCULATING FLUIDS CHAPTER X BLOOD: ITS GENERAL PROPERTIES (Bv R. G. PEARCE) v . . 85 Quantity of Blood in the Body, 85; Water Content, 86; Proteins, 87; Fer- ments and Antiferments, 89. CHAPTER XI THE BLOOD CELLS (BY R. G. PEARCE) 91 Red Blood Corpuscles, or Erythrocytes, 91 ; Origin, 92 ; Rates of Regeneration, 93; Hemolysis, 95; Leucocytes, 96; Blood Platelets, 97. CHAPTER XII BLOOD CLOTTING 98 Visible Changes in the Blood During Clotting, 98; Methods of Retarding Clotting, 99; Nature of the Clotting Process, 101; Influence of Calcium Salts, 103; Influence of Tissues, 104. BLOOD CLOTTING (COXT'D) 106 Theories of Blood Clotting, 106; Intravascular Clotting, 107; Measurement of the Clotting Time, 108; Blood Clotting in Various Physiologic Conditions, 110; Blood Clotting in Disease, 111 ; Hemorrhagic Diseases, 112 ; Thrombus Forma- tion, 113. CHAPTER XIV LYMPH FORMATION AND CIRCULATION 115 General Considerations, 135; Experimental Investigations, 118; Edema, 120. CONTENTS xiii PART III CIRCULATION OF THE BLOOD CHAPTER XV PAGE BLOOD PRESSURE 122 The Mean Arterial Blood Pressure, 123; Mercury Manometer Tracings, 123; Spring Manometer Tracings, 12(5; Clinical Measurements, 128. CHAPTER XVI THE FACTORS CONCERNED IN MAINTAINING THE BLOOD PRESSURE 134 Pumping Action of the Heart, 134; Peripheral Resistance, 134; Amount of Blood in the Body, 135; Effects of Hemorrhage and Transfusion, 139; Viscos- ity of the Blood, 140; Elasticity of Vessel "Walls, 142. CHAPTER XVII THE ACTION OF THE HEART 144 The Pumping Action of the Heart, 144 ; Intracardiac Pressure Curves, 146 ; Comparison of the Curves, 148. CHAPTER XVIII THE PUMPING ACTION OF THE HEART (CONT'D) 151 Contour of the Intracardial Pressure Curves, 151 ; Ventricular Curve, 151 ; Auricular Curve, 153; The Mechanism of Opening and Closing of the Valves, 154; The Heart Sounds, 157: Causes of Sounds, 157; Records of Sounds (Electrophonograms) , 1 58. CHAPTER XIX THE NUTRITION OF THE HEART 161 B-lood Supply, 161; Perfusion of the Heart Outside the Body, 161; Resuscita- tion'of the Heart in Situ, 164; Relationship of the Chemical Composition of the Perfusion Fluid in Cold-blooded and Warm-blooded Hearts, 165. CHAPTER XX PHYSIOLOGY OF THE HEARTBEAT 170 Origin and Propagation of the Beat, 170; Myogenic Hypothesis, 171; Neuro- genic Hypothesis, 172; The Pacemaker of the Heart and Heart-block, 174; Physiologic Characteristics of Cardiac Muscle, 176. CHAPTER XXI PHYSIOLOGY OF THE HEARTBEAT (CONT'D) 182 Origin and Propagation of the Beat in the Mammalian Heart, 182 ; Conduct- ing Tissue in the Mammalian Heart, 182; Site of Origin of Beat, 187. CHAPTER XXII PHYSIOLOGY OF THE HEARTBEAT (CONT'D) 101 Alode of Propagation of the Beat in the Auricles ami from the Auricles to the Ventricles, 191 ; Spread of Beat in the Ventricle, 193 ; Fibrillation of the Ven- tricles ami Auricles, 195. XIV CONTENTS CHAPTER XXIII PAGE THE BLOODFLOW IN THE ARTERIES 198 The Pulses, 198; General Characteristics, 198; Rate of Transmission of Pulse Waves, 198; Contour of the Pulse Curve, 200; Velocity Pulse, 200; Palpable Pulse, 202; Analysis of the Curve, 202; The Dicrotic Wave, 203; Causes of Disappearance of the Pulse in the Veins, 205. CHAPTER XXIV RATE OF MOVEMENT OF THE BLOOD IN THE BLOOD VESSELS 206 Velocity of Flow, 206; Mass Movement of the Blood, 208; The Visceral Blood- flow in Man, 212; Work of the Heart, 212; Circulation Time, 213; Movement of Blood in the Veins, 214. CHAPTER XXV THE CONTROL OF THE CIRCULATION 216 Nerve Control, 217; Vagus Control in the Cold-blooded and the Mammalian Heart, 217; Tonic Vagus Action, 221; Afferent Vagus Impulses, 222; Mechan- ism of Vagus, 224 ; Termination of the Vagus Fibers in the Heart, 22o ; Sym- pathetic Control, 227. CHAPTER XXVI THE CONTROL OF THE CIRCULATION (CONT'D) . 229 Nerve Control of Peripheral Resistance, 229; Detection of Vasomotor Fibers in Nerves, 231; Origin of Vasomotor Nerve Fibers, 232; Vasomotor Nerve Centers, 235 ; Independent Tonicity of Blood Vessels, 236. CHAPTER XXVII THE CONTROL OF THE CIRCULATION (CONT'D) 237 Control of the Vasomotor Center, 237; Hormone Control, 237; Nerve Control, 238; Pressor and Depressor Impulses, 239; Reciprocal Innervation of Vascular Areas, 243; Influence of Gravity on the Circulation, 244. CHAPTER XXVIII PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA . . . 247 Circulation in the Brain, 247; Anatomical Peculiarities, 247; Physical Condi- tions of Circulation, 249; Vasomotor Nerves, 252; Intracranial Pressure, 253; Circulation through the Lungs, 253; Circulation through the Liver, 255; The Coronary Circulation, 257. CHAPTER XXIX CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS 259 Electrocardiograms, 259; The Ventricular Complex, 262. • CHAPTER XXX CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (CONT'D) 266 Electrocardiograms of the More Usual Forms of Cardiac Irregularities, 266; Sinus Arrhythmia, 266 ; Sinus Bradycardia, 266 ; The Extrasystole, 266 ; Parox- ysmal Tachycardia, 269; Auricular Fibrillation, 269; Auricular Flutter, 269; Heart-block, 270. CONTENTS XV CHAPTER XXXI PAGE CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (CONT'D) 273 Polysphygmograms, 273; Venous Pulse Tracings, 273; Simultaneous Arterial Pulse Tracings, 276; Abnormal Pulses, 276. CHAPTER XXXII CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (CONT'D) 281 Measurement of the Mass Movement of the Blood, 281; The Normal Flow, 282; Clinical Conditions Which Affect the Blood flow, 283. CHAPTER XXXIII 'SHOCK 287 Gravity Shock, 287; Hemorrhage Shock, 288; Anesthetic Shock, 288; Spinal Shock, 288; Nervous Shock, 289; Surgical Shock, 289; Experimental In- vestigation of Shock, 289; Treatment, 295; Cause of Secondary Symptoms, 295. CHAPTER XXXIV RESPIRATION 299 The Mechanics of Respiration, 299; Pressure and Amount of Air in the Lungs, 299; Respiratory Tracings, 303; The Intrapleural Pressure, 304; Influence on Blood Pressure, 306. CHAPTER XXXV THE MECHANICS OF RESPIRATION (CONT'D) (BY R. G. PEARCE) 310 Variations in Dead Space, Residual Air and the Mid- and Vital Capacities in , Various Physiologic and Pathologic Conditions, 310. CHAPTER XXXVI THE MECHANICS OF RESPIRATION (CONT'D) (BY R. G. PEARCE) 315 The Mechanism of the Changes in Capacity of the Thorax and Lungs, 315; The Movements of the Ribs, 315; The Action of the Musculature of the Ribs, 319; The Action of the Diaphragm, 320; The Effects of the Respiratory Move- ments on the Lungs, 325. CHAPTER XXXVII THE CONTROL OF RESPIRATION 327 The Respiratory Centers, 327; Reflex Control of the Respiratory Center, 331. CHAPTER XXXVIII THE CONTROL OF RESPIRATION (CONT'D) 335 Hormone Control of the Respiratory Center, 335; Tension of CO., and O2 in Arterial Blood, 337; Tension of CO., and O2 in Alveolar Air, 339; Tension of CO, in Venous Blood, 342. XVI CONTEXTS CHAPTER XXXIX PAGE THE CONTROL OF RESPIRATION (CONT'D) (By R. G. PEAJICE) 344 Estimation of the Alveolar Gases, 344; Method for Normal Subjects, 345; Clinical Method, 347. CHAPTER XL THE CONTROL OF RESPIRATION' (CONT'D) 349 The Nature of the Respiratory Hormone, 349 ; Relationship between CO, of Inspired Air and Pulmonary Ventilation, 350; Possibility that CO., Specifically Stimulates the Center, 352; Relationship among Acidosis, Alveolar CO, and Respiratory Activity, 354. CHAPTER XLI THE CONTROL OF RESPIRATION (CONT'D) 356 The Constancy of the Alveolar CO, Tension under Normal Conditions) 256; Sensitivity of the Center to Changes in the CO, Tension of the Alveolar Air, 357; Alveolar CO2 Tension during Breathing in a Confined Space, 357, in Rarefied Air, 360, and in Apnea, 362. CHAPTER XLII THE CONTROL OF RESPIRATION (CONT'D) 366 The Effect of Muscular Exercise on the Respiration, 356. CHAPTER XLIII THE CONTROL OF RESPIRATION (CONT'D) 371 Periodic Breathing, 371; Types of Periodic Breathing, 371; Causes of Periodic Breathing, 372. CHAPTER XLIV RESPIRATION BEYOND THE LUNGS 378 Transportation of Gases by the Blood, 379; Transportation of Oxygen, 379; Dissociation Curve, 383; Difference between Curves of Blood and Hemoglobin Solution, 383; Rate of Dissociation, 386; Dissociation Constant, 388. CHAPTER XLV RESPIRATION BEYOND THE LUNGS (CONT'D) 390 Means by Which the Blood Carries the Gases, 390; Oxygen Requirement of the Tissues, 393; Mechanism by Which the Demands of the Tissues for Oxy- gen Are Met, 397. CHAPTER XL VI THE PHYSIOLOGY OF BREATHING IN COMPRESSED AIR AND IN RAREFIED AIR . . . 399 Mountain Sickness, 399; Compressed Air Sickness (Caisson Disease), 402; Practical Application in Treatment, 406. CHAPTER XLVII THE CIRCULATORY AND RESPIRATORY CHANGES ACCOMPANYING MUSCULAR EXERCISE 410 Mechanical Factor, 410; Nervous Factor, 412; Hormone Factor, 413. CONTENTS XVii CHAPTER XLVIII PAGE GENERAL PHYSIOLOGY OF THE DIGESTIVE GLANDS 418 Microscopic Changes during Activity* 418; Mechanism of Secretion, 420; Other Changes during Activity, 421 ; Control of Glandular Activity, 422 ; Nervous Control, 423. PHYSIOLOGY OF THE DIGESTIVE GLANDS (CONT'D) 425 Hormone Control, 425; Nervous Control of the Pancreas, 427. CHAPTER L PHYSIOLOGY OF THE DIGESTIVE GLANDS (CONT'D) 430 Normal Conditions of Secretion, 430 ; Normal Secretion of Saliva, 431 ; Secre- tion of Gastric Juice, 432; The Intestinal Secretions, 441. CHAPTER LI THE MECHANISMS OF DIGESTION 444 Mastication, 444; Deglutition, 445; The Cardiac Sphincter, 448; Vomiting, 449. CHAPTER LII THE MECHANISMS OF DIGESTION (CONT'D) 451 Movements of the Stomach, 451 ; Character of the Movements, 451 ; Effect of the Stomach Movements on the Food, 454; Emptying of the Stomach, 456 ; Control of the Pyloric Sphincter, 456 ; Rate of Emptying of the Stomach, 458; Influence of Pathologic Conditions on the Emptying, 450; Gastroenter- ostomy, 461. CHAPTER LIII THE MECHANISMS OF DIGESTION (CONT'D) . 463 Movements of the Intestines, 463 ; Movements of the Small Intestine, 463 ; Movements of the Large Intestine, 468; Effect of Clinical Conditions on the Movements, 470. CHAPTER LIV HUNGER AND APPETITE 471 Hunger Contractions of Stomach, 471; Remote Effects of .Hunger Contrac- tions, 474 ; Hunger during Starvation, 475 ; Control of the Hunger Mechanism, 476. CHAPTER LV BIOCHEMICAL PROCESSES OF DIGESTION 481 Digestion in the Stomach, 481; Functions of the Hydrochloric Acid, 482; Amount and Source of the Acid, 482; Action of Pepsin, 485; Clotting of Milk in the Stomach, 488. XV111 CONTENTS CHAPTER LVI PAGE BIOCHEMICAL PROCESSES OF DIGESTION (CONT'D) 489 Digestion in the Intestines, 489 ; Pancreatic Digestion, 489 ; The Bile, 492 ; Chemistry of Bile, 494. CHAPTER LVII BACTERIAL DIGESTION IN THE INTESTINE 499 Bacterial Digestion of Protein, 501; Botulism, 503. PART VI THE EXCRETION OF URINE CHAPTER LVIII THE EXCRETION OF URINE (BY R. G. PEARCE) 507 Structure of Kidney, 507; Mechanism of the Excretion of Urine, 510; Theories of Renal Function, 511; Diuretics, 518; Albuminuria, 519; Influence of the Nervous System on the Secretion of Urine, 519. • CHAPTER LIX THE AMOUNT, COMPOSITION AND CHARACTER OF TJIIE URINE (BY R. G. PEARCE) . 521 Amount, 522; Specific Gravity, 522; Depression of Freezing Point, 523; Re- action, 524; Solid Constituents, 525. PART VII METABOLISM CHAPTER LX METABOLISM 534 Energy Balance, 535 ; Methods for Measuring Energy Output, 536 ; Normal Values, 538 ; Influence of Age and Sex, 541 ; Influence of Diseases, 542 ; The Material Balance of the Body, 543; Methods for Measuring Output, 543; Calculation of the Results, 544. CHAPTER LXI THE CARBON BALANCE 547 Respiratory Quotient, 547; Influence of Diet, 547; Influence of Metabolism, 549; Magnitude of the Respiratory Exchange, 550; Influence of Body Tem- perature, 551. CHAPTER LXII A CLINICAL METHOD FOR DETERMINING THE RESPIRATORY EXCHANGE IN MAN (BY R. G. PEARCE) 554 The Valves, 555; Tissot Spirometers, 556; Douglas Bag, 558; Haldane Gas- analysis Apparatus, 559; Calculations, 562. CONTENTS XIX CHAPTER LXIII PAGE STARVATION* 566 Excretion of Nitrogen, 566 ; Energy Output, 568 ; Nitrogenous Metabolites, 568 ; Excretion of Purines, 569; Excretion of Sulphur, 569; Normal Metabolism, 570; Nitrogenous Equilibrium, 571; Protein Sparers, 571. CHAPTER LXIV NUTRITION* AND GROWTH 574 The Food Factor of Growth, 574; Relationship of Proteins to Growth and Maintenance of Life, 574. CHAPTER LXV NUTRITION AND GROWTH (CONT'D) 583 Relationship of Carbohydrates and Fats to Growth, 583; Accessory Food Factors, or Vitamines, 584; Relationship of Inorganic Salts, 586. CHAPTER LXVI DIETETICS 588 Calorie Requirements, 588; The Protein Requirement, 590; Accessory Food Factors, 593; Digestibility and Palatability, 593. CHAPTER LXVII THE METABOLISM OF PROTEIN 595 Introductory, 595; Chemistry of Protein and of the Amino Acids, 597. CHAPTER LXVIII THE METABOLISM OF PROTEJN (CONT'D) 606 Amino Acids in the Blood and Tissues, 606; Fate of the Amino Acids, 610. CHAPTER LXIX THE METABOLISM OF PROTEIN (CONT'D) 613 End Products of Protein Metabolism, 613; Urea and Ammonia, 615; In- fluence of Acidosis on Ammonia-urea Ratio, 616 ; Influence of Liver on Am- monia-urea Ratio, 617; Perfusion of Organs, 618; Clinical Observations, 620. CHAPTER LXX THE METABOLISM OF PROTEIN (CONT'D) 622 Creatine and Creatinine, 622; Essential Chemical Facts, 622; Metabolism, 624; Influence of Food, Age, and Sex, 624; Origin of Creatine and Creatinine, 626. CHAPTER LXXI THE METABOLISM OF PROTEIN (CONT'D) 629 Undetermined Nitrogen and Detoxication Compounds, 629; Ethereal Sulphates and Glycuronates, 632. CHAPTER LXXII URIC ACID AND THE PURINE BODIES 634 Chemical Nature of the Purines, 634 ; Chemical Nature of the Substances Containing Purine and Pyrimidine Bases, 637; History of Nucleic Acid in the Animal Body, 638; Balance between Intake and Output o,f Purine Substances under Various Physiologic and Pathologic Conditions, 641. XX CONTENTS CHAPTER LXXIII PACE URIC ACID AND THE PURINE BODIES (CONT'D) 643 Source of Endogenous Purines, 643 ; Influence of Various Physiologic Con- ditions, of Drugs, and of Disease on the Endogenous Uric-acid Excretion, 647; Uric Acid of Blood, 648. CHAPTER LXXIV METABOLISM OF THE CARBOHYDRATES 652 Capacity of the Body to Assimilate Carbohydrates, 652 ; Assimilation Limits, 652; Saturation Limits, 654; Digestion and Absorption, 656; Sugar Level in the Blood, 657; Value of Blood Examinations in Diagnosis of Diabetes, 659; Relationship Between Blood Sugar and the Occurrence of Glycosuria, 660. CHAPTER LXXV METABOLISM OP THE CARBOHYDRATES (CONT'D) 662 Fate of Absorbed Glucose, Gluconeogenesis, 662 ; Storage of Sugar, 662 ; Sources of Glycogen, 662; Gluconeogenesis in Normal Animals, 667. CHAPTER LXXVI METABOLISM OF THE CARBOHYDRATES (CONT'D) 669 Fate of Glycogen, 669 ; Regulation of the Blood Sugar Level, 671 ; Nerve Control and Experimental Diabetes, 672; Nervous Diabetes in Man, 674; Hormone Control and Permanent Diabetes, 676; Utilization of Glucose in Tissues, 677; Relation of the Pancreas to Sugar Metabolism, 678; Diabetes and the Ductless Glands, 678; Diabetic Acidosis or Ketosis, 683; Starvation Treatment, 684. CHAPTER LXXVII FAT METABOLISM 686 Chemistry of Fatty Substances, 686; Digestion of Fats, 690; Absorption of Fats, 691. CHAPTER LXXVIII FAT METABOLISM (CONT'D) 696 Fat of Blood, 696; Methods of Determination, 696; Variations in Blood Fat, 697; Depot Fat, 700; Fat in the Liver, 701. CHAPTER LXXIX FAT METABOLISM (CONT'D) 707 Production of Fatty Acid Out of Carbohydrate, 707; Method by Which the Fatty Acid is Broken Down, 709. CHAPTER LXXX CONTROL OF BODY TEMPERATURE AND FEVER 714 Variations in Body Temperature, 714; Factors in Maintaining the Body Tem- perature, 715 ; Control of Temperature, 719 ; Fever, 721 ; Causes, 721 ; Changes in the Body during .Fever, 723 ; Heat-icgulating Center, 725 ; Significance of FevOr, 726. CONTENTS XXI PART VIII THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS CHAPTER LXXXI PAGE THE • ENDOCRINE ORGANS, OR DUCTLESS GLANDS 729 Methods of Investigation; 730; Adrenal Gland, 731; Cortex, 731; Medulla, 732; AdrenalectOmy, 733; Suprarenal Extracts, 734; Physiologic Action, 734. CHAPTER LXXXII ADRENAL GLAND (CONT'D) 738 Variations in Physiologic Activity, 738; Assaying the Epinephiine Content of the Gland, 738; Epinephrine Content of the Blood, 739; Autoinjection Method, 743 ; Adrenalemia, 745 ; Association of the Adrenal with Other En- docrine Organs, 746. CHAPTER LXXXIII THYROID AND PARATHYROID GLANDS -. . 749 Structural Relationship, 749; Thyroid Gland, 750; Condition of Gland, 750; Experimental Thyroidectomy, 752; "Disease of the Thyroid, 753; Relation with Other Endocrine Organs, 757; Parathyroids, 758; Experimental Parathy- roidectomy, 758 ; Relationship with Other Endocrine Organs, 761. • CHAPTER LXXXIV PITUITARY BODY 762 Structural Relationships, 762; Functions, 764; Clinical Characteristics, 771; Relationship, with Other Endocrine Organs, 773. CHAPTER LXXXV THE PINEAL GLAND AND THE GONADS 776 Pineal Gland, 776; Gonads or the Generative Organs, 776; Generative Glands of the Male, 776; Generative Organs of the Female, 778. PART IX THE CENTRAL NERVOUS SYSTEM CHAPTER LXXXVI THE EVOLUTION OF THE NERVOUS SYSTEM 781 CHAPTER LXXXVII PROPERTIES OF EACH PART OF THE REFLEX ARC 788 Receptor, 788; Epicritic and Protopathic Receptors, 790; Peculiarities of the Separate Sensations, 791 ; Temperature, 791 ; Touch, 793 ; Pain, 795. CHAPTER LXX XVIII THE PROPERTIES OF EACH PART OF THE REFLEX ARC (CONT'D) 796 The Nerve Network, 796; Network on Skin Nerves, 796; The Synapsis, 797; The Nerve Cell, 799; The Intermediate or Internuncial Neuron, 802. XX11 CONTENTS CHAPTER LXXXIX REFLEXES OF THE SPINAL ANIMAL AND SPINAL SHOCK 803 Spinal Shock in Laboratory Animals, 803; Spinal Shock in Man, 806; Cause of Spinal Shock, 807. CHAPTER XC PHYSIOLOGIC PROPERTIES OF THE SIMPLE REFLEX ARC 809 Latent Period, 809; Grading- of Intensity, 809: After-effect, 810; Summation, 810; Irreversibility of the Direction of Conduction, 810; Refractory Period, 811; Successive Degeneration, 813. CHAPTER XCI RECIPROCAL INNERVATION •. 814 Reciprocal Inhibition, 814; Action of Strychnine and Tetanus Toxin, 819. CHAPTER XCII INTERACTION AMONG REFLEXES . , 821 Integration of Allied Reflexes, 822; Integration of Antagonistic Reflexes, 824; Other Factors Which Determine Occupancy of Final Common Path, 824; Irradiation, 826. CHAPTER XCIII THE TENDON JERKS; SENSORY PATHWAYS IN SPINAL CORD 828 The Tendon Jerks, 828 ; Afferent Spinal Pathways, 830. CHAPTER XCIV EFFECTS OF EXPERIMENTAL LESIONS OF VARIOUS PARTS OF THE NERVOUS SYSTEM . 835 Anterior Roots, 835 ; Posterior Roots, 836 ; Spinal Cord, and Braiii Stem, 839 ; , Medulla, 839; Corpora Quadrigemina, 840; Removal of the Cerebral Hemi- spheres, 840. CHAPTER XCV CEREBRAL LOCALIZATION 843 Ablation of the Motor Centers, 843; Stimulation of the Motor Centers, 844; Clinical Observations, 849.' CHAPTER XCVI CEREBRAL LOCALIZATION (CONT'D) 850 Sensory Centers, 850; Sense Centers, 851; Association Areas, 852. CHAPTER XCVII CONDITIONAL AND UNCONDITIONAL REFLEXES 856 CHAPTER XCVIII HIGHER FUNCTIONS OF TIIE CEREBRUM IN MAN; APHASIA 860 Psychopathological Applications, 862. CHAPTER XCIX FUNCTIONS OF THE CEREBELLUM 865 Localization of Function, 867; Circumscribed Extirpation, 869; Clinical Ob- servations, 870. CONTENTS XX111 CHAPTER C THE CEREBELLUM AND THE SEMICIRCULAR CANALS; FUNCTIONAL TESTS .... 873 Association between the Eye Movements and the Semicircular Canals, 875. CHAPTER CI THE AUTONOMIC NERVOUS SYSTEM 877 General Plan of Construction, 877; Thoracicolumbar Outflow, or Sympathetic System Proper, 880 ; Bulbosacral Outflow, or the Parasympathetic System, 882 ; Axon Reflexes, 883; Functions of Autonomic Nerves, 884; Afferent Fibers of the Autonomic System, 885. ILLUSTRATIONS 1. Diagram of osmometer ... ................. 5 2. Hematocrite ........................ 7 3. Plasmolysis in cells from Tradescantia discolor ...:.. ..... 9 4. Apparatus for measurement of the depression of freezing point of solution . 11 5. Diagram of conductivity cells ................. 18 6. Wheatstone Bridge for the measurement of electric resistance ..... 18 7. Diagram to show type of electrodes used in studying electromotive force . . 30 9. Chart of tints as used in eolorimetric measurement of H-ion concentration. (Color Plate.) ...................... 34 8. Diagram of apparatus for the measurement of the H-ion concentration . . 31 10. Diagram of apparatus for saturating, blood and plasma with expired air . 43 11. Van Slyke's apparatus for measuring the CCycombining power of blood in blood plasma ...................... 44 32. Ultramicroscope (slit type) for the examination of colloidal solutions . . 52 13. To show diffusion into gelatin of a crystalloid stain, and the noridiffusion of a colloid stain .................... 53 14. Diagram from W. Ostwald showing the relative size of various particles and colloidal dispersoids compared with a red blood corpuscle and an anthrax bacillus . . . '. ................ 54 15. Capillary analysis of colloids ................. 56 16. Diagram to show structure of gels ................ 61 17. Diagram to illustrate surface tension .............. 64 18. Traube's stalagmometer .................... 65 19. Diagram of the graphic coagulometer ............... 109 20. Coagulometer ........................ 110 21. Mercury manometer and signal magnet, arranged for recording the mean ar- terial blood pressure in a laboratory experiment ......... 124 22. The arterial blood pressure recorded with a mercury manometer (lower trac- ing) along with a tracing of the respiratory movement of the thorax . 325 23. Hurthle's spring manometer .................. 126 24. Arterial pressure recorded by a spring manometer .......... 126 25. Diagram based on experiments on dogs to show the systolic, diastolic and mean blood pressures at different parts of the circulatory system . . 127 26. Apparatus for measuring the arterial blood pressure in man ...... 129 27. Effect of cutting the vagus nerve on the arterial blood pressure ..... 135 28. Effect of stimulating the peripheral end of the right vagus on the arterial blood pressure . . ................... 136 29. Effect of stimulation of the left splanchnic nerve on the arterial blood pres- sure ......................... 137 30. The effect of rapid and slow hemorrhage on the arterial blood pressure . . 138 31. Diagram of experiment to show that the diastolic pressure depends on the elasticity of the vessel wall ................ 1^3 32. Diagram of Wiggers' optical manometer ............. 146 XXVI ILLUSTRATIONS FIG. PAGE. 33. Optical records of intraventricular pressure 147 34. Superimposed pressure curves after being graduated 149 35. Von Frank's maximal and minimal valve, which is placed in -the course of the tube between heart and mercury manometer 152 36. Diagram to show the positions of the cardiac valves 155 37. Diagram showing the position of the cardiac chambers and valves during presystole and during the sphymic period 156 38. Elcctrophonograms along with intraventricular pressure curves from three different experiments 159 39. One form of apparatus for recording tracings from an excised heart . . 163 40. Volume curve of ventricles of cat (lower curve) in a heart-lung perfusion preparation 169 41. Heart and cardiac nerves of Limulus polyphemus 173 42. Heart-block produced by applying clamp 175 43. Tracing of contraction of ventricle, showing the effect of the local appli- cation of heat to the auricle . •. 175 44. Frog heart showing the position of the first and second ligatures of Stannius 176 45. Effects of stimuli of increasing strength on skeletal and cardiac muscle to illustrate the "all or nothing" principle in the latter 177 46. The effects of successive stimuli on skeletal and cardiac muscle to show the prominence of the staircase phenomenon, or treppe, in the latter . . 178 47. The effects, of successive stimuli and of tetanizing stimuli on skeletal muscle and cardiac muscle 179 48. Myograms of frog's ventricle, showing effect of excitation by break induc- tion shocks at various moments of the cardiac cycle 180 49. Heart of tortoise as suspended 183 50. Dissection of heart to show auriculoventricular bundle 184 51. Photograph of model of the auriculoventricular bundle and its ramifications, constructed from dissections of the heart 184 52. Diagram of an auricle showing the arrangemoit of the muscle bands; the concentration point; and the outline of the node 186 53. Diagram to show the general ramifications of the conducting tissue in the heart of the mammal 186 54. Diagram to illustrate the development and spread of the wave of negativity in a strip of muscle (curarized sartorius) when stimulated at the end . 188 55. Simultaneous electrocardiograms to show the cause for extrinsic deflections 190 56. Diagram of experiment by Lewis showing the times at which the excitation wave appeared on the front of the heart 194 57. Diagram of Chauveau's dromograph 200 58. Diagram to show principle of Pitot's tubes for measuring velocity pulse . . 201 60. Dudgeon's sphygmograph 201 61. Pulse tracing (sphygmogram) taken by sphygmograph 202 62. Forms of apparatus for measurement of blood velocities 207 63. Plethysmograph for recording volume changes in the hand and forearm . 210 64. Simultaneous tracings from auricle and ventricle of turtle's heart . . . 218 65. Effect of vagus stimulation on heart of turtle 218 66. Tracing to show that vagus stimulation may diminish transmission from auricles to ventricles 219 ILLUSTRATIONS XXV11 FIG. PAGE 67. Tracing to show that vagus stimulation may facilitate transmission from auricles to ventricles 220 68. Diagram to show the innervation of the heart in the frog or turtle. (Color Plate.) 224 69. Frog heart tracing showing the action of nicotine 226 70. Schematic representation of the innervation of the heart of the mammal. (Color Plate.) 226 71. Tracings showing the effects on the heartbeat of the frog resulting from stimulation of the sympathetic nerves prior to their union with the vagus nerve 228 72. Roy 's kidney oncometer 230 73. Fall of blood pressure from excitation of the depressor nerve 239 74. The effect of strong stimulation (heat) of the skin of the foot on the ar- terial blood pressure and respiratory movements 241 75. Diagram showing the probable arrangements of the vasomotor reflexes . 242 76. Aortic blood pressure, showing the effect of posture 245 77. Tracing to show the effect of gravity on the arterial blood pressure . . 245 78. The effect of gravity on the aortic pressure after division of the spinal cord in the upper dorsal region 246 79. Schema to show the relations of the Pacchionian bodies to the sinuses . . 248 80. Tracing showing simultaneous records of the arterial blood pressure, the venous pressure, the intracranial pressure, the pressure in the venous sinuses . . 251 81. Electrocardiographic apparatus as made by the Cambridge Scientific Ma- terials Co 260 82. Normal electrocardiogram 261 83. Electrocardiogram (dog) taken simultaneously with curves from auricle and ventricle 262 84. Eecords of electrocardiogram and movement of ventricle of frog showing that when the apex is warmed a typical T-wave appears in place of a wave in the opposite direction appearing when the apex is cooled . . 264 85. Sinus bradycardia 267 86. Auricular extrasystole 267 87. Ventricular extrasystoles arising in the right ventricle 267 88. Ventricular extrasystole arising in the left ventricle 267 89. Paroxysmal tachycardia 268 90. Auricular fibrillation 268 9X. Auricular flutter 2-70 92. Delayed conduction 270 93. Partial dissociation 271 94. Complete dissociation 271 95. Polysphygmograph . . . . • 274 96. Normal jugular tracing 274 97. Eeduced tracings from carotid, aorta, ventricle, auricle and jugular, to show the general relationships of the various waves 275 98. Polysphygmograms including jugular, apex and radial tracings .... 275 99. Delayed conduction time 277 100. Dropped beats 277 101. Premature beats (extrasystoles) ventricular in origin 278 XXV111 ILLUSTRATIONS FIG. PAGE 102. Paroxysmal tachycardia 278 103. Auricular flutter .... 279 104. Auricular flutter 279 105. Auricular fibrillation 280 106. Showing the appearance of the blood vessels in the ears of a rabbit in a state of deep shock. (Color Plate.) . 290 107; Diagram showing amounts of air contained by the lungs in various phases of ordinary and of forced respiration 301 108. Pneumograph 303 109. Body plethysmograph for recording respiration 304 110. Effect of abdominal and chest breathing on the pulse and blood pressure of man 308 111. First dorsal vertebra, sixth dorsal vertebra and rib. Axis of rotation shown in each case 316 112. Lower half of the thorax from the 6th dorsal to the 4th vertebra, seen from the front • . . . . 318 113. Intercostal muscles of 5th and 6th spaces 319 114. Hamberger's schema to demonstrate the functional antagonism of internal and external intercostals • 319 115. Schema to demonstrate that the function of the internal intercar- tilaginous intercostals is identical with that of the external in- terosseous intercostals 320 116. Diagram to show the effect of high and low positions of the diaphragm on the costal angle 322 117. Diagram to show the effect of clinical displacements of the diaphragm on the costal angle 323 118. Diagram to show cuts required for isolation of the phrenic center . . . 328 119. Diagram to show certain positions in the medulla and upper cervical cord, where sections may be made without seriously disturbing the respirations 329 120. Diagram to show where cuts are made to isolate the chief respiratory center from afferent impulses 330 121. Diagram showing principle for measurement of the tension of CO2 in blood 338 122. The gas analysis pipette for the microtonometer shown in Fig. 123 . . . 339 123. Microtonometer, to be inserted into a blood vessel 339 124. Apparatus for collection of a sample of alveolar air by Haldane 's method 340 125. Fridericia's apparatus for measuring the CO2 in alveolar air 341 126. Curves to show the relationship between the O2 and CO., tensions in alveolar air and arterial blood 341 127. Same as Fig. 126, except that in this case the tension of CO2 in the alveolar air was experimentally altered 342 128. Arrangement of meters and connections of Pearce's method for measure- ment of CO3 of alveolar air in normal subjects 346 129. Curve showing the respiratory response to CO, in the deccrebratc cat . . 351 130. Tensions of O2 and CO2 in alveolar air at different altitudes 361 131. Curves showing variations in alveolar gas tensions after forced breath- ing for 'two minutes 364 132. Various types of periodic breathing ; 372 ILLUSTRATIONS Xxix PIG. PAGE 133. Quantitative record of breathing air through a tube 260 cm. long and 2 cm. in diameter 374 134. Barcroft's tonometer for determining the curve of absorption of oxygen by hemoglobin or blood 381 135. Barcroft's differential blood gas manometer 381 130. Barcroft blood gas manometer 382 137. Typical dissociation curve. (Color Plate.) 382 338. Average dissociation curves 384 139. Dissociation curves of hemoglobin 385 140. Dissociation curves of human blood 386 141. Curves showing relative rates of oxidation and reduction of blood as influenced by temperature and by tension of CO2 387 142. Curve of CO, tension in blood 392 143. Cells of parotid gland showing zymogen granules . 419 144. Parotid gland of rabbit in varying states of activity examined in fresh state 419 145. Diagrammatic representation of the innervation of the salivary glands in the dog. (Color Plate.) 422 146. Pancreatic acini stained with hematoxylin 427 147. Three preparations of pancreatic acini stained by eosinorange toluidin blue 428 148. Diagram showing miniature stomach separated from the main stomach by a double layer of mucous membrane 434 149. Typical curve of secretion of gastric juice collected in 5-minute intervals on mastication of palatable food for 20 minutes 437 350. Cubic centimeters of gastric juice secreted after diets of meat, bread, and milk 440 151. Digestive power of the juice, as measured by the length of the protein column digested in Mett's tubes, with diets of flesh, bread, and milk . 441 152. Loop of intestine after tying off the portions, cutting the nerves running to the middle portion and returning the loop to the abdomen for some time 442 153. The changes which take place in the position of the root of the tongue, the soft palate, the "epiglottis and the larynx during the second stage of swallowing 446 154. Schematic outline of the stomach 452 155. Diagrams of outline and position of stomach as indicated by skiagrams taken on man in the erect position at intervals after swallowing food impregnated with bismuth subnitrate 452 156. Outlines of the shadows cast by the stomach at intervals of an hour each after feeding a eat with food impregnated with bismuth subnitrate . . 453 157. Section of the frozen stomach (rat) some time after feeding with food given in three differently colored portions 455 158. Outlines of shadows in abdomen obtained by exposure to x-rays 2 hours after feeding with food containing bismuth subnitrate 458 359. Curves to show the average aggregate length of the food masses in the small intestine at the designated intervals after feeding 459 160. Apparatus for recording contractions of the intestine 464 161. Diagrammatic representation of the process of segmentation in the intestine 465 162. Intestinal contractions after excision of the abdominal ganglia and section of both vagi 466 XXX ILLUSTRATIONS FIG. PAGE 163. The effect of excitation of both splanchnic nerves on the intestinal contractions . 467 164. The effect of stimulation of right vagus nerve on the intestinal contractions 468 165. Diagram of time it takes for a capsule containing bismuth to reach the various parts of the large intestine 469 166. Diagram of method for recording stomach movements 472 167. Tracing of the tonus rhythm of the stomach three hours after a meal . . 473 168. Tracings from the stomach during the culmination of a period of vigorous gastric hunger contractions 473 169. Showing augmentation of the knee-jerk during the marked hunger con- tractions 475 170. Diagram of the uriniferous tubules, the .'arteries., and the veins of the kidney 508 171. Cross section of convoluted tubules from kidney of rat 509 172. Diagram of blood supply of Malpighian corpuscle and of convoluted tubules in amphibian kidney 515 173. Nerve supply of the kidney 520 174. Respiration calorimeter of the Russell Sage Institute of Pathology, Bellevue Hospital, New York ! 536 175. Chart for determining surface area of man in square meters from Aveight in kilograms and height in centimeters according to formula . . . 540 176. Diagram of At water-Benedict respiration calorimeter 543 177. Nose clip, face mask, and mouthpiece 555 178. Diagram of respiratory valves 556 179. The Tissot spirometer 557 180. The Douglas bag method for determining the respiratory exchange . . 558 181. Haldane gas apparatus and Pearce sampling tube 559 182. Curve constructed from data obtained from a man who fasted for thirty- one days 567 183. Curves of growth of rats on basal rations plus the various proteins indicated 576 184. Curves of growth of rats on basal rations plus the proteins indicated . . 577 185. Photographs of rats of same brood on various diets 579 186. Vividiffusion apparatus of .T. J. Abel 607 187. Curves showing the amount of amino nitrogen taken up by different tis- sues after the cutaneous injection of amino acids 608 188. Curves showing the concentration of amino-acid nitrogen in the blood dur- ing fasting and protein digestion 609 189. Curves showing the percentage of glucose in blood after a constant injec- tion of an 18 per cent solution into a mesenteric vein 658 190. Arrangement of apparatus for recording contractions of a uterine strip, intestinal strip, or ring, etc 740 191. Tracing showing the effect of epinephrine on the intestinal contractions and on the arterial blood pressure 741 192. Arrangement of apparatus for perfusion of the vessels of .a brainless frog 742 193. Microphotographs of thyroid gland of dog 751 194. Cretin, nineteen years old 754 195. Case of myxedema before and after treatment 755 196. Drawing from a photograph of a mesial sagittal section through the pitui- tary gland of a human fetus 763 ILLUSTRATIONS XXXI FIG. PACK 197. Tracing showing the action of pituitrin on the uterine contractions and blood pressure in a dog 768 198. Tracing showing the constricting action of pituitrin on the bronchioles and its effect on blood pressure in a spinal dog 769 199. Showing the appearance before and after the onset of acromegalic symptoms 771 200. Hand of a person affected with acromegaly 772 201. Diagram showing gradual evolution of nervous system in sponge, sea anemone, and earthworm 783 202. Diagram of nervous system of segmented invertebrate, supraesophageal ganglion, subesophageal ganglion, esophagus or gullet 784 203 Schema of simple reflex arc 785 204. Thermoesthesiometer 791 205. Cold spots and heat spots of an area of skin of the right hand .... 792 206. Diagram to show axon reflex of sensory nerve fiber of skin 797 207. Arborization of collaterals from the posterior root fibers around the cells of the posterior horn 798 208. Normal cell from the anterior horn, stained to show Nissl's granules . . 799 209. Part of an anterior cornual cell from the calf's spinal cord, stained to show neurofibrils 800 210. Living nerve cells examined by the ultramicroscope 801 211. Tracing from the hind limb of a spinal dog during the scratching move- ments produced by applying stimuli at two skin points 812 212. Eecord from myograph connected with the extensor muscle of the knee . 815 213. Diagram showing the muscles and nerves concerned in reciprocal inner- vation 816 214. Eeciprocal innervation 817 215. Sherrington 's diagram illustrating the mechanism of reciprocal innervation 818 216. Diagram showing the reflex arcs involved in the scratch reflex .... 822 217. Showing region of body of dog from which the scratch reflex can be elicited 823 218. Diagram showing the segmental arrangement of the sensory nerves . . 837 219. Outer aspect of the brain of the chimpanzee 847 220. Three sections through different parts of the cerebral cortex 852 221. The location of the chief motor and sensory areas on the outer and mesial aspects of the human brain 853 222. Footprints after destruction of the cerebellum in a dog 866 223. Diagrams to represent respectively a ventral view of the left half and a dorsal view of the right half of the human cerebellum illustrating the scheme of subdivision according to Bolk 868 224. Schema of the parts of the mammalian cerebellum spread out in one plane 869 225 and 226. The inferolateral and the posterior aspect of the human cerebellum indicating certain cerebellar localizations according to Barany . . . 871 227. The semicircular canals of the ear, showing their arrangement in the three planes of space 874 228. Diagram illustrating the different arrangements of the internuncial neurons of the voluntary and involuntary nervous systems 878 229. Diagram of the sympathetic nervous system to be used along with Fig. 232. (Color Plate.) 878 230. Diagram showing the manner of connection of the fibers composing the great splanchnic nerve. (Color Plate.) 878 XXX11 ILLUSTRATIONS FIG. PAGE 231. Diagram showing the manner in which a preganglionic fiber, emanating from the spinal nerve by the white ramus communicans, connects in a ganglion of the sympathetic chain with a nerve cell, the axon of which then proceeds as the postganglionic fiber by way of the gray ramus communicans back to the spinal nerve, along which it travels to the periphery. (Color Plate.) 880 232. Diagram showing the main parts of the autonomic nervous system to be used along with Fig. 229. (Color Plate.) . . 882 233. Schematic representation of the involuntary nervous system. (Color Plate.) 884 PHYSIOLOGY AND BIOCHEMISTRY IN MODERN MEDICINE PART I THE PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES CHAPTER I GENERAL CONSIDERATIONS The work of the physiologist consists, in large part, in ascertaining to what extent the known laws of physics and chemistry find application in explaining the phenomena of life. He gathers from the vast store- house of physical and chemical knowledge whatever is of value in the interpretation of the various mechanisms that work together to com- pose the living machine, and having added to this knowledge he passes it on for use by those who are concerned in the study and treatment of disease. Many of the most important steps in the advance of physiologic knowledge in recent years have depended upon the discovery of some hitherto unknown physical or chemical law, or upon the elaboration of some accurate method for the measurement of the phenomena upon which these or previously known laws depend. The discoveries of van't Hoff, Arrheiiius, and Ostwald of the so-called laws of solution were soon followed by important observations on their relationship to the movement of fluids and dissolved substances through cell mem- branes; the discoveries of Hardy, Willard Gibbs, etc., of the behavior of colloids and of the phenomena of surface tension found application in explaining many hitherto inexplicable peculiarities in the activities of ferments; the discovery by Nernst, etc., of methods for the measurement of the electro-motive force of dissolved substances was applied to de- termine the actual reaction or hydrogen-ion concentration of animal 1 PHYSICOCH^MICAL BASIS OF PHYSIOLOGICAL PROCESSES fluids, and to explain the generation of the electric currents which ac- company muscular, nervous, and glandular activity. It would be out of place here to devote much space to a detailed ac- count of such matters. They belong more properly in the domain of general than in that of human physiology. General physiology is con- cerned with the study of the essential nature of the vital processes; whereas human physiology is merely a branch of the subject in which special attention is devoted to the application of the truths of general physiology to the working of the human machine. For the physician and surgeon a knowledge of human physiology is as essential as is a knowledge of the construction of a piece of machinery for the engineer who attempts its repair, but obviously to acquire this knowledge the fundamental principles of general physiology must first of all be under- stood. For these reasons the introductory chapters are devoted to a brief review of the most important of the physicochemical principles upon which .the working of the cell depends. From the viewpoint of the physical chemist the cell consists of an envelope of more or less permeable material inclosing a dilute solution of crystalline substances in which colloid matter is suspended. It con- tains, in other words, a solution of crystalloids and colloids, in which these are in a state of equilibrium with each other. This equilibrium is readily altered by various influences that may act on the cell, and the resulting changes manifest themselves outwardly by alterations in the shape and volume of the cell — growth and motion; by the extrusion of some of its contents — secretion; or by the propagation to other parts of the cell, or its processes, of the state of disturbed equilibrium — nervous impulse. Besides the activities that are dependent upon physicochem- ical changes, purely chemical processes go on in the cell. Many of these consist in the breakdown and oxidation of complex unstable organic molecules, a process identical with that occurring in combustion outside the cell. Others involve the building up, stage by stage, of complex substances out of the elements or out of simpler molecules. Chemical transformations occur in the cell which, in the chemical laboratory, re- quire the most powerful reagents and physicochemical forces, either the •strongest of acids, alkalies, oxidizing agents, etc., or extreme degrees of heat, electrical energy, etc. But this is not all, for in the cell these chemical transformations are capable of being guided to a very remark- able degree of nicety so as to produce intermediate products that are used for some special purpose either by the cell that produced them or, after transportation by the blood, etc., by cells in other parts of the organism. It is customary to speak of the' cell as a chemical laboratory, but it LAWS OF SOLUTION 3 is more than this; it is a laboratory furnished not only with the equip- ment of the chemist but directed in the harmonious operation of its many activities by a guiding hand which far surpasses anything known to man. Chemical transformations that require for their accomplishment the greatest skill proceed without apparent difficulty in the cell. To what are these changes due? What is the nature qf the chemical rea- gents and forces, and what is the directive influence that guides them in their varied activities? To these, which are among the great ques- tions of general physiology, the reply may be given that the reagents are the ferments or enzymes, and that the directive influence operates through the susceptibility of enzymic activities to changes in the envi- ronment in which the enzymes are acting. In many cases these changes can be explained on a physicochemical basis as dependent upon the known laws of mass action or surface tension; in other cases they de- pend on purely chemical changes in the cell contents, such as changes in reaction or the accumulation of chemical substances that act like poisons on the enzyme. But there are still others that appear to depend on influences which as yet are quite unknown to the physical chemist, such as the changes in cell activity that can be brought about by the nerve impulse. These preliminary remarks will serve to indicate the problems with which we must first occupy our attention. They concern the physico- chemical nature of saline solutions and ef colloids, and the general na- ture of enzyme action. The knowledge which we acquire will be found to be of value, not only because it will help us to understand the nature of the workings of the normal healthy cell, but because, here and there, it will indicate possible causes for derangement in cellular function and suggest rational means by which we may attempt to rectify the fault. THE PHYSICOCHEMICAL LAWS OF SOLUTION The Gas Laws Three fundamental principles of general chemistry serve as the basis for an understanding of the nature of solutions. The first is that if we take a quantity of any gas equal to its molecular weight in grams (called a gram-molecule or for sake of brevity a mol), it will occupy ex- actly 22.4 liters at standard temperature and pressure ; the second is that, as we compress a gas, its pressure will increase in exactly the same proportion as the volume diminishes (the volume of a gas is inversely proportional to its pressure) ; the third is that all gases expand by 1/273 4 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES part of their volume at 0° C. for every degree C. that their temperature is raised.* The pressure of a gas is measured by connecting a pressure gauge or manometer with the vessel which contains the gas. Now, it is plain that if the 22.4 liters, which is the volume occupied by a gram-molecular quantity, were compressed so as to occupy a volume of 1 liter, its pressure would be 22.4 times that of 1 atmosphere, or 22.4 x 760 mm. Hg — the temperature remaining constant. Under these conditions we must im- agine that the molecules of gas are crowded together by the compression, and if we further conceive of these molecules as being in constant mo- tion, then we can understand why the pressure should increase just in proportion as we confine the space in which they can move. One other property of gases must be borne in mind — namely, their tendency to diffuse from places where the pressure is high to places where it is low until the pressure is the same throughout. OSMOTIC PRESSURE These fundamental facts regarding the behavior of gases suggested to van't Hoff the hypothesis that molecules of dissolved substances must behave in a similar manner to those of gases. To put this hypothesis to the test, it is necessary that we have some method for measuring the pressure of dissolved molecules. We can not, as in the case of a gas, use an ordinary manometer, for this would measure only the pressure of the solvent on the walls of its container and would tell us nothing of the pressure of the dissolved molecules. We must use some filter or membrane that will allow the molecules of the solvent but not those of the dissolved substance to pass through it. It is evident that if such a filter is placed, for example, between a solution of sugar in water and water alone, the molecules of the latter will diffuse into the solution until this has become so diluted that the pressure of the dissolved mol- ecules is equal on both sides of the membrane. Such a membrane is called semipermeable ; the diffusion of molecules through it is called osmosis, and the pressure which is generated, the osmotic pressure. If we prevent the water molecules from actually diffusing by opposing a pressure which is equal to that with which they tend to diffuse through the membrane, we can tell the magnitude of the osmotic pressure (Fig. 1). In applying these facts to test the hypothesis that molecules in solution *This implies that at -273° C. the gas would occupy no volume. Before this temperature is reached, however, the liquefaction of the gas sets in. The temperature -273° C. is known as absolute zero. An observed temperature phis 273° is called the absolute temperature. Another way of stat- ing the above law is therefore that the volume is directly proportional to the absolute temperature. At 273° C. the volume of a gas at 0° C. would be doubled, or if expansion were prevented the pressure would be doubled. LAWS OF SOLUTION obey the same laws as those in gaseous form, we must employ a semi- permeable membrane which is rigid enough to withstand the pressure and which forms part of the walls of a closed vessel connected with a manometer. If we place in such an osmometer a solution containing the molecular weight in grams of some substance dissolved in one liter of solvent, a so-called gram-molecular solution, it is obvious that, if the gas laws are to apply, the osmotic pressure should equal that of 22.4 liters of a gas compressed to the volume of one liter; in other words, it should equal 22.4 x 760 mm. Hg. Although there are very consider- able technical difficulties in making a semipermeable membrane that is strong enough to withstand such a pressure, yet this has been accom- ! M w Fig. 1. — Diagram of osmometer. The cylindrical vessel (O), with a bottom of unglazed clay, the pores of which are filled with a precipitate of copper fe,rrocyanide to form a semi- permeable membrane, is suspended in an outer vessel, and is closed above by a tightly fitting stopper pierced by a tube leading to a manometer (.M). O contains a strong solution of cane sugar, and W contains water. The water molecules tend to pass through the semipermeable membrane into the cane sugar solution, and since the cane sugar molecules can not pass in the opposite direction, the pressure in O rises and is recorded in M. This equals the osmotic pressure. plished, and the fundamental principle has therefore been firmly estab- lished that substances in solution obey the same laws as gases. Further proof that the gas laws apply to solutions has been secured by showing that the osmotic pressure (of a dilute solution) is directly pro- portional to the concentration of the dissolved substance (the solute) and to the absolute temperature. It also obeys the law of partial pres- sures, which states that the total pressure exerted by a mixture (of gases or dissolved molecules) is the sum of the pressures which each constit- uent of the mixture would exert were it alone present in the space occupied by the mixture. 6 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES Since the osmotic pressure is analogous to the pressure of a gas and is therefore proportional to the molecular concentration (i. e., number of molecules in unit space), it follows that a semipermeable membrane can be used to determine the relative concentration of two solutions of the same substance. When a watery solution of some substance is placed in an osmometer that is surrounded by a similar but more dilute solution, water molecules will diffuse into the osmometer until the pres- sure is equal on the two sides of the semipermeable membrane; that is, the water will pass from the solution having a lower osmotic pressure into the solution having the higher pressure. When two solutions have the same osmotic pressure, they are said to be isotonic; when that of one is greater than that of the other, it is kypertonic; and when less, hypotonic. Biological Methods for Measuring Osmotic Pressure A practical biological application of these principles can very readily be made if, instead of a rigid semipermeable membrane such as that figured in the diagram, we employ one that is extensible and takes the form of a closed sac ; then as diffusion of water occurs the sac will either distend when it contains a stronger solution than that outside, or shrivel or crenate when the reverse conditions obtain. Many animal and veg- etable protoplasmic membranes are semipermeable, including the en- velope of red blood corpuscles. Thus, if we examine blood corpuscles under the microscope and add to them a saline solution of higher os- motic pressure than blood serum, they will visibly diminish in size and become irregular in shape; whereas if the solution is of lower osmotic pressure, they will distend. If no change occurs, the osmotic pressure of the cell contents must equal that of the saline solution in which the cells are immersed, from which it is clear that we can readily determine the magnitude of the osmotic pressure if we know the strength of the saline solution. Instead of measuring the individual cells under the microscope, we can measure the space they occupy in the fluid in which they are suspended. For this purpose a portion of the suspension is placed in a graduated tube of narrow bore, which is rotated in a horizontal position by a cen- trifuge after being closed at one end. The graduation at which the upper edge of the column of cells stands after centrifuging is a measure of the relative amount of cells and fluid in the suspension. Having found this value for cells suspended in an isotonic solution, as for blood corpuscles in blood serum, we may then proceed to ascertain it for the same cells suspended in an unknown solution; if we find that the cells occupy a greater volume, the saline solution must have an osmotic pres- LAWS OF SOLUTION 7 sure that is lower than that of serum in approximate proportion to the readings on the tube in the two cases, and vice versa. The above apparatus, called a hematocrite (Fig. 2) has been very ex- tensively used in the collection of data concerning the relative osmotic pressures of different physiologic fluids. Hemolysis Another way for determining the relative osmotic pressure of dif- ferent solutions consists in placing equal amounts (a few drops) of blood in a series of test tubes containing solutions of different strengths, and after allowing the tubes to stand for some time, noting in which of them laking of the blood corpuscles occurs. In solutions which are isotonic or hypertonic with the contents of the corpuscles, the latter will settle to the bottom of the tube and the supernatant fluid will be untinted with hemoglobin, but in solutions which are distinctly hypotonic, the sediment will be less distinct and the supernatant fluid red. Fig. 2. — Hematocrite. The graduated glass tubes are filled with the two specimens of blood, or corpuscular suspension, and then rotated rapidly by a centrifuge. The relative heights at which the corpuscular sediment stands in the two tubes is proportional to the osmotic pressures of the fluid in which the corpuscles are suspended. By noting (1) the lowest concentration (percentage composition) of the solutions in which the corpuscles sink to the bottom and leave the supernatant fluid colorless, and (2) the highest concentration in which the corpuscles when they settle leave the supernatant fluid red, we can determine the limiting concentrations for solutions of different sub- stances. Thus, with bullock's blood the following results were obtained (Hamburger) : SUBSTANCE PERCENTAGE STRENGTH OF SOLUTION IN WHICH: I II SUPERNATANT FLUID SUPERNATANT FLUID WAS COLORLESS WAS R^D KN03 1.04 0.96 Nad 0.60 0.56 K2S04 1.16 1.06 C12H22O, „ (Cane sugar) 6.29 5.63 CH,COOH (Pot. acetate) 1.07 1.00 MgS04. 7H2O 3.52 3.26 CaCL, 0.85 0.79 8 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES The mean of these limiting concentrations is the critical concentration and indicates the strength of each solution that can be added to blood without causing any damage to the corpuscles. This critical concen- tration is not, as might at first sight be imagined, the same as that which is isotonic with the contents of the corpuscles, but distinctly below it. The reason for this becomes apparent if we observe the be- havior of corpuscles suspended in an isotonic solution which is then gradually diluted. As dilution proceeds, the corpuscles distend, until at last their envelopes burst and the hemoglobin is discharged. The lim- iting concentrations of a given salt vary for different corpuscles; thus, the concentration of sodium chloride solution that just causes laking of frog's blood corpuscles is 0.21 per cent, of human blood 0.47 per cent, and of horse blood 0.68 per cent. It is the strength of the corpuscular envelope rather than variations in the osmotic pressure of the contents that is responsible for these differences. The above described method of hemolysis, as it is called, can not be used for comparisons of osmotic pressure in cases in which the solution contains substances which alter the permeability of the corpuscular envelop ; for example, it can not be used when urea, or ammonium salts, or certain toxic bodies are present. This very fact is, however, put to a useful purpose in ascertaining whether a given substance does have a damaging influence on the corpuscular envelope by finding whether hemolysis occurs when we suspend the corpuscles in a solution that is isotonic with the corpuscular contents. We can further determine the degree of this toxic influence by estimating by color comparisons (colorimetry) the amount of hemoglobin that has diffused out of the corpuscles. Plasmolysis An analogous method for determining osmotic pressure is that of plasmolysis, in which the behavior of certain plant cells is observed microscopically while they are in contact with solutions of different strengths. When the surrounding solution is isotonic with the cell contents, the latter fill the cell and extend up to the more or less rigid cell wall (A in Fig. 3) ; but when the solution is hypotonic, the cell contents become detached from the cell wall at one or more places — plasmolysis (B and C). The semipermeable membrane in this case is therefore not the cell wall but the layer of protoplasm on the surface of the cell contents. The method can be used only for detecting solu- tions that are hypertonic, for with those that are hypotonic the cells merely become turgid and exert more pressure on the more or less rigid cell wall. Many of the conclusions that have been drawn from LAWS OF SOLUTION results obtained by the plasmolytic method have recently been called in question, because no regard has been taken of the power of the colloids of the cell to adsorb (imbibe) water (see page 62). The methods of hemolysis and plasmolysis have been used for the investigation of many problems in medicine. In the case of certain toxic fluids, such as snake venom, tetanus toxin, etc., determination of the hemolytic power has proved of value in roughly assaying the dam- aging influence on other cells than blood corpuscles. Studies in hemol- ysis have also been especially valuable in working out the mechanism by which cellular toxins in general develop their action, and the conditions under Avhich this action may be counteracted, as by the development of Fig. 3. — To show plasmolysis in cells from plasmolysis in 0.22 M. cane 'sugar; C, pronounc wall; p, the protoplasm. (After De Vries.) cells from Tradescantia discolor. A. norma discolor. A, normal cell; B, :cd plasmolysis in 1.0 M. KNO3; h, the cell antibodies. Furthermore, any solution that is to be injected into the animal body, either intravenously or subcutaneously, should first of all be tested by the above methods in order to find out whether it is isotonic with the body fluids. If a hypertonic solution is injected, it will result in the abstraction of water from the tissue cells, whereas a hypotonic solution will cause the water content of these to increase. Advantage has recently been taken of this water-abstracting effect of hypertonic solutions in the treatment of wounds. By constantly bathing them with strong saline solutions, an outflow of water is 'set up from the tissue cells that border on the wound, and this tends to bring to the focus of infection the defensive substances that are present in animal fluids. CHAPTER II OSMOTIC PRESSURE (Cont'd) Measurement by Depression of Freezing Point The limitations in the use of the plasmolytic and hemolytic methods in the precise measurement of the osmotic pressure of the body fluids have rendered it necessary to find some physical method that will be generally applicable. Because of technical difficulties, it is impracticable to measure the pressure directly by employing an osmometer, so that some indirect method, depending on a readily measurable physical prop- erty which varies in proportion to the osmotic pressure of the dissolved substances, must be used. Fortunately, one such exists in the property which dissolved substances have in lowering the temperature at which the pure solvent solidifies; the freezing point of pure water, for example, is lowered when substances are dissolved in it, and the extent of this lowering, with certain reservations which will be explained later (page 16), is proportional to the molecular concentration of the solution and independent of the chemical nature of the substance dissolved. This lowering of temperature is designated by the Greek letter A, and to measure it a thermometer is used which is not only extremely sensitive but in which the level of the mercury column can be adjusted so that it stands at a convenient level on the scale corresponding to the freezing point of whatever solvent was used in making the solution under investi- gation (Beckmann's thermometer) (Fig. 4). The exact position on the scale of this thermometer at which the pure solvent freezes having been ascertained, the observation is repeated with the solution whose osmotic pressure is to be determined. A gram-molecular solution in water (having therefore an osmotic pres- sure of 170,240 mm. Hg) has a freezing point that is 1.86° C. lower than that of pure water. This is known as the "freezing point constant," and it varies for different solvents, being 3.9 for acetic acid and 4.9 for benzene. If an unknown watery solution is found to have a freez- ing point that is A° C. lower than that of water, its osmotic pressure , Ax 17.024 _ will equal — — ^ mm, Hg, l.OQ 10 OSMOTIC PRESSURE 11 The depression of the freezing points produced by the various body fluids has been compared, the objects in view being to see whether osmotic pressure is a property which changes under different physiologic and pathologic conditions, and to find out by comparison of the osmotic pressures of the fluids in contact with a membrane, whether physical forces alone can be held responsible for the transference of substances through it from one fluid to the other. The Role of Osmosis, Diffusion, and Allied Processes in Physiologic Mechanisms An account of some of the investigations in which the foregoing methods have been used will illustrate their value in revealing the Fig. 4. — Apparatus for measurement of the depression of freezing point of solutions. The solution is placed in the large test tube with the side arm, and in it is suspended the bulb of a Beckmann thermometer with a platinum loop to serve for stirring. The upper end^of the mercury column of the thermometer is shown magnified at the upper left corner. The amount of mercury in the thermometer tube can be regulated by tapping the upper end with the thermometer in various positions. The test tube is protected by an outer tube, which is then placed in a vessel containing a freezing mixture. mechanism involved in the transference of water and dissolved sub- stances through cell membranes, as occurs in absorption of food in the intestine, in the formation of lymph and urine, and so forth. In em- ploying physical methods in the elucidation of such problems, it is always most necessary to proceed with great care, since the physical 12 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES chemist works with pure solutions, while the physiologist has to use fluids that are always complicated and frequently very variable in com- position. We must simplify the problem as far as possible by having clearly before us the exact nature of the biological problem which a com- parison of physicochemical values, such as osmotic pressure, may ena- ble us to elucidate, and we must consider the other physical forces which may assist or modify the particular one we are investigating. In the physical experiments described above, the s-emipermeable mem- brane may be conceived of as composed of pores of such a size that they permit only the smallest of molecules — those of water — to pass through them. Semipermeable membranes with larger pores may, how- ever, exist — that is, membranes which permit water molecules and mole- cules of simple chemical substances to pass, but hold back those com- posed of large complex molecules. Such a semipermeable membrane would allow -the saline constituents but not the proteins of blood serum to pass. It is, however, no longer semipermeable towards all of the dis- solved substances, and the process of diffusion through it is more gener- ally designated as one of dialysis than of osmosis. Since the passage of dissolved molecules through membranes de- pends upon the principle of diffusion, its rate will be proportional to the osmotic pressures of the solutions on the two surfaces of the mem- brane and to the size of the molecules, small molecules diffusing more quickly than large . ones. Suppose a membrane permeable to sodium chloride and water is placed between two fluids containing sodium chloride in solution, but in greater concentration in one of them than in the other: the sodium chloride will diffuse from the stronger to the weaker solution, and water will diffuse still more quickly (because its molecules are smaller) in the opposite direction, until the number of sodium-chloride molecules in a given volume of solution is equal on both sides of the membrane. For a time, therefore, the volume of the stronger solution will increase. The differences which exist in the dif- fusibility of dissolved molecules are analogous to those which have long been known to exist in the diffusibility of gases, but the relation between rate of diffusibility and molecular weight is not so simple as the ratio between these two quantities in gases. These relationships, however, indicate several further possibilities in the explanation of the mechanism of exchange of substances through membranes, and must not be overlooked, as they often are, in the interpretation of physiologic phenomena. An excellent review of the possible conditions is given by Starling in his " Human Physiology."4 For example, let us suppose the substances on the two sides of a semipermeable membrane, such as the peritoneal, to be different in diffusibility, as cane sugar, OSMOTIC PRESSURE 13 which does not readily diffuse, and sodium chloride, which diffuses quickly; the osmotic flow will take place from the sodium-chlorid solu- tion to the cane sugar even when the sodium-chloride solution is stronger than the sugar. In such a case, water molecules will pass from the fluid having the higher osmotic pressure (NaCl) toward a fluid in which this is lower (sugar). Furthermore, the simple laws of osmosis may be upset by an attrac- tive influence of the membrane toward certain substances [due to their becoming dissolved or adsorbed in it (see page 65)] but not toward others. Many membranes of this nature are known to the chemist (e. g., rubber membranes in contact with gases, pyridine solutions, etc.), and it is probable that such a property of selective solubility may play a not unimportant role in the transference of substances across animal membranes (Kahlenberg5). These few conditions which may modify the direction of the osmotic flow, are indicated here to show how involved such problems are, and how careful we must be not to assume that, because a substance is trans- ferred through a living membrane contrary to the simpler laws of os- mosia and diffusion, it must involve the expenditure of forces different from those operating in dead membranes. Another force comes into operation under certain conditions — namely, that of filtration. This is a purely mechanical process, in Avhich mole- cules are forced through the pores of a filter (i. e., membrane) by dif- ferences in pressure on its two sides. We are now in a position to consider in how far the above physical forces explain certain physiologic problems. 1. Is the absorption, into the blood and lymph circulating in the intes- tinal walls, of substances in solution in the intestinal contents entirely dependent upon the processes of filtration, diffusion and osmosis? The absorption of weak solutions of highly diffusible substances is probably very largely a matter of osmosis and diffusion, and water passes quickly into the blood because of osmotic attraction, but that other forces ordi- narily come into play is very clearly established by the following ob- servations. If a piece of intestine is isolated from the rest by placing two ligatures on it, and the isolated loop filled either with a solution con- taining the same saline constituents in similar proportions as in blood serum, or better still, with some of the same animal's blood serum, it will be found after some time that all of the solution becomes absorbed into the blood ; the contents of the loop are therefore absorbed into the blood, even though the osmotic pressures of the dissolved substances are the same on both sides of the membrane (Weymouth Reid6). The intestinal membrane seems to possess towards readily diffusible 14 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES substances a permeability which varies, not at all with the physical diffusibility of the substance, but with its value from a physiologic standpoint. Thus, sodium sulphate and sodium chloride diffuse through ordinary membranes with about equal facility, and yet if a solution con- taining these two salts is placed in the intestine, the chloride will be absorbed into the blood much more quickly than the sulphate. Sodium sulphate in watery solution diffuses through a membrane fifteen times more quickly than cane sugar, but from the intestinal lumen, cane sugar is absorbed ten times more quickly than sodium sulphate. If. however, the vitality of the epithelium is destroyed, as by first of all bathing it with a solution of sodium fluoride, then the sulphate and chloride will be absorbed at an equal rate. Although diffusion and osmosis can not therefore play any significant role in the normal process of absorption from the intestine, we must not entirely discount them; under certain circumstances, these physical forces may assert their influence as, for example, when concentrated saline solutions are present. Such solutions will attract water from the blood, and, other things being equal, more will be attracted the less permeable the epithelium happens to be towards the saline employed. Sulphates and phosphates will attract more water than chlorides or acetates. This property of the saline solutions to attract water coun- teracts the natural tendency for the water to be absorbed, and the large volume of fluid stimulates peristalsis. 2. Do the physical processes of filtration, diffusion and osmosis suf- fice to account for the production of urine ~by the kidneys? Under normal conditions the molecular concentration of the urine, as determined by the depression of freezing point, is considerably greater than that of the blood. This indicates that excretion must have occurred contrary to the laws of osmosis; in other words, that the renal cells must have compelled dissolved molecules to be transferred from the blood to the urine, although the difference in osmotic pressure would cause them to pass in the opposite direction. This force, sometimes called for want of a better name "vital activity," must depend on the operation of processes that are quite distinct from those of diffusion, etc.; but that they are necessarily of a nonphysical nature (e, g., vital) is less probable than that they depend on some physical process the nature of which our present knoAvledge does not permit us to understand. By comparing the osmotic pressures of urine and blood, attempts have been made to measure the work done by the kidney in the produc- tion of urine. Thus, it has been found that A for normal urine (human) is about 1.8, and for blood about 0.6, from which it may be calculated that in the production of 1 kilogram of urine 150 kilogrammeters of OSMOTIC PRESSURE 15 work are expended.* But that such comparisons of the osmotic pres- sure of blood and urine are fallacious as an indication of the work of the kidney is evidenced, not alone by the results of the above calcula- tions, but also by the fact that under certain circumstances (as after copious diuresis) the osmotic pressure of the urine may be considerably lower than that of the blood. That opposite relationships should exist indicates that differences in osmotic pressure between blood and urine can signify little if anything regarding the work done by the kidney. For some time after the application of osmotic pressure measurements to the study of biological problems, it was thought that determination of A in urine might be of clinical value as a criterion of renal efficiency, especially in one kidney as compared with the other. For this purpose A was determined in samples of urine removed from each ureter by catheterization. The tests of renal efficiency based on the rate of excre- tion of potassium iodide, phenolphthalein, etc., have however been found of much greater value. 3. Is the formation of lymph purely a physical process? The osmotic pressure of normal lymph is nearly always somewhat below that of blood serum, although occasionally it has been found to be a trifle higher. Physical processes, such as filtration, might therefore suffice to account for its formation under most conditions. But when we con- sider the excessive production of lymph that occurs 'as a result of cel- lular activity or following the injection of certain substances, called "lymphagogues," it is not so easy to explain the .production in such terms, although some interesting attempts have been made to do so by those that are wedded to the mechanistic view. For example, the very marked increase in lymph flow which occurs as a result of muscular exercise or glandular activity has been attributed to the fact that dur- ing such processes large molecules become broken down into small ones in the cell protoplasm, so that the osmotic pressure is raised and water is attracted into the the cell until the latter becomes distended and a process of filtration into the neighboring lymph spaces occurs (see page 119). There are several other physiologic processes of secretion and excre- tion which might be considered in the present relationship, but the above instances will suffice to illustrate the general principle upon which all of them have to be considered. "Osmotic pressure corresponding to A = -0.6° C. equals 5,662 mm. Hg (75 m. of H2O), and that corresponding to A = -1-8° C. equals 16,986 mm. Hg (225 m. H2O). The difference is there- fore equal to a column of water 150 m. high. According to these calculations it would appear that the kidney in producing the average daily output of 1500 c.c. urine performs 225 kilogrammeters of work in comparison with the 14,000 kilogrammeters which the heart is computed to perform in the same time (page 212). CHAPTER III ELECTEIC CONDUCTIVITY, DISSOCIATION, AND IONIZATION The osmotic pressure is not infrequently found to be considerably greater than that expected from the strength of the solution. Although A of a gram-molecular watery solution of cane sugar (342 gm. to the liter) is 1.86 (see page 10), that of sodium chloride (58.5 gm. to the liter) is considerably greater. If the hypothesis regarding the relationship of molecular concentration to osmotic pressure is to hold good, it becomes necessary to explain this apparent inconsistency; one must account for a greater number of dissolved units than is represented by the actual number of dissolved molecules (i.e., weight of dissolved substances). It was observed that the power to conduct the electric current — electric conductivity — in the case of solutions (e. g., of sugar) which have an osmotic pressure that corresponds to the weight of dissolved substances is practically nil, whereas the conductivity of those solutions which give higher osmotic pressure is quite pronounced. Arrhenius made the hy- pothesis that the conductivity depends on the splitting of molecules into two or more portions or ions, each of which carries either a positive or a negative electric charge, and that it is only when such dissociation occurs that the electric current can be conducted through the solution, the ions serving as it were as floats carrying the electric current. When sodium chloride is dissolved in water, it splits into Na carrying a positive charge and Cl carrying a negative charge, or Na H Cl -, as it is written ; on the other hand, when sugar is dissolved, the molecules remain unbroken and no electric charges are set free. Substances which thus dissociate are called electrolytes, and those which do not, nonelectrolytes. When the electric current is passed through a solution of electrolytes, the ions which carry a positive charge move to the electrode or pole by which the current leaves the solution — that is, in the same directions as the current; and since this electrode is called the cathode, these are called cations. Hydrogen and the metals belong to this group. The ions carrying a negative charge go in the opposite direc- tion, against the current — that is, towards the electrode by which the cur- rent enters, or the anode; they are therefore called anions. They include oxygen, the halogens and the acid groups, such as S04, C03, etc. It must be understood that this dissociation into ions is already present in the solution before any electric current passes through it, the ions 16 ELECTRIC CONDUCTIVITY, DISSOCIATION, IONIZATION 17 being however uniformly distributed throughout — that is, arranged so that the negative charges of the anions precisely neutralize the positive charges of the cations. The electric current causes the electrodes to be- come charg-ed, the one positively, the other negatively, so that an attrac- tive force is exerted on the ions of opposite sign. This causes the nega- tively charged ions to migrate towards the positive electrode, and the positively charged, towards the negative electrode. It is this migration of the ions that endows the solution Avith conducting qualities. In water, or in a solution of a nonelectrolyte, molecules of H20 or non- electrolyte exist thus: H,0 H20 H20 H20 H20 H20 H20 H20 H20 In a solution of an electrolyte, the molecules split into ions thus: Na+ Cl- Na+ Cl- Na+ 01- Na* Cl- Na+ Cl- Na* 01- Na+ Cl- Na+ Cl- Na+ 01- When an electric current passes through a solution of an electrolyte, the ions tend to arrange themselves thus: Cathode- Anode* Na> Na+ Na* Cl- Cl- 01- Na+ Na+ Na+ Cl- Cl- 01- Na+ Na+ Na+ Cl- Cl- 01- It follows from the above considerations that the conductivity of a siib- stance in solution will depend on the degree to which it undergoes dissocia- tion. Furthermore, if we assume that in so far as osmotic pressure phenomena are concerned, each ion behaves in the same way as a mole- cule, then it follows that the electrical conductivity must be proportional to the extent to which the osmotic pressure is greater than we should ex- pect it to be from the amount of substance actually dissolved. In the Determination of the Conductivity it is obviously necessary to use standard conditions of depth and width of the fluid through which the current is passed, and to have some standard of comparison. The value is then known as the specific conductivity, the standard for comparison being the conductivity of a hypothetical liquid which, if enclosed in a centimeter cube, would offer a resistance of 1 ohm between two opposite sides of the cube acting as electrodes. The actual determination is usu- 18 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES ally made in a cylindrical vessel of hard glass (from soft glass enough alkali might be dissolved to affect the results), the electrodes being circu- lar plates of platinum firmly cemented at a known distance from each other (Fig. 5).* This conductivity cell, as it is called, is- connected with a suitable electric apparatus for measuring the resistance offered Fig. 5. — Diagram of conductivity cells. The platinum discs are represented by the thick black lines. They are held in position by thick-walled glass tubes, through which they are connected with the terminals by platinum wires. (From Spencer.) by the solution to the passage of an electric current (Wheatstone Bridge) (see Fig. 6). The resistance is of course inversely proportional to the conductivity. As a saline solution is progressively diluted, its specific conductivity naturally decreases (since there are now fewer molecules between the Fig. 6. — Wheatstone Bridge for the measurement of electric resistance: a-b, bridge wire; c, the movable contact. two opposite faces of the centimeter cube, and the space between ions or molecules is increased). This result will not, however, tell us whether the salt itself is undergoing any alteration in conducting power as a con- sequence, for example, of greater dissociation. To ascertain this we must *This distance is determined not by direct measurement but by calculation from results obtained by testing the actual resistance of a solution whose specific resistance is accurately known. ELECTRIC CONDUCTIVITY, DISSOCIATION, IONIZATION 19 obtain figures relating to the same quantity of salt at each dilution. If we multiply the specific conductivity by the volume of solution in c.c. which contains 1 gram-equivalent (see page 22), a value will be secured which represents the conducting power of a gram-equivalent. This is known as the equivalent or molecular conductivity* and is represented by the sign X. When it is determined for progressively diluted solutions, A gradually increases, indicating that the efficiency of the electrolyte itself as a conductor increases with dilution, because it dissociates more. The extent of this increase is found to become less and less as dilution proceeds. By plotting the values of the molecular conductivity of suc- cessive dilutions as a curve, the value at infinite dilution can be ascertained by extrapolation. This value is represented by A oc . Now, let us see how these facts bear out the theory of electrolytic dissocia- tion. According to this hypothesis the conductivity depends on the num- ber of ions (see page 17), and since it is at a maximum at infinite dilu- tion, the value A°c must represent the total number of ions that can be pro- duced by the dissociation of 1 gram-equivalent, and A that at some other dilution. If, therefore, we divide A by A 29 30 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PKOCESSES equal and in opposite directions to each other. On the other hand, should the concentration of the metallic ion in the solutions be unequal, the electromotive force will flow from the one electrode to the other, and the pressure with which it flows will be equal to the difference in con- centration of the two solutions. This is the principle of a concentration cell, and if we know the concentration of one of the solutions composing it, and then proceed to measure the electromotive force, we can obtain the concentrations of the other solution by difference. To do this we must employ a formula which takes into consideration the relation be- tween the potential and the concentration of the solution. The potential of an unknown electrode composed of a metal in con- tact with a solution of one of its salts- may also be determined by making it one pole of a battery of which the other pole is composed of a stand- ard electrode of unchanging known potential. An electrode of the latter Fig. 7. — Diagram to show type of electrodes used in studying electromotive force. The metal in each electrode is connected (through a glass tube) with a platinum wire, to which the apparatus for measurement of the voltage is connected. The metal dips into a solution contained in the electrode vessel and filling the side tube. The latter dips into an inter- mediate vessel containing saturated KC1 solution. The currents flow through the circuit under the following conditions: (1) dissimilar metals dipping into the same fluid; (2) similar metals dipping into different fluids; (3) dissimilar metals dipping into different fluids. type can most readily be made by bringing pure mercury in contact with a saturated solution of calomel (Hg2Cl2) in normal potassium chlo- ride solution. Under suitable conditions (i. e., when the circuit is com- pleted), a potential of +0.560 v. is developed in this so-called calomel electrode* — that is, positive ions of mercury are deposited on the mercury from the calomel solution, at this pressure. Suppose that we connect a calomel electrode, through the intermediation of some solution which *The calomel electrode consists of a suitably shaped glass vessel containing pure mercury, con- nected by means of a platinum wire with a conductor, and filled with a saturated solution of pure mercurous chloride in normal KC1 solution up to such a level that it also fills a side tube connected with a vessel containing a saturated solution of potassium chloride. Into this vessel also runs a similar side tube from the unknown electrode. By having an excess of utidissolved calomel in the solution in the calomel electrode its saturated condition is maintained during the chemical changes which accompany the production of the electric current. HYDROGEN-ION CONCENTRATION 31 will serve as a good conductor, with another electrode, the two elec- trodes being also connected by wires with electrical apparatus for measuring the total potential of the battery; then by adding +0.560 v. to or subtracting this value from the total potential (depending on the sign of the unknown electrode) we can tell the potential of the unknown electrode. We have discussed these principles of electrochemistry because they form the basis upon which depends the standard method for the deter- mination of the H-ion concentration of fluids. Suppose, for example, that in place of using a metal in the construction of one electrode, we use an electrode consisting of a layer of pure hydrogen gas in contact with a solution in which are free H ions; then the rate at which H ions Fig. 8. — Diagram of apparatus for the measurement of the H-ion concentration. The cur- rent generated in the battery (composed of calomel electrode, connecting vessel with KC1 solu- tion, and the H electrode) or that from the normal element is transmitted through a reversing key to the bridge wire, where the voltage is compared with a steady current flowing through the bridge wire from an accumulator. The capillary electrometer is used to detect the flow of current at various positions of the movable contact on the bridge wire. (Modified from Sorensen.) become added to the solution from the H layer, or taken from it, will de- pend on the concentration of H ions in solution. In order to secure a hydrogen electrode fulfilling the above requirements, it is necessary to employ some means by which a layer of hydrogen may be furnished, and fortunately this can be done by taking advantage of the property which spongy platinum possesses of absorbing large quantities of this gas. It is also necessary to keep an atmosphere of pure H in contact with the fluid. As is the case of the simpler cells described above, there are two types which we might use for measuring the electromotive force gen- erated in the unknown electrode: a concentration cell composed of two 32 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES hydrogen electrodes, of which one contains a solution of known H-ion concentration, and the other the solution in which this is unknown; and a cell of which one electrode is a standard calomel electrode and the other, a hydrogen electrode containing the unknown solution. The exact arrangement of the apparatus in which the calomel elec- trode is used will be seen in the accompanying sketch. The hydrogen electrode, it will be noticed, is a very small V-shaped tube, in which is suspended a platinum wire coated with spongy platinum and dipping into a solution which nearly fills the tube. The space above the solution is filled with pure hydrogen. This and the calomel electrode are con- nected with suitable electric measuring instruments, and the circuit is completed by connecting the two electrodes by means of an intermediate vessel containing a saturated solution of potassium chloride. This con- necting solution is used because it has been found that the electric cur- rents set up at the contact between different solutions are so small that they can be disregarded.* As outlined above, the hydrogen electrode is that which is used to determine the H-ion concentration of blood, the particular point about it, in comparison with the apparatus used for simpler solutions, being that the hydrogen is not changed in the course of the experiment. This precaution is to prevent the carbon dioxide of the blood from being "washed out" of it by a frequently changing atmosphere of hydrogen. Many inaccuracies in the earlier results obtained by this method were due to the removal of carbon dioxide, which, as we shall see later, is one of the chief acids contributing to the H-ion concentration of blood. The Indicator Method As pointed out in a previous chapter (page 22), the method of titra- tion for acidity or alkalinity in which a standard solution of alkali or acid is added until a certain change in the color of a suitable indicator is detected, does not afford any information regarding the H-ion con- centration actually present in the solution. It tells us the total con- centration of available acid or base, both dissociated and undissociated. By modification of the method of procedure, however, we may also use indicators for determining the H-ion concentration. The principle of this method depends on the fact that there are certain dyes which change quite distinctly in tint with very slight changes in the H-ion concentration, so that if we use dyes which possess this property at a point near that of neutrality (i. e., between PH6.5 and Pn8), we can es- *A description of the technic for measuring the electric potential developed by the cell would be out of place here. Suffice to say that the strength of the current is compared with that of a current of known strength furnished by a normal cell, the comparison being made by a bridge wire F, a capillary electrometer II being employed to detect the direction and degree of current. HYDROGEN-ION CONCENTRATION 33 timate the H-ion concentration of the body fluids with very remarkable accuracy, provided certain precautions are taken to circumvent the disturbing influence which the protein and salts in these fluids may have on the color change. To understand this use of indicators, it is important to bear in mind that one solution reacting neutral to one indicator may have a H-ion concentration which differs very markedly from that of another solu- tion reacting neutral to another indicator. This is because indicators react to different H-ion concentrations. A solution that is neutral to phenolphthalein has a PH of about 9, whereas one neutral to methyl or- ange has a PH of about 4. This can be very clearly shown by titrating a solution of phosphoric acid with decinormal alkali. After a certain amount of alkali has been added it will be noticed that methyl orange changes from red to yellow, but after it has changed and is therefore alkaline as judged by this indicator, it still remains distinctly acid to- wards phenolphthalein (shows no red tint) even though considerably more alkali is added. The methyl orange is, therefore, itself unrespon- sive to weak acids such as remain after the greater part of the phos- phoric acid has been neutralized by the alkali. The series of indicators which has been employed for this purpose is given in the accompanying table, along Avith the PH limits through which they change in color. LIST OF INDICATORS CHEMICAL NAME COMMON NAME CONCEN- TRATION COLOR CHANGE RANGE PH Thymol sulfon phthalein per cent (acid range) Thymol blue 0.04 Red-yellow 1.2-2.8 Tetra bromo phenol sul- fon phthalein Brom phenol blue 0.04 Yellow-blue 3.0-4.6 Ortho carboxy benzene azo di methyl aniline Methyl red 0.02 Red-yellow 4.4-6.0 Ortho carboxy benzene azo di propyl aniline Propyl red 0.02 Red-yellow 4.8-6.4 Di bromo ortho cresol sulfon phthalein Brom cresol purple 0.04 Yellow- purple 5.2-6.8 Di bromq thymol sulfon Brom thymol phthalein blue 0.04 Yellow-blue 6.0-7.6 Phenol sulfon phthalein Phenol red 0.02 Yellow-red 6.8-8.4 Ortho cresol sulfon phthalein Cresol red 0.02 Yellow-red 7.2-8.8 Thymol sulfon phthalein Thymol blue 0.04 Yellow-red 8.0-9.6 (see above) Ortho cresol phthalein Cresol phthalein 0.02 Colorless-red 8.2-9.8 These dyes may now be obtained in this country. (W. M. dark and H. A. Lubs.)» 34 PHYSICOCHliMICAL BASIS OF PHYSIOLOGICAL PROCESSED Briefly stated the method for measuring the H-ion concentration Con- sists in preparing a series of solutions containing known concentrations of H-ion — that is to say, of known PH — and adding to each solution an equal amount of an indicator which exhibits easily distinguishable changes in tint at H-ion concentrations approximating those believed to be present in the unknown solution. The same indicator is added to the unknown solution, which is then placed side by side with the stand- ards to find with which of them it most closely matches. The series of solutions of known H-ion concentration is prepared by mixing fif- teenth normal solutions of Na2HP04 and KH2P04 in varying propor- tions as given in the following table: PREPARATION OF STANDARD SOLUTIONS The solutions are mixed in the proportions indicated below to obtain the desired PH:* PH 6.4 6.6 6.8 7.0 7 1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 8.0 8.2 8.4 Primary Potas. 73 63 Phos., c.c. Secondary Sodium 27 37 Phos., c.c. 51 49 37 63 32 68 27 73 23 77 19 81 15.8 84.2 13.2 11 86.8 89 8.8 91.2 5.6 94.4 3.2 2.0 96.8 98.0 (From Levy, Rowntree and Marriott.) 'Standard phosphate mixtures are prepared according to Sorensen's directions as follows: 1/15 mol. acid or primary potassium phosphate.- — 9.078 grams of the pure recrystallized salt (KHoPCU) are dissolved in freshly distilled water and made up to 1 liter. 1/15 mol. alkaline or secondary sodium phosphate. — The pure recrystallized salt (Na2HPO4. 12H2O) is exposed to the air for from ten days to two weeks, protected from dust. Ten molecules of water of crystallization are given off and a salt of the formulnl Na«HPO4 .2112O is obtained; 11.876 grams of this are dissolved in freshly distilled water and made up to 1 liter. The solution should give a deep rose red color with phenolphthalein. If only a faint pink color is obtained, the salt is not sufficiently pure. The indicator method is extremely accurate when used with 'pure solutions of acids, but, as mentioned above, it is apt to be inaccurate, at least with most indicators, when protein or inorganic salts are pres- ent in the solution, and of course it is quite unusable with colored fluids such as blood. In order to overcome these difficulties, the dialysis method has recently been evolved. It consists in placing the fluid — blood, for example — in a dialyser sac composed of celloidin and about as large as a small test tube. The sac is placed in a wider test tube of hard glass containing an isotonic solution of sodium chloride that has been carefully tested to ascertain that it is strictly neutral. The amount of blood or serum required for this method is only 2 or 3 c.c., and the amount of salt solution placed outside the sac should be about the same. It takes only from five to ten minutes for dialysis to occur. The celloidin sac is then removed, a few drops of the indicator are thoroughly mixed with the dialysate, and the tube compared with the series of standards until the corresponding tint is matched. This indicates the H-ion concentration in the dialysate. The tints produced by using sulphonephenolphthalein are reproduced as nearly as possible PH7-o PH7-/ PH7-2 PH7-3 10 PH7-5 PH7-6 Fig. 9. — Chart showing approximately the tints produced by adding sulphophenolphthalein to a series of phosphate solutions of the H-ion concentrations indicated in each case by PH. HYDROGEN-ION CONCENTRATION 35 in the accompanying chart. The H-ion concentration of the unknown solution is that of the tint with which it matches in the series. It might be thought that this method would be inaccurate because of the loss of carbon dioxide from the blood. By actual experiment, how- ever, it has been found that, if the blood is collected with certain pre- cautions, the error is negligible. The method is, therefore, a most useful one clinically. The following table gives the hydrogen-ion concentration or true reaction of the body fluids. FLUID PH FLUID PH Blood 7.4 Muscle juice (fresh) 6.8 Urine 6.0 Muscle juice (autolyzed) Variable Saliva 6.9 Pancreas extract 5.6 Gastric juice (adult) 0.9-1.6 Peritoneal fluid 7.4 Gastric juice (infant) 5.0 Pericardial fluid 7.4 Pancreatic juice (dog) 8.3 Aqueous humor 7.1 Small intestinal contents 8.3 Vitreous humor 7.0 Small intestinal contents (infant) 3.1 Cerebrospinal fluid 7.2 Bile from liver 7.8 Cerebrospinal fluid 8.3 Bile from gall bladder 5.3-7.4 Amniotic fluid 7.1 Perspiration 7.1 Amniotic fluid 8.1 Perspiration 4.5 Milk (human) 7.0-7.2 Tears 7.2 Milk (cow) 6.6-6.8 Milk (goat) 6.6 Milk (ass) 7.6 (W. M. Clark and H. A. Lubs.) CHAPTER VI THE REGULATION OF NEUTRALITY IN THE ANIMAL BOD.Y AND ACIDOSIS Nothing is more constant in the animal economy than the H-ion con- centration (On) of the fluids which bathe the tissues. This regulation is fundamentally of a physicochemical nature, depending on the inter- action of alkalies with acids, of which carbonic and phosphoric acids and the proteins are the most important.* When different amounts of acids or alkalies are added to water, the range of variation in H ion is very extensive, whereas in blood the range is very limited indeed, not extending beyond PH7 and PH7.52 (i. e., CH never goes above that of a 0.000,000,1 N solution or below that of a 0.000,000,03 N solution). In other words blood can withstand considerable additions of acid or al- kali without much change. Buffer Substances. — The chemical reactions upon which this remark- able constancy in reaction depends have been explained by Lawrence J. Henderson.10 The fundamental equations are as follows: M,HPO4 + HA — MH2PO4 4 MA, and MHC03 4- HA = H2CO, 4- MA, when M = a basic radicle, and A, an acid radicle. Now it has been discovered that weak acids, like carbonic and phos- phoric, possess the remarkable property of maintaining the reaction constant when they are present in a solution which also contains an excess of their salts. Under these circumstances the concentration of ionized hydrogen is almost exactly equal to the product of the dissocia- tion constantt of the acid (see page 10) multiplied by the ratio be- tween free acid and salt; in other words, K V[HA] 7K X[BA]' If carbonic acid is present in a solution of bicarbonates so that there 'According to circumstances, proteins may act either as acids or as alkalies. They are there- fore called amphoteric. tThe ionization constant has already been referred to as a figure which expresses the tendency of a weak acid or base to dissociate in an aqueous solution. "It expresses the proportion in which the nondissociated part is capable of existing in the presence of its ions," and therefore is a gauge of the strength. The dissociation constant amounts to about 0.000,000,5 for carbonic acid ; that is, the dissociation of HoCOs into H'-f-HCQg' at room temperature will be such that the concentra- tion of H-ion equals a 0.000,000,5 N solution. 36 ACIDOSIS 37 are equivalent quantities of free H2C03 and bicarbonate — i. e., r^.-. =~r 1-bAj l — the H-ion concentration will be exactly the same as the dissociation constant of carbonic acid; therefore 0.000,000,5 N (PH = 6.31), or about five times the value of neutrality, 0.000,000,1 N (PH = 7.31). If ten times as much free carbonic acid as bicarbonate is present, then the H-ion concentration will be fifty times that of neutrality, i. e., •,_ . ,- =^- x 0.000,000,5 = 0.000,005 (PH = 5.31); if there is ten times less carbonic acid than bicarbonate, the H-ion concentration will be one-half that of neutrality, i. e., [!*fj = -^ * 0.000,000,5 = 0.000,000,05) (PH = 7.31) ; or L-t£.£\.J .LU if twenty times less, one fourth (PH = 7.6). Since a large amount of bicarbonate is actually present in blood (enough to yield from 50 to 65 c.c. C02 per 100 c.c. of blood) (see page 391), and the free carbonic acid undergoes fluctuations which are only trivial when compared with those which have been chosen in the above examples, it is clear that there must be very little change in the H-ion concentration of the blood in comparison with the variations which would occur were no bicarbonate present. Another weak acid which acts like carbonic in maintaining neutral- ity is acid phosphate (MH2P04), and for the same reason — namely, that its dissociation constant is of similar magnitude to the H-ion concen- tration. Although the blood plasma itself contains much less phosphate than bicarbonate, the tissues contain a considerable amount, which en- ables them to maintain their neutrality. This action of bicarbonates and phosphates is styled the buffer action, meaning that it serves to damp down the effect on the H-ion concentration which additions of acids or alkalies would otherwise have. As pointed out by Bayliss, however, a better word to use would be "tampon action," since the substances actually soak up much of. the added H- or OH' ions. It is not confined to the fluids of the higher animals, but is very widely distributed throughout nature ; for example, in the ocean and in the fluids of marine organisms and animalcules (see L. J. Henderson).11 Although the actual reaction by which neutrality is maintained is purely of a physicochemical nature, some provision must obviously be made so that the acid and basic substances that take part in it may be supplied and those produced by the reactions removed as occasion re- quires. The source of supply is partly exogenous and partly endogenous. The exogenous source is the basic and acid substances present in the food; and although we do not ordinarily attempt to control the amounts of these substances ingested, we may do so, as, for example, by the persistent administration of soda in cases of pathologic acidosis. The endogenous source depends on the constant production in metabolism 38 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES of acids such as carbonic, phosphoric, lactic, and sulphuric, and of alkalies such as ammonia. Amphoteric substances, like amino acids and proteins, may functionate either as acids or as alkalies. Whatever may be its source, a considerable reserve of alkali is undoubtedly available in the animal organism. This required store of alkali appears to be automatically liberated as a result of the physicochemical process. The removal is affected by three pathways: (1) through the lungs gaseous carbonic acid is eliminated; (2) through the kidneys, the fixed acids; and (3) through the intestines, some of the phosphoric acid. Carbonic acid is produced in large amounts in the normal process of metabolism, and is excreted in a gaseous condition by the lungs. Varia- tion in its excretion is the most important mechanism for controlling temporary changes in CH. In order to make this clear, it may be well to revert for a moment to the physicochemical equation by which carbonic acid is enabled to maintain neutrality. This may be written: CH = TT pQ molecular ratio 2 rr(~ • . The ratio may be increased either by adding NaMCL^ free carbonic acid to the blood (as by causing an animal to respire some of the gas), or by the addition of some other acid (e. g., oxybutyric, as in diabetes) which will decompose some of the NaHC03 and produce H2C03. The increase which these changes would cause in CH of the blood is prevented by the remarkable sensitivity of the respiratory cen- ter to changes in CH. An increase which is much less than can be measured by physicochemical means stimulates the center, causing in- creased pulmonary ventilation, so that the carbonic acid is immediately eliminated through the lungs. This elimination does not stop when the old level of carbonic-acid concentration is reached, but proceeds until TT r an(^ ^H consequently become greater, the NaHLOg condition is called uncompensated acidosis, but if the center does respond so that CH is held constant (although NaHC03 is decreased), the condition is one of compensated acidosis. For practical reasons, therefore, the study of pathologic acidosis de- pends on an estimation of the bicarbonate content of the blood or, since it is simpler to carry out and is of equal value, of the plasma. When plasma is obtained by removing blood from a vein of the arm and cen- trifuging immediately out of contact with air (so that C02 may not be lost from it) it contains approximately 60 vols. per cent of C02. Since we know that the partial pressure of C02 in blood is equal to 42 mm. Hg (ascertained from determinations of the alveolar C02) (see page 344), we can calculate how much of the 60 vols. per cent must be in simple solution by application of the law of solution of gas in a liquid (page 336). It has been found that plasma at body temperature and at 760 mm. Hg (atmospheric pressure) dissolves 0.54 per cent C02, so that at 42 42 mm. it will dissolve _pr. x 100 x 0.54 = 3 vols. per cent. Transcribing 7bO [H2C03] 3 1 the figures to our equation we get [NaHC03] 60 20 This definition of acidosis leaves out of regard all conditions that may TT /-1Q raise the ratio 2 3 by the addition of H2C03 without decomposing any of the NaHC03, such, for example, as occurs Avhen an excess of free carbonic acid is present in the blood plasma. Since increases in free C02 are not infrequent in both health and disease — e. g., asphyxial con- ditions — the above definition is not sufficiently comprehensive. When we come to study the control of the respiratory center, we shall see that TT pQ an increase in the ratio — 2^pof sufficient magnitude to cause an IN clii-vy v/3 actual increase in CH can be produced by causing an animal to respire air *This agrees sufficiently with the result as calculated from the known values of the equation N HCO~ ~ ~TC~ ' Thus, if we take CH as 0.35 xlO-7, \ as 0.605 for blood conditions, and L H2CO3 0.605 x 0.35 x 10-7 _ 1 fc as 4.4 x 10- (M.chaehs and Rona),, we get = 4.4 x IQ-T = J\ 40 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES containing an excess of C02 — a true acidosis, but one for which no place is found in the above definition. Nevertheless, Van Slyke's definition has a real value, because it em- phasizes the importance of a determination of the bicarbonate as a cri- terion of the degree of the forms of acidosis usually met with in disease. The bicarbonate under such conditions may become reduced either be- cause of the appearance of improperly oxidized fatty acids, like /?-oxy- butyric and acetoacetic, when carbohydrate metabolism is upset as in diabetes or starvation, or because the acids produced by a normal metabolism are inadequately eliminated by the kidneys, as in nephritis. Accordingly, if the respiratory mechanism and increased mass move- ment of the blood (for an increase in CH accelerates this also) should TT C*(\ fail to eliminate C02 quickly enough so as to keep the -A-nrf ratio at JNa-ri.L'U3 one twentieth, then CH will rise. This is not likely to happen until a large part of the NaHC03 has been used up, so that an estimation of that actually present must be a reliable index of the proximity to this condition. A sustained increase in CH is incompatible with life. The NaHC03 is the buffer, the factor of safety which prevents its occurrence. Although it is only in arterial blood (i. e., after elimination of excess of C02 by TT r<(\ the lungs has been accomplished) that constancy in the ratio can be expected, it is fortunate, for practical reasons, that venous blood collected during muscular rest and without stasis should be only slightly different. When acids are added to the blood, they will first of all be neutralized by the "buffers" of the plasma — namely, NaHC03 and protein, as we have seen. But this is only the first line of defense against acidosis, for buffer substances present in the corpuscles may also be used. This intra- corpuscular reserve of alkali is mobilized partly by transference of K and Na from corpuscle to plasma, but mainly by that of HC1 from the plasma into the corpuscle, so releasing base in the former to combine with the added acid (e. g., H2C03), according to the equation: H2C03 + NaCl ?=> NaHC03 + HC1. The HC1 on entering the corpuscle reacts with phosphates according to the equation: HC1 + Na2HP04 ?± NaH2P04 + NaCl. This is a particularly important detail of the buffer action of the blood, for it shows us how the phosphates of the corpuscles are rendered available for neutralizing acids added to the plasma, where there are practically no phosphates. Indeed the transference of acid through the corpuscular envelope indicates that the same sort of thing must go on with the other cells of the body, so that the plasma, itself rather poor in buffer substances, has all those of the body at its disposal. ACIDOSIS 41 THE MEASUREMENT OF THE RESERVE ALKALINITY 1. Titration Methods There are several methods by which the reserve alkalinity of the blood may be measured. The simplest in theory consists in seeing how much standard acid must be added to a measured quantity of blood plasma in order to reach the neutral point as judged by change in tint of some indicator. The indicators employed (e. g., methyl orange) are such as change their tints at H-ion concentrations that are well to the acid side of neutrality (i. e., at a high CH or low PH). To bring the plasma to this point of neutrality the added alkali will need to neutralize, not only the bicarbonate of the plasma, but other acid-binding substances as well. This will give us a false impression of the acid-binding powers of the plasma, since, at the normal CH of the blood, proteins do not absorb acids to anything like the extent they do at higher degrees of CH. Another objection to the method is that the proteins interfere with the sensitive- ness of the indicators. The objections can be removed by determining the end point electro- metrically or by indicators that change tint at about PH7. The most practical way is to determine the change in CH produced by adding a known volume of standard acid to blood plasma. The resulting change in CH will then be greater the less the alkaline reserve. In the electro- metric method irregularities that might be caused by variable amounts of carbonic acid in the blood to start with are best controlled by removing the C02 from the plasma after adding the standard acid. The procedure therefore consists in mixing 1 c.c. plasma with 2 c.c. N/50 HC1 in a small separating funnel, which is then evacuated so as to remove the C02, after which the fluid is transferred to a hydrogen electrode and CH measured (see page 29). In normal blood this should be 10 5-6 (PH5.6). In acidosis, where there is a depleted alkaline reserve, the 2 c.c. of acid will cause a much greater change in CH — in diabetic blood to below 5 or lower. The technic involved in the above method is, however, too exacting for routine clinical work. For such purposes the colorimetric method of Levy and Rowntree may be employed. THE METHOD OF LEVY AND ROWNTREE.IS — A test tube made of hard ("nonsol") glass of about 20 c.c. capacity, containing about a gram of powdered neutral potassium oxalate, is filled with newly drawn blood, immediately stoppered and placed on ice. Quantities of 2 c.c. each of the blood are then placed in a series of seven small (nonsol) test tubes and allowed to stand for five to six minutes in order to permit a narrow 42 PHYSICOCHSMICAL BASIS OF PHYSIOLOGICAL PROCESSES layer of plasma to separate on the surface (this prevents laking of the blood during the subsequent addition of acid or alkali). The blood in the first tube is used for the determination of the normal H-ion. In each of the next three tubes are added respectively 0.1, 0.2 and 0.3 c.c. N/50 HC1, and to the last three, similar quantities of N/50 NaOH. After inverting the tubes so as to mix the contents, the blood in each is trans- ferred to celloidin sacs and the CH determined according to the method described elsewhere (page 32). The tubes are noted in which a change in tint from that of the normal blood is evident, and the results are expressed as the c.c. of N/50 HC1 or NaOH which must be added to blood to change its CH. Thus, the alkali buffer is the c.c. of N/50 NaOH which can be added to 2 c.c. of blood without change of CH of the dialysate, and the acid buffer the c.c. of N/50 HC1. The method suffers from the following drawbacks: 1. Very small quantities of acid and alkali are employed. 2. It is often difficult to tell just exactly when a slight difference in tint has been produced. 3. Even with the precautions described above, it is impossible to be sure that the amount of C02 in the different samples of blood is the same, which means, of course, that some bloods will, on this account alone, be able to bind more alkali than others. THE METHOD OF VAN SLYKE. — A method based on somewhat the same principle, but which is more accurate because it meets the above objec- tion, is that suggested by Van Slyke, Stillman and Cullen.14 Plasma is freed of CO2 by placing it in a vacuum, and, is then mixed w7ith an equal volume of N/50 HC1 (or NaOH) and the CH determined by the electric method (see page 29). In the case of normal blood, after such an addi- tion of acid, a practically normal CH will be found, whereas in the blood of cases of acidosis it will be very distinctly increased (i. e., PH lower). 2. C02-combining Power The above objections to the titration of blood plasma or dialysate with standard solutions of acids are removed if we measure the com- bining power of the blood alkali towards carbonic acid itself at normal blood reaction. This may be done either in blood immediately after its removal from the animal or in blood that has been first of all saturated outside the body with carbonic acid at a partial pressure equal to that existing in the body. Since for practical reasons venous blood must be used — in the clinic at least— the former of these methods suffers from the fault that varying amounts of carbonic acid will be added to the blood during its passage through the tissues, and the error thereby ACIDOSIS 43 incurred will become greatly aggravated if venous stasis has been pro- duced in drawing the specimen for analysis. But the chief reason why this method has not been extensively employed, as pointed out by Van Slyke, is the technical difficulty of making the necessary analysis. It is most satisfactory to collect venous blood after a period (one hour at least) of muscular rest (so that there is no excess of C02) and without venous stasis, and to centrifuge without permitting any considerable loss of carbonic acid. The latter precaution is necessary because there is a migration of acid radicles, e. g., HC1, from plasma into corpuscles when the C02 of the former is increased, and in the reverse direction when the C02 is decreased. If the C02 in the blood were not the same during cen- trifuging as it is in the body, the separate plasma would not contain the same amount of alkali — i. e., its reserve alkalinity would be altered. Although theoretically, therefore, centrifuging should be performed in Fig. 10. — Diagram of apparatus for saturating blood or plasma with expired air. The glass beads in the bottle condense excess of moisture. The separating funnel, as soon as it has been filled with expired air, should be closed by a stopper and the stopcock turned off. It is then rotated so that the blood forms a film on its walls. an atmosphere containing the same partial pressure of C02 as exists in the body (i. e., the alveolar air) (see page 344), this has been found im- practicable for general use, and is unnecessary if loss of C02 from the specimen of blood is prevented by allowing it to flow into the syringe very slowly (without any suction). It is mixed in the syringe with powdered (neutral) potassium oxalate (enough to make a 1 per cent solution with the blood), and immediately delivered into a centrifuge tube under paraffin oil, which by floating on its surface serves to diminish free diffusion of C02 to the outside air (even though such oils dissolve more C02 than water). To mix the blood with the oxalate, the syringe should be moved backward and forward several times, but it must not b*1 shaken. After centrifuging, about 3 c.c. of plasma are removed and saturated with C0? at the same tension as in alveolar air (i. e., 5.5 per cent). 44 PHYSICOCHKMICAL BASIS OF PHYSIOLOGICAL PROCESSES is done by placing the plasma in a separating funnel of 300 c.c. capacity, laying the funnel on its side and displacing the air in it by alveolar air secured by quickly making as deep an inspiration as possible through the tube and bottle containing glass beads (Fig. 10). The glass beads remove excess of water vapor from the air. The funnel must be restop- pered before the end of the expiration, so that no outside air enters. It is then rotated, for about two minutes, in such a way that the plasma forms a film on its walls. If it is necessary to postpone the saturating of the plasma, this should be pipetted off from the corpuscles and pre- served in hard glass test tubes coated with paraffin. From ordinary glass Fig. 11. — Van Slyke's apparatus for measuring the COa-cotnbining power of blood in blood plasma. For description, see context. enough alkali is soon dissolved out to vitiate the results. After saturation of the plasma with C02, the funnel is placed in the upright position and the plasma allowed to collect in the narrow portion, after which 1 c.c. is removed with an accurate pipette and analyzed for C02. The analysis may be done by using either the Van Slyke or the Hal- dane-Barcroft apparatus. The Van Slyke method is as follows: The apparatus is filled to the top of the graduated tube with mercury (Fig. 11) by raising the mercury reservoir F, care being taken that D and E are also filled. One c.c. of the CO2-saturated plasma is then de- ACIDOSIS 45 livered into A (which has been rinsed out with C02-free ammonia water), and the stopcock / turned so that by cautiously lowering the level of the reservoir F, the plasma runs into B (but no trace of air). The same procedure is repeated with 1 c.c. water, so as to wash in all of the plasma, and finally 0.5 c.c of 5 per cent H2S04 is sucked in, after which stopcock 7 is turned off. The reservoir F is then lowered sufficiently to allow all of the mercury, but none of the blood, to run out of B and C. A vacuum is thus produced in B and C. As the level of the mercury falls in B and C, the plasma effervesces vio- lently,* because it is exposed to a vacuum. To be certain that all traces of C02 have been dislodged from the solution, the apparatus is inverted several times. To ascertain how much C02 has been liberated, stopcock // is now turned so as to bring C and E into communication, and by cautiously lowering the reservoir the fluid in C is allowed to run into the bulb E. Stopcock II is thereafter turned so as to connect C and D, and the reser- voir raised so that the mercury runs into C as far as the C02 that has col- lected in the burette will permit it to go. After bringing the level of the mercury in F to correspond to that in the burette, the graduation at which this stands is read. It gives the c.c. of C02 liberated from the plasma. Under the above conditions normal plasma binds about 75 per cent of its' volume of C02 ; therefore, since the total capacity of the pipette is 50 c.c., the mercury should stand at 0.375 c.c. on the burette. For accurate measurement it is necessary to allow for the C02 that remains dissolved in the water, etc., as well as for barometric pressure and temperature. This is best done by the use of a table based on the known solubility of C02 under the various conditions obtaining, which is given in Van Slyke's paper.12 The Haldane-Barcroft apparatus that is most suitable for the above analysis is shown in Fig. 136, page 382. t One c.c. of C02-free ammonia water is placed in the bottle and the 1 c.c. of plasma delivered beneath it. *This may be prevented by adding a small drop of caprylic alcohol. tThis form of Haldane-Barcroft apparatus is not quite the same as the differential manometer that is used for measurement of the (^-combining power of hemoglobin (page 382). In the form used for the present purpose, a side tube at the bend of the U-tube is connected with a small rub- ber bag, which Ann be compressed by a screw. When the gas is evolved in the bottle, it presses down the fluid in the proximal limb of the manometer correspondingly and raises that in the distal limb. Since the calculation of the amount of gas evolved depends on finding the pressure produced without any change in volume, it is necessary after the gas has been evolved to compress the rubber bag until the meniscus of fluid in the proximal limb of the manometer is brought back to its original level. The height at which the fluid stands in the distal limb then obviously corresponds to the pressure created by the evolved gas. The equation for determining the amount of gas evolved depends on the gas law, which states that the pressure of a gas is inversely proportional to its volume (page 336). Suppose that the volume of gas evolved was equal to the volume of the bottle, then, since the volume has been kept constant, the pressure would be doubled — that is, the fluid in the distal limb would equal that of 1 atmosphere, or 10,400 mm. of water or 10,000 of clove oil, which is the fluid actually used to fill the manometer. Any other observed pressure would therefore correspond to the volume of evolved gas according to the equation, vol. of bottle (and tubing to meniscus) 10,000 (when clove oil is used) _ In using the apparatus in the above manner, only one of the bottles is employed, and the tartaric acid is added from a pocket in the stopper by a simple manipulation. 46 PHYSICOCH^MlCAL BASIS OF PHYSIOLOGICAL PROCESSES The bottle is then connected with the manometer with the precautions described elsewhere in this volume. "When temperature conditions have been alloAved for, saturated tartaric acid is mixed with the plasma solu- tion and the gas evolved measured by the displacement of the fluid in the manometer. The apparatus may also be used with blood in place of plasma. In this case, however, it is necessary that the oxygen be removed before adding the tartaric acid. This precaution is necessary, since acid can dislodge some of the 02 from hemoglobin. The blood is therefore first of all laked with ammonia containing some saponin, then shaken with 0.25 c.c. saturated potassium ferricyanide solution, and finally with the saturated acid solution. If blood is used, the estimations must be made on strictly fresh blood, since on standing the C02-combining power greatly deteriorates. 3. Indirect Methods There are several other methods by which the alkaline reserve may be measured. These include: 1. Determination of the Tension of C02 in Alveolar Air (page 344).— Since this method is employed more particularly in investigating the hormone control of the respiratory center, we shall defer a description of it until later. The alveolar C02 tension corresponds to the C02 ten- sion in arterial blood and this is proportional to the alkaline reserve as determined by Van Slyke's method as is proved by the fact that the ratio, plasma C02 . , . ,, ., . — -.. : — , is satisfactorily constant. alveolar C02 tension 2. The Measurement of the Acid Excretion by the Kidney. — As might be expected, the acid-base equilibrium of the body may also be gauged by measurement of the acid excretion of the urine, in which the acids are contained partly in combination with ammonia or a fixed base, and partly in a free state. We shall first of all consider the methods 'of acid excretion and then examine the evidence showing that the total acid excretion is proportional to the alkaline reserve as measured by the above described methods. EXCRETION OF ACID. IN COMBINATION WITH AMMONIA. — The production of ammonia is essentially an endogenous process, and when excessive quantities of acid make their appearance in the organism, the fixed alkali may not be sufficient to neutralize it all, so that ammonia, derived from the breakdown of amino acids (page 616), instead of being converted into urea is employed to neutralize the excess of acid. Most workers have in this way explained the very large ammonia excretion that has long been known to occur in such conditions as diabetic acidosis. Some recent workers are, however, inclined to question the significance of ammonia in this connection, believing that the increased ammonia ex- cretion is, like the acetone bodies themselves, a product of perverted metabolism. Be this as it may, it is no doubt true that ammonia is used for neutralizing acid in disease, although it may not be an important factor in the maintenance of neutrality under normal conditions. It is a factor of safety, in that it helps to care for an increase in acid when the normal mechanism of the body is overtaxed. EXCRETION OF PHOSPHATES. — The more permanent control of neutrality depends on the excretion of phosphates by the kidney. The principle governing this process is exactly the same as that already discussed in connection with carbonic acid. In the one case it is the volatile acid C02, and in the other, the fixed phosphoric acid that is concerned in the reaction. The ratio between the acid salts of phosphoric acid, MH2P04, and the alkaline salts, M2HP04, in blood is approximately 1 to 5, but in the urine this ratio varies according to the amount of H ion that must be eliminated from the blood. In other words, a definite amount of phos- phoric acid is enabled to carry variable amounts of H ion out of the body by causing the amount of alkali excreted in combination with it to be- come altered. For example, in the form of MH2P04 a given amount of P04 obviously carries out more H ion than when it is excreted as M2HP04. The adjustment between these two salts is a function of the kidney. We may accordingly measure the amount of alkali retained by the organism by finding how much standardized alkali must be added to a given quantity of urine until the reaction of the blood is obtained. Since the latter value is constant, the titration can be done simply by titrating the urine with an indicator such as sulphonephenolphthalein, which changes tint at about PH of blood. A more serviceable indicator to use, however, is phenolphthalein, be- cause its end point is such that when human urine just reacts neutral to it — that is, when the titrable acid approaches zero — the C02-absorb- ing power of the plasma is at its maximum of 80 vols. per cent and the ammonia excretion by the urine is zero (Van Slyke). It is advantageous, therefore, to use this indicator, because it happens to have its turning point situated for a reaction which is well to the alkaline side of neu- trality, and which is reached in urine when the blood is at its maximal acid-combining power and no ammonia is being used for neutralization purposes. As the C02-combining power of the blood decreases, there should, therefore, be a proportionate increase in ammonia and in the titrable acidity of the urine. Although a general parallelism exists between these values in cases of diabetes, etc., there is no strict proportionality. The expedient has therefore been tried of comparing the alkaline reserve of the blood with 48 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES the excretion rate of acid as determined by an application of Ambard's equation for chlorides and urea, and with curiously satisfactory results (Fitz and Van Slyke). This equation is: Blood concentration = constant x V^VC; where D is the excretion rate, W the body weight, and C the concentration of excretory prod- uct in the urine. For the present purpose D is therefore the number of c.c. of N/10 alkali (or acid) required to bring the urine to the neutral point of phenolphthalein plus the NH3 expressed as c.c. of an N/10 solution, for the twenty-four hours, and C is c.c. of N/10 alkali and of N/10 NH3 per liter of urine. If we assume that the acid accumulation in the blood is proportional to the fall of the plasma C02 figure below the maximal figure of 80, the above equation becomes: / L) _ Retained acid = 80 - plasma C02 = constant x y \\fVC. For practical purposes it is best to make the necessary analysis on a sample of urine collected over a period of one to four hours, and to col- lect the blood for determination of its reserve alkalinity in the middle of this period. The twenty-four-hour rate of excretion is then computed (D) from the analysis. The value calculated by the above equation has been found to agree with that of the C02-combining power of the plasma to within 10 vol- umes per cent, except when bicarbonate is being taken by the person, when the blood bicarbonate is much higher than indicated by the urine. 3. Determination of Alkali Retention. — Another valuable criterion of the alkaline reserve is the amount of alkali required to change the re- action of the urine. In health the CH of the urine varies from 0.000,016 N (PH = 4.8) to about 0.000,000,035 N (PH = 7.46) with a mean of about 0.000,001 N (PH = 6). These extremes are rarely overstepped in disease, but frequently the average is considerably different. In car- dio-renal disease, for example, the mean acidity may be approximately 0.000,005 N (PH = 5.3), or five times the normal value. A certain de- gree of acidosis is therefore common enough in this condition — a fact which has indicated the advisability of administering sodium bicarbon- ate. It has been found that 5 grams or less of soda, given by mouth to a normal person, causes a distinct diminution in the CH of the urine, whereas in pathologic cases it may be necessary to give more than 100 grams before a similar effect is observed (L. J. Henderson and Palmer15 and Sellards16). For this very large holding back of alkali, the organism and not the kidney is responsible. This is indicated by the fact that, when the administration of alkali is discontinued, the acidity of the urine soon ACIDOSIS 49 regains its old level, although now if a smaller dose of alkali is given, the CH of the urine will immediately be lowered. These facts indicate that for the moderate degrees of acidosis common in chronic disease, the properly controlled administration of soda is very probably a most advan- tageous treatment. CHAPTER VII COLLOIDS Substances which can be obtained in the crystalline state and which, when "in solution, are capable of readily diffusing through membranes, are designated as crystalloids, and are to be distinguished from another, larger group of substances not having these characteristics or having them only in very minor degree — the colloids. In every field of chem- istry the properties of colloids have been studied extensively during recent years, but in no field more than in that which covers the chem- istry of biological fluids and tissues, into whose composition colloids enter much more extensively than crystalloids. The subject of colloidal chemistry has indeed become so extensive that an attempt to do more than indicate some of the most important characteristics of colloids would take us far beyond the limitations of this book. The far-reaching applications of the subject in physiology and medicine are only begin- ning to be realized. The term "colloid," or "colloidal," does not refer to a class of chemical substances, but rather to a state of matter which is quite independent of the chemical composition of the substance. We are familiar with more colloids in the organic than in the inorganic world, yet they are plentiful in both, and the same substance may at one time be colloidal and at another noncolloidal. Indeed, under appropriate conditions prob- ably all substances may assume the colloidal state — not solids and liq- uids alone, but gases as well. It is mainly with liquids, however, that we are concerned in biochemistry. CHARACTERISTIC PROPERTIES The distinction between molecular* and colloidal solutions is a rela- tive one. Suppose, for example, that we take a piece of gold in water and divide it up into smaller and smaller parts. At a certain stage, the particles will be so fine that they will remain in suspension and be in- visible, by ordinary means. They are then said to be in the colloidal state. If we divide 'them further until they become molecules of gold, a molecular solution will be obtained. In the colloidal state, there are *Molecular solutions include those of nonelectrolytes, such as sugar, and electrolytes, such as inorganic salts. 50 COLLOIDS 51 two distinct phases in the solution, one solid and the other liquid, and between the two, because of the great subdivision of the original par- ticle, is an enormous surface of contact. The solution is heterogeneous, and at the interface between the two ' ' phases ' ' the physical forces which depend on surface — e. g., surface tension (see page 64) — are enormously developed, and are responsible for the peculiar properties of colloidal solutions as compared with those of molecular solutions, which may, therefore, be styled homogeneous. The solutions of crystalline substances which we have hitherto been concerned with, are homogeneous. Between these two groups of solutions is an intermediate one — namely, suspensions (as suspensions of quartz or carbon, or oil emulsions). Be- sides being turbid in transmitted light, the solutions may be seen by means of the ultramicroscope to contain particles. These can be sepa- rated by filtration from the fluid they are suspended in, except in the case of many emulsions in wrhich the particles can squeeze their way through the filter pores by changing their shape. On standing or being centrifuged suspensions may also separate into their constituents, al- though this can be greatly hindered by the addition of a suspending substance such as gelatin or certain bodies having a so-called protec- tive action (as peptone, proteose, etc.). True Colloidal Solutions 1. The Solution Is More or Less Turbid. — Frequently this can be recog- nized by holding the solution in a thin-walled glass vessel against a dark background, but the turbidity may be so slight that it requires for its detection the use of the Tyndall phenomenon. This is familiar to all in the effect of a beam of sunlight let in through a small aperture into an otherwise darkened room. In the course of the beam suspended dust particles, which are invisible in an equally illuminated room, be- come visible, and thus render very distinct the pathway of the beam. If a colloidal solution contained in a glass vessel, preferably with paral- lel sides, is held in the course of such a beam, the Tyndall phenomenon will be seen in the liquid, which is not the case with molecular solutions. Focused artificial light may be employed for intensifying the effect. The light that is sent out at right angles to the beam is plane-polarized, which means that the particles reflecting the light must be smaller than the mean wave length of the light forming the beam. It should be under- stood that the individual particles themselves may not be rendered visible to the naked eye by the beam, although in such cases they can often be seen by using intense illumination and a dark-field (ultramicro- scope) combined with suitable magnification (Fig. 12). 2. Colloids Do Not Readily Diffuse. — To demonstrate this, test tubes 52 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES are half filled with a 5 per cent solution of pure gelatin or a 1 per cent solution of pure agar, and, after the jelly is set, the solution under examination is poured on the surface; or, when it is of high spe- cific gravity, the tube of gelatin, etc., is placed mouth downwards in the solution. In the case of colloidal solutions very little if any diffu- sion into the gelatin or agar will occur, even after several days; whereas true molecular solutions will diffuse for. a considerable distance. When colored solutions are used, the diffusion can readily be recognized by visual inspection (see Fig. 13), but when they are colorless, the presence or absence of diffusion must be determined by removing the column of gelatin or agar and dividing it into slices of equal size, which are then examined chemically for the substance in question. A further test is afforded by the failure of colloids to diffuse through membranes (dialysis). This was the method originally used by Thomas Graham to distinguish between molecular and colloidal solutions. The solution under examination is placed in a dialyzer, which is then im- mersed in a wide vessel containing the pure solvent. The older forms Fig. 12. — Ultramicroscope (slit type) for the examination of colloidal solutions. The arrange- ment of diaphragms, etc., in this form removes the absorptive effects of the surfaces of the glass vessel or slide used to contain the colloidal solutions. of dialyzer consisted in general of a bell-shaped glass vessel closed be- low with parchment paper, but more recently so-called diffusion sacs have been adopted. These consist of pig or fish bladders or of col- lodion sacs. The latter are made by placing some collodion dissolved in ether in a test tube, which is then tilted so that the collodion runs out except for a thin layer which remains adherent to the walls. When the collodion has set, the sac can be removed after loosening it by allow- ing a little water to flow between the sac and the walls of the test tube. The sac containing the colloidal solution is then suspended in water or some of the solvent used in preparing the colloidal solution, care being taken that the menisci of the fluids inside and outside of the sac stand at the same level. Sometimes, especially when collodion sacs are used, some colloid may at first diffuse through, but if the outer fluid (the dialysate) is renewed and the dialysis allowed to proceed, this ceases. COLLOIDS 53 When a fluid solution exhibits both of the above properties (i. e., the Tyndall phenomenon and indiffusibility) , there can be no doubt as to its being in a true colloidal state, but there are substances, such as congo red or protein solutions of certain strengths, which may exhibit a very slight diffusibility in a dialyzer but not show the Tyndall phenomenon. Substances of this group constitute transitional types between molecular and colloidal solutions, and to determine their true nature it is neces- Fig. 13. — To show diffusion into gelatin of a crystalloid stain in b and the nondiffusion of a colloid stain in a. (From W. Ostwald.) sary to employ refined methods such as those of ultramicroscopy, ultra- filtration, etc., which can not be described here. 3. The Size of Colloidal Particles. — It will be apparent that the essential property upon which the above-mentioned phenomena depend is the size of the particle. Particles which can still be seen under the microscope are called microns. They have been computed to have a dimension of 0.1 /u, (0.001 mm.) or more, and they form suspensions. Particles which are invisible microscopically under the ordinary conditions of illumina- 54 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES tion, but are still visible when the ultramicroscopic illumination is used, are called submicrons. They have a dimension between 0.1 p. and 1 fjLfjt. (0.000,001 mm.),* and they constitute the colloids. Particles smaller than 1 fjifj, are called amicrons, this term being used to include the mol- ecules and ions present in molecular solutions. (The amicron of hydro- gen is, for example, computed to be 0.067 to 0.159 p,^, and that of water vapor, 0.113 /x/u,.) This classification of dissolved substances according to the size of the particles and molecules shows the relationship of one Fig. 14. — Diagram from W. Ostwald showing the relative size of various particles and colloidal dispersoids compared with a red blood corpuscle and an anthrax bacillus. class of substances to others. An idea of the relative sizes of colloidal particles and molecules in comparison with. such familiar objects as a blood corpuscle and an anthrax bacillus is given in Fig. 14. The fluid in which the "particle" is suspended is called the dispersion medium, or external phase, and the particle itself the dispersoid, or internal phase. It is the enormous development of surface which determines the dif- *H — 0.001 mm., and up, = 0.000,001 mm. COLLOIDS 55 ference in the properties of a colloidal solution from those of a suspen- sion of the same substance. Thus, the difference between a colloidal solution of platinum (prepared by allowing an electric arc to form be- tween platinum electrodes in water) and pieces of platinum in water depends on the fact that the surface of the platinum in the former case has been increased many million times. When the subdivision becomes still greater and the particles gain the size of molecules, the phenomena due to surface development become suppressed and those due to con- centration in unit volume become accentuated. The properties depend- ent on osmotic pressure, diffusibility, etc., are exhibited by all dispersoids, whether ions, molecules or particles, but some of these properties are much more pronounced when the dispersoids are of large dimensions, and others when they are small. In other words, the phenomena due to surface, such as those of surface tension (see page 64), become apparent only when the dispersoids have the properties of matter in mass; when the dispersoids become molecular in size, they manifest the properties characteristic of true solutions. 4. Electric Properties of Colloids. — Most colloids carry a charge, which may be either positive or negative toward the dispersion medium. Both crystalloids and colloids therefore .carry electric charges; in the former case, however, the charge does not reveal itself until the molecules in solution have become dissociated, when each ion carries a charge of opposite sign (see page 16), whereas1 in the case of colloids, each col- loid particle usually carries a charge which is always of one sign, either positive or negative. Colloids may therefore be grouped into positive and negative, according to the charges which they carry, and there is a third group in which the charge may be either positive or negative ac- cording to the nature of the dispersion medium. A colloid not carrying a charge to begin with can be caused to assume one by the action of electrolytes, for the electrical properties of colloids, as well as those of inert powders suspended in water, are readily in- fluenced by the charges present in the ions of the dispersion medium. The II • and OH' ions are especially liable to exert this influence. The particles of inert powders in suspensions (kaolin, sulphur, etc.) carry a positive charge when the water- in which they are suspended is acidi- fied, and a negative charge \vhen it is made alkaline. In general, it may be said that suspensions of most powders and of insoluble organic acids in water (e. g., charcoal, cellulose, kaolin, caseinogen, mastic, free acid of congo red, etc.) are electro-negative. Of true colloids ferric hydrox- ide (ferrum dialysatum) and serum globulin are positive in acid solu- tions; arsenious sulphide and serum globulin are negative in alkaline solution, and serum globulin in neutral solutions has no charge. 56 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES To ascertain the nature of the charge various methods may be em- ployed, of which the following are important: 1. The method of electrophoresis. The colloid solution is placed in a U-tube, each side of which carries a platinum electrode dipping into the solution. After a strong continuous electric current has been allowed to pass for some time through the solution, it will be found that the colloid collects at the anode (where the current enters) when it is a negative colloid (since unlike electric charges attract each other), and at the cathode when it is positive. In the case of colored solutions, the migration can be readily seen, but otherwise it may be necessary to ana- lyze the solution at the two poles. Fig. IS. — Capillary analysis of colloids. Strips of filter paper, after being suspended with the lower ends dipping into colloidal solutions. Those on the right hand were positive colloids, Which did not rise in the strips, but formed a sharp line of demarcation at • the lower end on account of precipitation. Those on the left hand were negative • colloids. (From W. Ostwald.) 2. The method of capillary analysis. For this purpose a long strip of filter paper is arranged vertically over the solution, with its lower end dipping into it. In the case of negative colloids the colloid, as well as the dispersion medium, rises uniformly on the strip of paper (it may be to a height of 20 cm.) ; whereas with positive colloids the dispersion medium alone rises, the colloid itself doing so only to a very, slight ex- tent, but becoming so highly concentrated at the interface between the solution and the paper that it coagulates on the end of the strip of paper, where it forms a sharp line of demarcation (Fig. 15). 3. The method of mutual precipitation of colloids. When a positive COLLOIDS 57 and a negative colloid are mixed in such proportions that the electric charges are neutralized, precipitation usually occurs. When it does so, we can tell the nature of the electric charge of an unknown colloid by its behavior when a colloid of known electric sign is added, to it. For example, if ferric hydroxide (positive) causes a precipitate to form when it is added to an unknown colloidal solution, the electric charge of the latter must be negative; if it does not precipitate with ferric hydroxide, but does so with arsenious sulphide (negative), it must be positive. 5. Brownian Movement. — Like the particles in fine mechanical suspen- sions, those of colloidal solutions, especially when examined ultra- microscopically, exhibit the so-called Brownian movements, which have been described as "dancing, hopping and skipping." These movements occur in straight lines, which are suddenly changed in direction and are quite independent of external sources of energy, such as change in temperature (although they become quicker as the temperature of the solution is raised), earth vibrations, chemical changes, or the electric charge of the colloid. The movements become more rapid the smaller the particles, and they become sluggish as the viscosity of the solution in- creases. Addition of electrolytes decreases the movement by causing the particles to clump together. The density and viscosity of the disper- sion medium, the electric charge of the dispersoid and the presence of Brownian movements, are the forces which operate together to prevent sedimentation of the particles in a colloidal solution. 6. Osmotic Pressure. — As one of the distinguishing properties of col- loids we have seen that their diffusibility, as into gelatin or agar jel- lies, is extremely slow when compared with that of a molecular solution. This does not mean, however, that colloids are possessed of no power of diffusibility if left long enough. Indeed the existence of the Brownian movement indicates that such diffusion must occur, and therefore it should be possible, by the application of the same principles as those which govern molecular solutions (e. g., by using a semipermeable mem- brane), to measure the osmotic pressure. Many studies of the osmotic properties of colloidal solutions have been undertaken, especially by those who are interested in the possibility that the colloids of blood serum (serum albumin and globulin) may cre- ate an osmotic pressure. If this should prove to be the case, it would be necessary for the osmotic pressure to be overcome by mechanical pressure such as that supplied by the heart (i. e., the blood pressure) in the various physiologic processes of filtration and diffusion taking place through cell membranes (as in the formation of urine in the kidney). For measuring the osmotic pressure of colloids, osmometers similar 58 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES to those already described (page 4) can be employed. Most of the recent work has been done either with collodion sacs, or with unglazed clay cups impregnated with some gel, such as silica cr gelatin. When such an osmometer, filled with some colloidal solution (like a solution of pure albumin) and provided with a vertical glass tube, is placed in an outer vessel containing water, the fluid will be seen to rise in the ver- tical tube, the height to which it rises being proportional to the osmotic pressure. But the observed pressure does not necessarily give us the osmotic pressure of the pure colloid, for to this, even when highly purified, there is almost certain to be attached a considerable amount of inorganic salt, which may be responsible for the osmosis. It has indeed been maintained by some observers that electrolytes form an integral part of certain colloids, being bound to them perhaps by adsorption (see page 65), and that they are essential to the maintenance of the colloidal state. In any case, since electrolytes are always present, the osmotic pressure of the pure colloid can be measured only when means are taken to discount their influence. Several devices have been used, of which the following may be mentioned: 1. Addition to the fluid outside the osmometer of a percentage of salt equal to that found by chemical analysis to be present in the col- loid. (This method is untrustworthy.) 2. The use of a limited quantity of fluid on the outside of the osmom- eter so that equality of saline content soon becomes established, by diffusion, in the fluids on the two sides of the membrane. 3. The use of a membrane which is permeable to electrolytes but not to colloids. Even when the greatest care is taken in its measurement, the osmotic pressure of a given colloid has been found to vary considerably not only according to the method used in its preparation, but also accord- ing to the amount of mechanical agitation (shaking, stirring, etc.) to which the colloid solution has been subjected. Regarding the influ- ence of the method of preparation, it was found in one series of experi- ments that albumin that had been repeatedly washed (but still con- tained considerable ash) gave no osmotic pressure, whereas another preparation that had been purified by crystallization tAvice (and con- tained much less ash) had a pressure of 3.38 mm. Hg. According to these results the ash content of the colloid is not fundamentally re- sponsible for its osmotic pressure. As to the influence of mechanical agitation, the osmotic pressure of a gelatin solution is increased by shaking, while that of a solution of egg albumin is. decreased. The property upon which the osmotic pressure depends is undoubtedly COLLOIDS 59 the state of dispersion of the colloid particles, and until we know all of the factors which may influence this, measurements of osmotic pressures of colloids can scarcely be of very much value. Nevertheless, that this property has some physiologic bearing is clear from the effect which col- loids have in restoring the blood pressure after hemorrhage (page 141). Further evidence that the osmotic pressure of colloids has not the significance that it has in the case of molecular solutions is furnished by the fact that the osmotic pressure is only approximately proportional to the concentration of the solution; it may either increase or decrease relatively to 'the strength of the solution. Temperature also has quite a different influence on the osmotic pressure of colloids from that which it has on the osmotic pressure of molecular solutions, and it frequently has an influence which persists after the solution is brought back to its original level. The influence of added substances on the osmotic pressure of colloidal solutions is of considerable interest to the biologist, for, whereas in the case of molecular solutions this is purely additive, in the case of col- loids the added substance may at one time cause the osmotic pressure to increase, at another, to decrease. It has been found that the osmotic pressure of gelatin solutions at first decreases, then rapidly increases as the H-ion concentration is raised. The addition of alkali increases the osmotic pressure until a maximum is reached, beyond which it begins to fall. Both acids and alkalies lessen the osmotic pressure of egg albu- min. Electrolytes always decrease the osmotic pressure of gelatin and albumin solutions, and the degree to which they exert this influence depends on the nature of the cation and anion composing the electrolyte. In the order of their depressing influence the cations arrange them- selves: Heavy metals > alkaline earths > alkalies; and the anions: S04 > Cl > N02 > Br > I > CNS. The influence of a given electrolyte varies extraordinarily with the reac- tion of the colloid, a fact which must be carefully regarded in all work in this field. ! CHAPTER VIII COLLOIDS (Cont'd) SUSPENSOIDS AND EMULSOIDS According to whether colloids form solutions that are more or less viscid than the suspension medium, they are divided into emulsoids and suspensoids. Examples of the former class are silicates and gelatin, and of the latter, dialyzed iron and arsenious sulphide. The following char- acteristics are used to distinguish between suspensoids and emulsoids: 1. Measuring the time it takes, at a standard temperature, for a given volume of the fluid to flow out of a standard pipette (10 c.c.) shows the viscosity to be, roughly, inversely proportional to the time of outflow. In the case of suspensoids the viscosity is no different from that of the dispersion medium alone, and does not vary much when the solution is cooled. The viscosity of emulsoids even in very dilute solutions is, on the other hand, considerably greater than that of the dispersion medium itself, and it becomes greatly increased by cooling. 2. Suspensoids are much more readily coagulated by the addition of electrolytes than emulsoids. This is particularly true when water is the dispersion medium (so-called hydrosols), and when electrolytes hav- ing a polyvalent ion (such as Al or Mg.) are employed. Thus, practically all suspensoids are coagulated in the presence of 1 per cent of alum, which has no influence on emulsoids. We shall return to this phase of our subject later on. The division of colloids into emulsoids and suspensoids is more or less arbitrary, since one class may be changed into the other, the determining factor being the water content of the dispersoid. The water content of suspensoids is low (lyophobe), while that of emulsoids is high. By changing the relative amounts of water and solid of which a colloidal solution is composed, the nature of the dispersoid may be changed. If the water is diminished, the dispersoid behaves as a suspensoid and be- comes readily precipitated. The practical importance of this fact is that it explains the salting out of proteins — a process extensively used in their separation. Ordinarily these behave as emulsoids, but the addi- tion of salt raises the osmotic pressure of the dispersion medium, and thus attracts water from the dispersoids, with the result that they come 60 COLLOIDS 61 to behave as suspensoids, and are accordingly precipitated by the elec- trolytes. Another property of emulsoids of biological' importance is the pro- tection which they can afford against the precipitating influence of electrolytes on suspensoids. If a colloidal solution of gold is mixed with a trace of gelatin, the subsequent addition of salts will be found to produce no precipitation. The explanation of this is that the emulsoid becomes distributed as a film on the suspensoid particles, thus practically converting them into emulsoids. Gelatinization One of the best known properties of emulsoids is that of gelatiniza- tion, which has an interesting bearing on many problems of biology. After the gel has set, an enormous pressure is required to squeeze out any water from it, indicating that the water no longer forms the con- tinuous phase but must be enclosed in vesicles formed of more solid material. Fig. 16. As a gelatin solution cools, the gel at first forms a polarized cone of light, but the very fine particles which are responsible for this effect soon increase in number and size so that they obstruct one another in their Brownian movements and adhere, giving an appearance of fine felt-like threads throughout the solution. A sort of impervious sponge work of the more solid phase is therefore formed, the more fluid phase being inclosed in the meshes. If, as in the accompanying diagram, the dispersion medium is repre- sented by white and the dispersoid in black, the relationship between the two in a suspensoid is as in A, and that in a gel as in B. To express any of the dispersion medium in B, it will require a pressure sufficient to 62 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES cause the more fluid phase to penetrate the more solid. If the gel is treated with reagents like formaldehyde, the liquid can be readily pressed out. This occurs during fixation for histological purposes. Imbibition Closely related to gel formation is the process of imbibition — the power of taking up large quantities of water without actually forming liquid solutions. Besides gelatin the dried tissues of plants and animals exhibit the phenomenon, and it is undoubtedly of importance in many physiologic processes such as growth and the passage of water into cells, etc. The materials present as vacuoles in plant cells attract water from the environment of the cell by imbibition, and thus exert on the cell wall a pressure which, acting along with the osmotic pressure, maintains the turgor of the cell. The initial growth of pollen is also dependent upon imbibition, and important observations on this process under varying conditions, are likely to furnish us with useful informa- tion concerning the significance of imbibition in connection with growth of cells in general. By measuring the rate of increase in length of long, narrow strips of gelatin placed in Petri dishes containing solutions of varying composi- tion, the factors that influence the imbibition process can be -quantita- tively investigated. Working in this way, F. H. Lloyd17 has found that for all acids there is a certain concentration (about N/320 H2S04) which Induces a maximum rate of swelling, and another, much weaker (N/2800 H2S04), in which the rate of swelling is even less than in pure \vater. In higher concentrations of acid than N/320, the gelatin at first swells very quickly, but the rate slows off so that it soon comes to be less than that with intermediate concentrations. Regarding alkalies, at high concentrations the effect is similar to that of acids. Salts alone seem to repress the swelling below that of water. It should be pointed out that the concentrations of acid and alkali in the above observations are much greater than those that could occur in the animal body. The experiments recall the attempts made some years ago by Martin Fischer to explain edema as due to excessive imbibition of water by the pro- teins of the tissues because of increased acidity of the blood and tis- sue fluids. That imbibition might possibly play some role in such processes is not denied, but Fischer disregards entirely the now well-estab- lished facts that hydrogen-ion concentration is one of the most constant properties of the blood, that very low concentrations of acid may dimin- ish rather than increase imbibition, and that it is manifested only in the absence of inorganic salts.* Moreover, the fluid in edema can often 'Determinations of the hydrogen-ion concentration of the blood recently published from Fischer's laboratory do not inspire confidence. COLLOIDS 63 be drained off by hollow needles, and it passes by gravity from one part of the blood to another, neither of which processes would be possible if imbibition were the essential factor concerned. If further evidence against this hypothesis should be demanded, it might be found in the utter failure of the therapeutic measures — alkali administration — that are recommended to combat the edema. Action of Electrolytes on Colloids (apart from their effect on osmotic pressure). — It has been stated above that the charge which a colloidal particle assumes may be neutralized by a charge of opposite sign car- ried by an ion present in the dispersion medium. The neutralization of the electric charge causes coagulation of the suspensoids but not of the emulsoids. Of the positive and negative ions into which the elec- trolytes dissociate, the one producing the coagulation is that which is opposite in sign to the electric charge of the colloidal particle. A quantity of electrolyte which is capable of producing complete pre- cipitation when added all at once to suspenroids will be ineffective when added in small quantities at a time. This phenomenon, which is also known to be exhibited when toxins and antitoxins are mixed together, is probably owing to the fact that precipitation depends on inequality and irregular distribution of electric charges, a condition which becomes established when the electrolyte is suddenly added, but not so when it is gradually added. The particles in the latter case become, as it Avere, acclimated to the electric charges introduced by the addition of the electrolyte. Proteins as Colloids. — The most prominent colloids in the field of bio- chemistry are the proteins. On account of complexity of structure, however, certain factors intervene which render the investigation of their behavior very difficult. As we shall see later, proteins are made up of combinations of amino acids, each of which contains basic (NH2) and acid groups (COOH). The various amino acids are linked together in protein by the COOH of one uniting with the NH2 of another, with the elimination of -water — thus, CO j OH + H ; HN — but some NH2 and COOH groups are left uncombined. According to the relative number of these uncombined radicles, the protein (or polypeptid, page 601) will exhibit faintly acid or basic or neutral properties. With acids, for example, a salt will be formed by union with the NH2 groups, Avhich will dissociate into the anion of the acid and a large organic cation; whereas with alkalies union will occur with the COOH group, and the salt on dissociating will form a small cation of the metal of the salt and a large complex anion. We may therefore obtain the protein with either a positive or a negative electric charge by altering the chemical nature of 64 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES the fluid in which it is dissolved, so that the reaction towards other colloids and towards electrolytes will vary. One feature of proteins of importance in this connection is that known as the isoelectric point, at which the protein exists with a maximum of electrically neutral molecules. This point is reached by adding acid to a protein solution. The acid represses the dissociation of the protein acting as an acid, and therefore diminishes the number of free hydrogen ions ; and at the same time it combines with the NH2 groups and neutral- izes the basic characteristics. The alteration in electric charge thus in- duced alters the water-absorbing powers of the protein and therefore all of the properties which we have seen to be associated therewith (page 63). SURFACE TENSION Before we consider a very important property of colloids known as adsorption, by means of which they are able to perform many reactions that do not conform with the laws of mass action, it will be well to A. Fig. 17. — Diagram to illustrate surface tension. The rings A and B inclose soap films in which a very fine loop of silk is suspended. In A it is loose but in B, where the film inclosed in the loop has been broken, it is drawn into a circle by the tension of the soap film. (From Bayliss.) say a few words concerning the physical phenomenon upon which this depends — namely, surface tension. The creation of this force is due to the fact that, whereas the molecules within a liquid are subjected to equal forces of attraction on all sides, at the surface these forces act on one side of the molecules only, and therefore tend to pull them inwards. This causes the surface to pull itself together so as to occupy the least possible area, and it is this force which constitutes surface tension. The surface behaves as if stretched. There are various simple experi- ments that reveal the presence of surface tension. If a film is made on a loop of wire by dipping it in soap solution, a fine silk thread can be floated in the film, so that it forms a loop that is quite loose. If the portion of the film inside the loop is destroyed by touching it with filter paper, the film will break in the loop, which will now be pulled into a circular shape by the tension of the film around it (Fig. 17). For the measurement of surface tension, various methods are used. COLLOIDS 65 The size of drops of liquid falling from an orifice is dependent on sur- face tension; the larger the drops, the greater the surface tension. If the number of drops obtained by allowing a liquid to drop from a stand- ard orifice in a given time is counted, we have a measure of the surface tension. Account must of course also be taken of the specific gravity of the liquid. The instrument used for this purpose is called a stalagmometer (Fig. 18). Another method depends on the fact that the height to which a fluid rises in a capillary tube is dependent on surface tension (and inversely on the diameter of the capillary). The difference in the heights to which two liquids rise in capillary tubes of known bore permits us to compare their surface tensions, and if this is known for one of the solutions, it can be determined for the other. Besides existing between liquid and air, surface tension also exists at the interface between two immiscible liquids, and at that between sus- Fib. 18.- — Traube's stalagmometer. The surface tension is proportional to the number of drops formed in a given time. The right-angled tubes are for thin liquids, and the straight one for blood and other viscous fluids. pended solid particles and liquid, as in colloidal solutions. Since, as we have seen, the surface area (interface) is enormously increased in these solutions, a very great surface energy is present, for this is equal to the surface tension multiplied by the surface area. ADSORPTION The surface tension between liquid and air is lowered when organic substances are dissolved in the liquid, but is slightly raised when inor- ganic salts are dissolved. The degree of lowering varies markedly ac- cording to the organic substance dissolved, being very pronounced with bile salts, upon which fact the well-known (Hay) test for the presence of bile in urine is based. Between liquid and liquid, as well as between 66 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES solid and liquid, the surface tension is always lowered ~by dissolving sub- stances in the liquid. Now, at the interfaces between solid particles and liquid there must be a local accumulation of free surface energy, which will be equal to the surface tension multiplied by the surface (inter- face) area. A constant tendency exists for such free energy to be de- creased and, since dissolved substances have this effect, they will become concentrated at the interface. This is known as the principle of Willard Gibbs, and it is of fundamental importance to the biochemist, because on it depends the phenomenon known as adsorption, which in the case of colloidal solutions may therefore be denned as the local concentra- tion or condensation of dissolved substances at the interface between the two phases. The amount of substance concentrated at the interface can be calculated by a formula which takes into account the concentra- tion of the dissolved substance, the temperature, and the surface tension at the interface (the Gibbs formula). After absorption has occurred, vari- ous reactions of a chemical, electrical or purely physical nature (e. g., dif- fusion) may follow at a rate which depends on the amount of the condensation. Every-day Reactions Which Depend on Adsorption 1. Decolorization of liquids by charcoal. That no chemical reaction oc- curs in such a case is readily shown by the ease with which the pigment can be extracted from the charcoal. 2. Adsorption of gases by such solids as charcoal and spongy platinum. In these cases there must be great condensation, even a liquefaction of the gas, during which heat must be evolved. By adsorbing oxygen and hydro- gen, spongy platinum causes these gases to combine and form water. The hemoglobin of blood may take up oxygen by a similar process. 3. Formation of solid surface films on solutions of protein, saponin, etc. The condensation may lead to coagulation, which explains why, if the froth produced by beating the white of an egg is allowed to stand, it can not be again beaten into a froth, the albumin having gone out of solution by surface coagulation. An interesting phenomenon depending on the surface tension occurs when the protoplasmic contents of a ciliated infusorian is pressed out in water. A new membrane forms on the protoplasm because of surface con- centration of all constituents which lower surface energy. By application of the principle of Willard Gibbs, A. B. Macallum18 concludes that not only adsorption, as exhibited in a colloidal solution, but also the local accumula- tions of material often seen in cells, are associated with changes in sur- face energy. His conclusions are based largely on microscopic studies of various forms of cell exhibiting different degrees and types of activity, COLLOIDS 67 and ingeniously stained for potassium by cobalt hexanitrite. By such a means the potassium stains intense black. In vegetable cells, local accumulations of potassium occur either near the interface between the clear and the chlorophyl-containing parts of the cell (spirogyra) or under a portion of the cell wall from which later a protrusion grows out to form the first stage in conjugation. The outgrowth from the cell, as well as the accumulation of potassium, may be the result of a low surface tension. In unicellular animal organisms, such as Vorticella, much less potassium is present, being confined to the base of the cilia, which Macallum believes indicates that the structures are produced as an outcome of low surface tension. In the cells of higher animals, deposits of potassium are also- localized ; in striated muscle, for example, they occur in a zone at each end of the doubly refractive band and immediately adjacent to the singly refrac- tive band. Changes in surface tension, associated with changes in the distribution of potassium, are believed by many to be responsible for muscular contraction. In nerves and nerve cells, potassium is concen- trated at the axon and at the surfaces of the cells. Interesting sugges- tions are offered to explain the relationship among changes in surface tension at the terminations of axons (synapses, terminations in gland and muscle cells) brought about by the nerve impulse acting as a change in electric potential. Surface condensation of potassium has also been observed at the lumen border of gland cells (pancreas), and on the lu- men surface of the cells of the renal tubules. Such observations indicate in what way surface tension may be called into play to control cellular activities. The field is new and almost unexplored, but there is already much to indicate that surface energy plays a most important role in the performance of many cellular activities. Conditions That Influence or Are Influenced by Adsorption Electric Changes.— Besides mere concentration, other forces come into play to assist or retard adsorption. One of the most important of these is electrical. Most solids when present as particles in a fluid carry a negative charge of electricity, some a positive one. In conformity with the AVillard Gibbs law, a constant tendency will exist for this free energy to be diminished by the neutralization of the electric charge. This can occur by deposition on the interface of other particles carrying an electric charge of opposite sign or by the action of that present on ions. Charcoal in suspension in water, for instance, has a negative charge. If colloidal iron, which has a positive charge, is added to the solution, it will become deposited on the charcoal, as will also the cations of an inorganic salt. On account of electric adsorption, dyestuffs and bile 68 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES salts are adsorbed much more freely than they would be if the process depended solely on surface condensation ; that is, if the Gibbs formula is used to calculate the adsorption, it will give values that are much below those actually found. If the dissolved substance and the particles both have the same electric sign, adsorption will not occur. Filter paper, for example, has a nega- tive charge and can not therefore adsorb a negative dye such as congo red (as shown by the depth to which it becomes stained) ; whereas it readily adsorbs night blue, which is positively charged. If the negative charge of the paper is lowered, it becomes capable of adsorbing some of the negative congo red. This can be effected either by placing the paper in alcohol or by adding inorganic salts (NaCl) to the water with which it is in contact. The positive-charged ions of Na, produced by dissocia- tion, neutralize some of the negative charge on the paper, and allow a certain amount of adsorption of the negative-charged congo red to oc- cur. As would be expected, acids and alkalies are capable of greatly altering the electric charges by the H and OH ions which they contribute. Chemical Forces. — If the nature of the phase at the surface of which adsorption occurs is such that it can enter into chemical combination with the substance adsorbed, reactions will occur that do not obey the laws of mass action. By adsorption, reactions of a certain type may be encouraged over other reactions, even although the necessary reacting substances may be present in the solution (specific adsorption). The adsorbing substance itself is not, however, usually susceptible of chem- ical change even when it exists as very minute particles, as in the case of colloidal solutions. Nevertheless, adsorption may accelerate chemical reactions by bringing together in concentrated form substances of high chemical reactivity. In such cases the adsorbing substance itself does not enter into the chemical reaction, and can be recovered at the end in an unchanged condition. It acts as a catalyst (page 72). As we shall see later, enzymes act in this way — i. e., their rate of reaction is controlled by adsorption.* The distinguishing feature of such adsorption phenomena is that a curve of the reaction (drawn by plotting amount of chemical change 'Another instance of the influence of surface energy on the course of chemical reactions is seen in the accelerative influence of charcoal on such reactions as the oxidation of formic acid, glycerol, etc. Surface tension may also cause retardation of chemical reactions, as is- seen in the turbidity (due to the separation of chloroform) which gradually develops when a —^ — Na2CO» solution is M mixed with a -^r— chloral hydrate solution. The surface remains clear, because surface energy has prevented the reaction. An important effect of surface tension on chemical reactions is also seen in the relationship between it and the absorption coefficient of gases (volume of gas dissolved by unit volume of liquid). The lower the surface tension, the greater the solubility of the gas. Oxygen and nitrogen are, for example, much more soluble in alcohol, hydrocarbons or oil than in water. This shows the futility of attempting to prevent the loss of gases from fluids such as blood by covering them with oils or hydrocarbons. COLLOIDS 69 against concentration of reacting substances) is a parabola, indicating that the laws of mass action (page 23) are no longer followed. In order to be able to determine whether some particular process — as, for example, a fermentation process, or the absorption of oxygen by blood — is caused by adsorption, we must compare its curves, constructed ac- cording to the same principles, with the typical adsorption curve. A formula may be used in constructing the curves. In arriving at this formula, two facts have to be remembered: (1) As adsorption proceeds and less and less of the free energy on the adsorbing surface remains to be neutralized, the reaction slows off, until equilibrium is reached. The more dilute the solution, the greater is the proportion of its con- tents to be adsorbed, which means that if a is the amount of substance adsorbed from a certain solution, then, from a solution of twice that strength, somewhat less than 2 a will be adsorbed — i. e., a multiplied by some root of 2. Although the formula is one belonging to the class known as parabolic, it must not be assumed that every reaction which happens to give such a parabolic curve (such as the combination of 0., with hemoglobin under certain conditions) (see page 383) must be one dependent on adsorption. It must be understood that although the substance that is removed from a solution by adsorption is no longer capable of contributing to the conductivity or the osmotic pressure of the solution, it is nevertheless not so firmly fixed that it can not be set free again by purely mechanical means, as by constant dilution of the fluid. If charcoal which has ad- sorbed sugar is placed in a dialyzer made of membrane the pores of which allow sugar but not charcoal to pass through, the sugar will gradually be removed if the dialyzer is immersed in running water. A certain equilibrium exists between the substance adsorbed and the same substance still remaining in solution. If the latter is constantly dimin- ishing by dialysis, the adsorption compound must break down to main- tain the equilibrium. It is clear, however, that the process of removal will be extremely slow. The ability of adsorbed substances to withstand removal by washing is taken advantage of by nature in holding back foodstuffs in the soil. Physiologic Processes Depending on Adsorption Instances in which adsorption undoubtedly plays a most important part in physiologic processes are as follows: 1. The action of enzymes (see page 71). 2. The combination of toxin with antitoxin occurs according to the laws of adsorption rather than those of mass action. In this case it is im- portant to note that when the toxin of diphtheria is added in small sue- 70 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES cessive quantities to diphtheria antitoxin, more toxin is neutralized than when the toxin is all added at once. A similar phenomenon can also be observed by adding filter paper to congo red, more of the pigment being adsorbed when the paper is added in small quantities than when added all at once. The explanation is that relatively more adsorption of a given substance occurs from a dilute than from a strong solution (cf. page 69). 3. The sensitizing of leucocytes by opsonins, as well as the subsequent ingestion of bacilli by the sensitized leucocytes, both of which follow the course of an adsorption reaction. 4. The formation of adsorption compounds, such as the inorganic salts and proteins and the complex lecithin compounds that can be extracted from egg yolk or brain tissue. In such compounds the laws of chemical proportion no longer hold, and properties may be exhibited that are quite different from those of either one of its components. When yolk of egg is extracted with ether, for example, a compound of lecithin with vitellin goes into solution, although vitellin itself is quite insoluble in ether.* There can be no doubt that adsorption compounds of this character are very abundant in living cells, and that they are constantly being formed and broken down. CHAPTER IX FERMENTS, OR ENZYMES One of the most striking developments of modern research in biochem- istry concerns the nature of enzyme action. So remarkable are many of the facts that have been brought to light that it can not fail to interest every one engaged in the study of life phenomena — whatever the nature of that study may be — -to know something of the main questions at present occupying the attention of investigators in this field. In this chapter a brief survey will be given of some of these questions; no at- tempt will be made at completeness, and only where necessary for the sake of example will reference be made to individual types of enzyme action. The discovery by Buchner that an enzyme can be expressed from yeast cells which is capable of instantly bringing about the alcoholic fermen- tation of dextrose solutions has been responsible for a great deal of the modern advance. Formerly, yeast cells were believed to bring about alcoholic fermentation as a result of their growth: it was believed to be a life phenomenon, or " vital process." Now we know that yeast cells produce an intracellular ferment or endo-enzyme* to which its sucroclastic properties are due and which can act apart from the cells that produce it. It is no great stretch of imagination to think of all chemical reactions mediated by cellular activity as due to a similar mechanism, and this thought has led to the hypothesis that all processes of intermediary metabolism in the animal and plant are caused by enzyme action. Before Buchner 's day we knew only of the extracellular enzymes (such, for example, as the digestive ferments), that is to say, of enzymes, produced indeed by cells, but secreted from them and acting outside their protoplasm ; now we must recognize intracellular enzymes acting where they are produced, in the protoplasm of the cell. But we must not permit this conception to carry us too far. Without further investigation we must not imagine that the riddle of life is thus solved. As an example of the role which extra- and intracellular enzymes are supposed to play in the animal economy may be cited the metabolism of protein. Proteolytic enzymes are very widely distributed in the active tissues of the animal and plant. By their agency in animal life, the com- *The terms "ferment" and "enzyme" are synonymous, but the latter is preferable as the noun, leaving the former to be used as the verb. 71 . 72 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES plex protein molecule is split up to render it absorbable from the intes- tine, and the tissues appropriate from the blood those of the degradation products that they require for the construction of protoplasm, which, later, they decompose so as to utilize the energy which the organism demands. All these processes are believed to be the work of enzymes. The Nature of Enzyme Action The changes brought about by enzymes can also be accomplished by ordinary chemical means, but these have often to be of a very energetic nature to accomplish what the enzyme can so quickly and quietly perform. It is the custom to regard "enzymes as catalysts. A catalyst is a sub- stance which accelerates (or retards) a chemical reaction which in its absence could proceed at a different (usually slower) pace. The action of catalysts has been aptly likened to that of a lubricant. A weight placed at the top of an inclined plane, so held that the weight only slowly slips down, has its velocity greatly increased if its under surface be oiled. The oil accelerates the action but does not affect the ultimate result. Catalysts do not combine with the final products of the reaction, these being, as a rule, the same as they would have been had no catalyst been added. Another characteristic is the tremendous amount of chem- ical change which even a trace of catalyst can induce. There are many examples of catalysts in the inorganic world, among which may be cited the action of spongy platinum on hydrogen peroxide. This substance normally tends to decompose into water and oxygen, but if a small amount of spongy platinum is added to it, the decomposition is greatly accelerated: H202 = H20 + 0. A very good example of the action of an inorganic catalyst is that of the hydrogen ion on cane sugar, or other disaccharides, in the presence of water. It accelerates the hydrolysis. Cane sugar solution at room temperature does not indeed, in sterile solution, undergo any appreciable hydrolysis, but at 100° C. it does, which leads us to believe that, though inappreciable, the action also occurs at room temperature. By adding a little hydrochloric acid, or other acid not having an oxidizing effect on sugar, we greatly accelerate the hydrolysis because of the hydrogen ions present in the acid solution. Within certain limits the rate of hy- drolysis is proportional to the amount of catalyst present. Enzymes, like other catalysts, produce their action when present in very small amounts (e. g., sucrase can hydrolyze 200,000 times its weight of cane sugar; diastase can convert starch to sugar in a dilution of 1-1,000,000) and there is a distinct relationship between the amount of enzyme present and the rate of the reaction. The final product of the FERMENTS, OR ENZYMES 73 reaction is, however, the same at whatever rate it proceeds, and the enzyme does not appear in the final products. Many enzymes such as diastase can be found unaltered in amount after they have completed their action. This is determined by adding a fresh supply of substrate (that is, of material to be acted on), when the enzymic action proceeds again in the usual way. The same is no doubt true for all enzymes, though as yet it can actually be proved for only a few of them. Enzymes, therefore, may be defined as catalysts produced by living organisms. The Properties of Enzymes Although enzymes are examples of catalysts, they exhibit many proper- ties that appear to differ from those of inorganic catalysts. It will, therefore, be advisable in considering each quality to compare it in catalysts and enzymes, for by this method a much clearer conception of the nature of enzyme action can be gained (Bayliss19). Those properties that are strictly peculiar to enzymes we shall consider later. 1. Most enzymes are remarkably specific in their action, whereas inor- ganic catalysts are very much less so. Thus, in the case of the enzymes which bring about inversion of disaccharides, this specificity is clearly shown. There is a special enzyme for each of the three disaccharides — maltose, lactose and cane sugar — :and one of these can not replace another. Still more strikingly is this specificity of enzyme action demonstrated in the fact that certain enzymes, such as zymase (expressed from yeast), will act only on bodies having a certain configuration, that is, having their side chains arranged in a certain way. Thus, there are two varie- ties of dextrose (a and /?), which differ from each other solely in the fact that the side chains are arranged in different positions with rela- tion to the central chain of carbon atoms. This form of isomerism is called stereoisomerism because the two bodies rotate the plane of polar- ized light to an equal degree in opposite directions. Zymase acts on one of these but not on the other, and there are innumerable examples of the same kind. Indeed, of all bodies that exist in two stereoisomers only one is found in living cells and it is on this variety alone that the enzymes in animals can act. A similar specificity exists between certain drugs and their pharmacologic action. Specificity of action is explained by supposing that a union occurs between the substrate and the enzyme, and for this union to take place the enzyme must possess a configuration which corresponds accu- rately with that of the substrate. The process has been compared to a lock and key; the key must be shaped to fit the lock, or it can not operate. The specificity does not, however, in itself disprove the close 74 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES relationship between enzymes and inorganic catalysts, for on the one hand there are several enzymes which do not exhibit this property, and on the other, there are inorganic catalysts which do. For example, lipase, the fat-splitting enzyme of pancreatic juice, decomposes not only fats but to a greater or less degree a number of bodies of the same gen- eral build (esters), and tyrosinase can decompose, not ty rosin alone, but all phenol compounds. Conversely, the hydrogen ion — to the pres- ence of which acids owe their catalytic powers — can decompose the ordi- nary esters (that is, of acids containing the carboxyl or COOH group) but it has no action on the sulphonic esters. However, enzymes are cer- tainly much more specific in their action than inorganic catalysts. 2. Temperature does not influence catalysis and enzyme action in the same way. As the temperature is raised in the case of inorganic catalysts, the reaction becomes about doubled in rapidity for each rise of 10° C., whereas in the case of enzymes it becomes increased up to a certain opti- mum temperature, beyond which, as the temperature rises, the reaction is first slowed and then disappears altogether. This peculiarity of enzymes as compared with inorganic catalysts need not in itself disprove the analogy between the two, because enzymes do not form true, but colloidal solutions. Colloidal solutions, as we have seen, are really fine suspensions of ultramicroscopic particles ; there is no splitting into ions of the dissolved substance, as is the case with true (molecular) solutions, but the colloid is suspended in the water or other solvent to form a heterogeneous system (page 51), on which account the surface area of the menstruum is enormously increased. Rise in temperature alters the extent of the surface area, and thereby intro- duces an influence which progressively opposes catalysis. Although inorganic catalysts in molecular solution show no optimum temperature but increase in activity in proportion as the temperature is raised, inorganic colloidal catalysts may show an optimum temperature. Thus, spongy platinum shows an optimum temperature in its action on a mixture of hydrogen and oxygen. It has therefore been suggested that it is because they are colloids that enzymes exhibit this property. 3. Inorganic catalysts frequently carry the reaction to a further stage than that attained by the action of enzymes. For example, acid breaks down the protein molecule much more completely than do the proteolytic enzymes. This difference is perhaps explained by the fact that enzymes are retarded in their activities when there comes to be a certain accumu- lation of the products of the reaction present. The final stages in the reaction may become so slow as to be almost inappreciable. This de- crease in activity is partly due to a union between the enzyme and the products of its activity. FERMENTS, OR ENZYMES 75 4. The velocity constant in the case of inorganic catalysts remains un- changed throughout the reaction, whereas in the case of enzymes it 'be- comes either less or greater as the process proceeds. When a substance is changed by catalytic action, it is, of course, constantly being diminished in concentration so that less and less of it remains to be acted on. This implies that there are fewer molecules present for the same amount of catalyst to act on and consequently that the amount changed in a unit of time is progressively less. At any moment, therefore, the rate of catalysis will be proportional to the amount of substance (substrate) left. To understand this we must refer back to what we have learned about mass action. If we suppose that two substances A and B react to form two other substances C and D, then, by the law of mass action, the reac- tion will not go on to completion but will stop when a certain equilibrium is reached. The reaction can be represented by the equation A + B +± C + D, which means that it proceeds at a rate proportional to the reacting molecules. In some cases this reaction goes on until either A or B has practically disappeared (that is, the equilibrium point is very near the right of the equation), as is the case in the inversion of cane sugar: C12 H22 On + H20 = C0 H12 06 + CG H12 06 Taking place as it does in an excess of water, and there being very little tendency for this reaction to go in the opposite direction (cf. re- versible action) (page 25), the only thing Avhich will influence its velocity is the concentration of cane sugar ; in other words, the velocity of the reaction at any moment will depend solely on the concentration, C, of the material still left undecomposed. This can be determined by means of an equation.* The value of such an equation is that it gives us a figure K, represent- ing the amount of inversion that would occur in each unit of time if the cane sugar were kept in constant concentration. When, for example, it is stated that K for a particular strength of acid acting on cane sugar solution is 0.002, this means that when volume, concentration of acid and *If x be the amount of sugar inverted in time *, and if we use a figure called a constant (K) to express the fundamental rate of the reaction (which will therefore be different for different reac- tions), then — — = KC. But C can not be the same at any two consecutive periods of time, because the reaction is going on continuously. This renders it necessary to use the notation of the differential calculus, and we have— — = KC. The sign 5 indicates that the reaction is a constantly changing ot one so that 5x and 5t represent such infinitely small amounts that they can not be measured. By methods of integration, however, it can be shown that the above equation may be written: K = 1 • log. nat. -^-, 1 2 1 *-2 thus permitting us to find the value of K (Ci C2 being the concentrations of cane sugar at the times TI T2). Any two determinations during; the course of the reaction can be used for calculating K. These equations apply only to cases in which but one substance is changing (monomolecular reaction). When two substances are involved, the equation is more complicated. 76 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES temperature are constant in a gram-molecular solution of sugar, 0.002 gram-molecule of sugar would be inverted the first minute and 0.002 gram each succeeding minute, provided we could keep the solution con- stantly a gram-molecular one, that is, provided we could add sugar just as quickly as it becomes inverted. At first sight it may appear of little practical importance to determine K. In our present discussion concerning the nature of enzyme action, it is however of great value for, whereas with inorganic catalysis K is really of constant value, with enzyme action it is not so. Thus, when cane sugar is inverted by sucrase — an enzyme present in the intestine and in yeast — the constant gradually rises; for most other unimolecular reactions mediated by enzymes it gradually falls ; for example, the action of trypsin on proteins. Where there is a great excess of substance to be acted on, in compari- son with the amount of enzyme present, it will be found that a more constant value than K is obtained when we compute the absolute amount of substance decomposed in a given time. In such a case, too, the amount of change in a given time will be proportional to the amount of enzyme present, indicating that some sort of combination between en- zyme and substrate must be the first step in the fermentative process. This fact has been noticed by us in connection with the hydrolysis of glyeogen in the liver. When there is an excess of glycogen present, the amounts which disappear in equal intervals of time after death are the same; when, on the contrary, there is not much glycogen, the amount which disappears gradually declines, but, if K be computed by the above equation, it is constant. To make these facts clear it may be well to pause for a moment to consider an illustration. The conditions obtaining when there is a large excess of substrate over enzyme may be compared to those governing the removal of a pile of bricks from one place to another by a number of men. The pile of bricks represents the substrate ; the men, the enzyme. If each man works up to his capacity, it is plain that the number of bricks transferred in a given time will not depend at all on the size of the pile to be transferred. When, however, the pile of bricks gets small, though the same number of men continue to work the number of bricks transferred in a given time falls off, because the men interfere Math one another's activities in securing their loads from the pile. When a similar stage is arrived at in enzyme processes, we have to use the velocity con- stant to show how much work could be done by the enzyme if the amount of substrate were maintained of constant amount. In the large volume of recent work which has been done with the object of discovering the cause of these variations in the velocity con- FERMENTS, OR ENZYMES 77 stant in the case of enzymes, four important conditions have been recog- nized: (1) reversibility; (2) gradual destruction of the enzyme; (3) com- bination of the enzyme with products of the reaction; (4) autocatalysis. Of these four influences the only one which could be held accountable for an increase in the activity of the enzyme is autocatalysis; in this process the enzyme by its action produces substances which intensify its own activity. In some cases at least — for example, the action of invertase on cane sugar — these are acid bodies, a moderate increase in acidity favoring the action of this enzyme. The other influences all tend to retard the reaction and progressively lower the value of K. Negative autocatalysis occurs when the enzyme produces products which interfere with its activity. Gradual destruc- tion of the enzyme and its union with the products of its activity will manifestly also decrease its power. There is plenty of evidence that both of these processes may occur. Reversibility of Enzyme Action But the most important of all the causes of retardation of enzyme activity is undoubtedly reversibility of action, which is an application of the law of mass action (page 25). If we take the saponification of an ester, the equation is: CH3CH2CH2COOC2HS + H2O ±5 CH3CH2CH2COOH + C,HBOH. (ethyl buty rate) (butyric acid) (ethyl alcohol) The equilibrium point is not so near the position of complete hydrol- ysis as in the case of the inversion of saccharose; in other words, the tendency for the bodies produced by the hydrolysis to reunite and form the original substances is quite marked, so that the reaction comes to an end before all the ethyl butyrate has been decomposed. For some time before the equilibrium point is reached, there will have existed a progres- sively increasing opposition to the breakdown of the ester, as a conse- quence of which, when enzymes are used to accelerate the reaction, the velocity constant, as determined by the above equation, will gradually fall as the reaction proceeds. Conversely, in a mixture of ethyl alcohol and butyric acid there is very slow synthesis to ethyl butyrate, and here again lipase accelerates the process; it induces a recognizable synthesis within a short time. Ethyl butyrate is usually employed for these ex- periments because, on account of its odor, the ester is readily recognized. Thus, if the alcohol and acid be mixed alone, no ester will be detectable, but if some lipase be added, it will soon become so. Similar synthetic action of lipase has also been demonstrated for mono- and tri-olein. 78 PHYSICOCHEMICAL BASIS OP PHYSIOLOGICAL PROCESSES It should be clearly understood that pure catalysts, such as the hydro- gen ion, in accelerating a reaction like the above, do so equally in both directions, so that the position of equilibrium remains unchanged. En- zymes may, however, cause this position to change because of their form- ing intermediate combinations. The reverse phase of certain reactions is probably the cause of at least some of the synthetic processes which occur in the animal body. A great difficulty in accepting such a view, however, is the fact that the equilib- rium point of all hydrolytic reactions, in the presence of an excess of water, is so near complete hydrolysis that very little synthesis can be possible. That is true so long as the substance synthesized is soluble, but if it is nearly insoluble in water, or if it is immediately removed from the site of the reaction by diffusion, or in any other way,, then it is obvious that it will go on being synthesized by the reaction. Thus, in the intestine neutral fat is hydrolyzed by pancreatic lipase into fatty acid and glycerin, which are absorbed into the epithelium, where they again come under the influence of intracellular lipase. This latter will tend to accelerate the synthesis of neutral fat from the fatty acid and glycerin until the equilibrium point of the system (fat acid + glycerin ^± neutral fat + H20) is again reached; but this point, although it is near the right hand of the equation, will really never be reached for the reason that the neutral fat, as quickly as it is formed, will become deposited in insoluble globules in the protoplasm and thus be removed from the equation. In support of this view it has been found that lipase is present in intestinal mucosa after all traces of adherent pancreatic juice have been washed away. By similar reactions the fat of the tissues becomes decomposed to fatty acid and glycerin and passes out of the blood when the concentra- tion of fat in this fluid falls below a certain level. Provided one of the substances synthesized is insoluble or can in some other way be removed from the reaction, it is plain that, even though the equilibrium point is very near to that of complete hydrolysis, yet the reversion will be suf- ficient to do all that is required of it. Results such as the above have prompted many to conclude that it is by such reversible action that all synthetic processes occur in the living organism. But the demonstrable synthesis of an ester must not be taken as evidence that all other syntheses are explainable on the same basis. For example, we have seen above that in the case of cane sugar the equi- librium point in the equation is so near that of complete hydrolysis, that no measurable amount of cane sugar is formed when dextrose and levulose are allowed to act on each other, and that cane sugar does not appear when sucrase is added to the mixture. If instead of sucrase we take another of the sugar enzymes — namely, maltase, which accelerates the FERMENTS, OR ENZYMES 79 decomposition of maltose into two molecules of glucose— there is, how- ever, evidence of synthesis as a result of the acceleration of a reversible reaction. To understand these results we must remember that ordinary dextrose is a mixture of two stereoisomers designated a and /?. When two molecules of a dextrose condense (that is, fuse togther with the loss of a molecule of water) maltose is formed, but when two molecules of /? dextrose condense isomaltose results. There is some controversy as to whether maltose is really responsible for the synthesis of a dextrose molecules to maltose, it being claimed by some that this is accomplished by another enzyme, emulsine. If this were true it would materially minimize the importance of reversible action as a factor in cellular syn- thesis. The latest evidence goes to show, however, that it is maltase and not emulsine that is responsible in the above case (cf. Bayliss). Evidence, both direct and indirect, is also steadily accumulating to show that enzymes may accelerate the synthesis of proteins. As pieces of direct evidence we have: (1) the retardation of the digestive action of trypsin, etc., which sets in after the process has gone on for a time, and (2) the recommencement of a digestive process apparently at an end, if the products of the digestion are removed by dialysis or other means. As direct evidence may be cited the formation of synthetic products when pepsin is added to concentrated solutions of peptone, and the diminution in the number of small molecules, as judged by meas- urements of electrical conductivity, when trypsin is added to the prod- ucts of tryptic digestion of caseinogen. Protamine — a simple form of pro- tein— has also been found to be produced when trypsin — obtained from a mollusc — was added to a tryptic digest of the same protamine. The significance of these facts in connection with the metabolism of the amino aids will be evident \vhen we come to study this subject (page 598).* Specificity of Enzyme Action Although in all of the above features of enzyme action there is nothing to contradict the vieAv that they are catalytic agents, there remains one peculiarity which at first sight seems uninterpretable on such a basis. This is with regard to their often remarkable specificity of action. Thus, as we have seen, maltase can hydrolyze maltose alone (which is com- posed of two a-dextrose molecules), but not isomaltose (composed of /^-dextrose). This means that mere difference in the configuration of molecules is sufficient to alter the influence of enzymes on them. Since such differences could not influence that of inorganic catalysts we must *We have been unable in this laboratory to demonstrate any synthesis of glycogen when gly- cogenase is added to a hydrolysis mixture of dextrine, maltose and glucose produced by the prolonged action of glycogenase on pure glycogen. 80 PHYSICOCHEMICAL BASIS OP PHYSIOLOGICAL PROCESSES explain the cause of the difference. This has been done on the basis either that enzymes are colloids or that the active (catalytic) group of the enzyme is attached to a colloid molecule. Before a substance can be acted on, it must combine with the colloid, which it does by the proc- ess of adsorption (see page 65). Thisi can occur, however, only when there is a harmony between the adsorbing substance and the substance adsorbed. Instances of the specificity of adsorption have already been given. In support of this view it has been found that of the two proteases, a and ft, in the spleen, one is adsorbed but not the other when a solu- tion containing them is shaken with Kieselguhr. Furthermore, when solutions of invertase are shaken with certain inert powders, the in- vertase is adsorbed by some of them but not by others. In strong sup- port of the adsorption hypothesis is also the fact that the same mathe- matical laws as apply in the process of adsorption are obeyed in the ratio which exists between the activity of an enzyme and its concen- tration in the solution. To sum up, then, catalysis as exhibited by enzymes involves three processes: (1) contact between the enzyme and the substrate, which will be dependent on their rates of diffusion; (2) adsorption between them, which will depend on their configurations (cf. the lock and key simile) ; and (3) the chemical change which itself probably takes place in two stages. In connection with the third process, it is probable that an initial com- pound of a definite chemical nature is first formed, followed by the hydrolytic or other chemical change, after which the enzyme group becomes free. It is very significant in this connection to. note that in their solubil- ities there exists a distinct relationship between the ferments and the substrates on which they react. Thus, trypsin is very soluble in water and acts on water-soluble proteins; lipase is soluble in fat solvents. Certain Peculiarities of Enzymes Notwithstanding the very strong case that is made out for the cata- lytic hypothesis, there are certain facts which many find it difficult to make conform with such a view. One of these is that dextrose can undergo three distinct and separate types of decomposition according to the enzyme allowed to act on it. These are alcoholic fermentation, butyric acid fermentation and lactic acid fermentation. It is difficult to see how simple catalytic action can be responsible for all three results. The enzyme must not only initiate the changes but also direct their course. Another peculiarity is that when certain enzymes — e. g., rennin, pep- FERMENTS, OR ENZYMES 81 sin, etc. — are inoculated in animals, they cause specific antienzymes to appear in the blood of the inoculated animal. Thus, when antirennin serum is added to milk it greatly hinders clotting on the subsequent addition of rennin. It is probable that powerful antienzymes are pro- duced in the animal body for the purpose of protecting the tissues from attack by enzymes. It is on account of the presence of antienzymes that intestinal parasites can exist in the intestine, and the immunity from digestion which the mucosa of the gastrointestinal tract enjoys, is believed to be due to the same cause. But there is considerable doubt regarding this claim. Fresh pancreatic juice when injected into the empty intestine digests its walls. When food is present in the intes- tine it evidently prevents digestion of the walls by diverting the enzyme to itself. Types of Enzyme Having learned something about the general nature of enzyme action, we may now turn our attention to certain details that have a practical importance. In the first place, with regard to nomenclature, in the earlier work each newly discovered enzyme received a name which was often quite inappropriate. Many of these names are retained, such as pepsin, trypsin, ptyalin, etc., but it is now customary to name the enzyme according to the substance on which it acts. This is done either by replacing the last part of the name of the substance acted on by the termination -ase (for example, the enzyme which inverts maltose is called maltase), or by merely adding -ase to the name of the substance acted upon (thus, the enzyme which hydrolyzes glycogen is called glycogenase). Most of the enzymes in the animal body accelerate hydrolytic proc- esses and are classified according to the chemical nature of the sub- strate on which they work. Thus, we have: 1. The amylases — accelerating the hydrolysis of polysaccharides, e. g., ptyalin (in saliva), amylopsin (in pancreatic juice), glycogenase (in liver), diastase (in malt). 2. The invertases — accelerating hydrolysis of disaccharides, e. g., malt- ase, lactase and sucrase (in succus entericus). 3. The proteinases — accelerating hydrolysis of proteins, e. g., pepsin (in gastric juice), trypsin (in pancreatic juice), erepsin, intracellular proteinases. 4. The Upases— accelerating disruption of neutral fats, e. g., steapsin (in pancreatic juice), intracellular lipases. 5. Arginase — accelerating hydrolysis of arginin into urea and or- nithin, (intracellular). 82 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES 6. Urease — accelerating hydrolysis of urea to ammonium carbonate (in many microorganisms and in the soy bean). 7. Glyoxylase — converting glyoxals into lactic acid (page 666). Other enzymes accelerate oxidative processes and are called oxidases and peroxidases. Others bring about the displacement of an ammo group by hydroxyl (desamidases) . Others cause coagulation (coagula- tive ferments], e.g., thrombin, rennin. One of the enzymes present in succus entericus acts by converting the zymogen (trypsinogen) into the enzyme (trypsin). Enzyme Preparations So far it has been impossible to prepare enzymes in a pure state al- though, being colloidal in nature, they are readily precipitated, or ad- sorbed along with other colloids. Since most enzymes exist in cells, it is necessary to break up the cells in order to isolate the enzyme. This is done in various ways. By one method the cells are ground in a mortar with fine sand, then made into a paste with infusorial earth (Kieselguhr), the paste enclosed in stout canvas and placed under an hydraulic press at about 300 atmospheres pressure; a clear fluid separates and this contains the enzymes. An- other way is to freeze the tissue with liquid air and grind it in a steel mortar by means of a machine. Still another and less expensive method, and one which we have found most useful for organs and tissues, con- sists in reducing the tissue to a pulp and, after sieving it to get rid of connective tissue, etc., spreading the pulp on glass plates and drying in a slightly warmed, dry air current. The scales of dried material are then ground in a paint mill with toluene, and the resulting suspension filtered ; the powder which remains on the filter, after thorough washing with toluene, is dried and kept for future use. The toluene removes all the fatty substances, so that when shaken with water, etc., the enzymes dissolve. Conditions for Enzymic Activity Reactions brought about by intracellular enzymes are very readily inhibited when there comes to be a certain accumulation of their prod- ucts of action. Thus, yeast ceases to ferment sugar when the alcohol has accumulated to a certain percentage. This action is partially due to a toxic action of the alcohol on the cell, which paralyzes its power of absorbing the substance to be acted on by the intracellular enzyme. If these products be not in some way removed, they will ultimately kill the cell and stop the fermentation. We have seen above how the ac- cumulation of products may interfere with the activities of enzymes in FERMENTS, OR ENZYMES 83 other ways in which the enzyme does not suffer destruction, as is shown by the fact that it resumes its original activities on removal of the products. Enzymes, both intracellular and extracellular, are very sensitive to- wards the inorganic composition of the medium in which they are act- ing. For the intracellular enzymes this is what we should expect when we bear in mind the profound influence of inorganic salts on the heart beat and on cell growth and division. This influence of salts and of reaction (acidity, etc.) on the life of the cell is so pronounced as to lead some observers to believe that abnormal cell multiplication in the body, as in the case of tumor formation, is due to changes in the inorganic composition of the tissue fluids. Extracellular enzymes are also very susceptible to the influence of inorganic salts but more especially so towards the reaction of the solution. In terms of modern chemistry we may say that the concentration of H- and OH' ions has a profound influence on the activities of enzymes. Most of the enzymes of the an- imal body perform their action normally in the presence of a slight ex- cess of OH' ions, that is, in faintly alkaline reaction. Indeed the only exception of importance to this is the pepsin of gastric juice, which nor- mally acts in an acid medium. An excess of either OH' or H- ions inhibits the activity of the enzyme and usually destroys it permanently. The activities of enzymes are also influenced by light, many of them being destroyed by sunlight; cells such as microorganisms are similarly affected. Before being secreted the digestive enzymes exist in the cells which produce them as inactive precursors called zymogens. The granules seen in resting gland cells are of this nature. The activation of the zymogen, or its conversion into the enzyme, occurs after it has left the cell, and this has been considered as another safeguard to digestion of the cell. Sometimes the activation does not occur until the zymogen has travelled some distance along the gland duct, as in the case of the proteolytic enzyme of pancreatic juice. Till it reaches the intestine, this exists as trypsinogen (the zymogen), but it is here acted on by another enzyme- like body produced by the intestinal epithelium and called enterokinase. PHYSICOCHEMICAL REFERENCES (Monographs and Original Papers) ^Bayliss, W. M.: Principles of General Physiology, Longmans, Green & Co., 1915. 2Philip, J. C.: Physical Chemistry, Its Bearing on Biology and Medicine, Arnold, ed. 2, 1914. 3McClendon, J. S.: Physical Chemistry of Vital Phenomena, Princeton University Press, 1917. ••Starling, E. H.: Principles of Human Physiology, ed. 2, 1915, Lea and Febiger. 84 PHYSICOCHEMICAL BASIS OF PHYSIOLOGICAL PROCESSES sKahlenberg, L.: Jour. Physical Chem., .1906, x, 141. , eReid, E., Weymouth: Jour. Physiol., 1898, xxii, Ivi. 7 Wilson, T. M.: Am. Jour. Physiol, 1905, xiii, 150. siialdane, J. S., and Priestley, J. G.: Jour. Physiol., 1916, 1, 296; Priestley, J. G : Ibid., p. 304. 9Clark, W. M., and Lubs, H. A.: Jour. Bacteriology, 1917, ii, 1 and 109. loHenderson, L. J.: The Excretion of Acid in Health and Disease, Harvey Lectures, J. B. Lippincott Co., 1915, x, 132. "Henderson, L. J.: The Fitness of the Environment, Macmillan, N. Y., 1913. i2Van Slyke, D. D.: Jour. Biol. Chem., 1917, xxx, 289, 347. isLevy, K. L., and Eowntree, L. G.: Arch. Int. Med., 1916, xvii, 525. i4Cull'en, G. E.: Jour. Biol. Chem., 1917, xxx, 369. ispalmer, W. W., and Henderson, L. J.: Arch, Int. Med., 1913, xii, 153. 16Sellards, A. W.: The Principles of Acidosis and Clinical Methods for Its Study, Harvard University Press, Cambridge, 1917. i^Lloyd, F. H.: Private communication. isMacallum, A. B, : Surface Tension and Vital Phenomena. University of Toronto Studies, No. 8, 1912; also Ergebnisse der Physiologic, 1911, ii, 598. isBayliss, W. M.: Enzymic Action, ed. 2. Monographs in Biochemistry, Longmans, Green & Co. PART II THE BLOOD AND THE LYMPH CHAPTER X BLOOD: ITS GENERAL PROPERTIES BY R. G. PEARCE, B.A., M.D. The blood, being the carrier of the nutritive and waste substances of the body's metabolism, must at one time or another contain all the ma- terials which compose the tissues in addition to those which are peculiar to the blood itself. It is a very complex fluid, and all of its constituents are not fully known. Structurally it is composed of water in which are dissolved various gases and organic and inorganic bodies, the corpuscles and platelets. The most accurate method of determining the volume of blood in the body is by bleeding and subsequently washing out the blood from the vessels and then estimating the amount of hemoglobin in the total fluid (Welcher's method). This method employed in the case of two criminals who had been decapitated gave the weight of the blood as 7.7 and 7.2 per cent of the body weight. Bloodless methods for deter- mining the total volume of blood are based upon the principle of add- ing a definite quantity of a knoAvn substance to the circulation and then estimating its concentration in a sample of blood withdrawn from the body shortly afterward. If the substance can not leave the blood vessels and does not cause fluid to be withdrawn from the tissues, the total quantity of blood in the body can be calculated from the concentration of the injected substance in the blood. The most accurate methods based on this principle are Haldane and Smith's, in which carbon monoxide gas is inhaled in a given amount and the carbon monoxide hemoglobin sub- sequently determined colorimetriealjy ; and Keith, Rowntree and Ger- aghty's, which employs vital red, a dye of low diffusibility. The dye remains long enough in the body to be thoroughly mixed with the blood, and its concentration in the plasma is determined colorimetrically 85 86 THE BLOOD AND THE LYMPH by comparing with a suitable standard mixture of dye and serum. These methods give the total amount of blood in the body as from 5 to 8.8 per cent of its weight. Meek has recently developed a method in which gum acacia is used. After mixing with the blood, the concentration of this substance is determined from the calcium content. Being colloid, none of the gum leaves the blood vessels. The newer methods have shown that the volume of the circulating fluid is maintained fairly constant in spite of influences tending to alter it. The body accomplishes this by drawing upon the reserve fluid in the tissues and by varying the rate of water excretion, particularly through the kidneys. Years ago the doctrine of an increased amount of blood in the body (plethora) gave rise to the therapeutic use of bleeding. Especially was this thought to be useful in conditions which we now recognize as chronic hypertension, and which show no increase in blood volume. Indeed variation in blood volume is not common, although plethora may occur in polycythemia, chlorosis, and anemias, and there may be a temporary reduction in the amount of blood in diseases in which there is a great depletion of water, as in Asiatic cholera, and fol- lowing very severe hemorrhage. While the total quantity of the blood in the body does not vary greatly, the concentration of its various constituents is subject to distinct change. The volume percentages of the corpuscles and the plasma can be approx- imately determined by allowing oxalated blood to sediment or by cen- trifuging in a graduated cylinder by the use of the hematocrit. Such methods are not very reliable, but may yield some important information. Normally 45 to 50 per cent of the volume of blood is composed of cor- puscles. It varies more or less directly with the number of red blood cells. THE WATER CONTENT OF THE BLOOD Since the blood plasma is essentially a watery solution, some idea of its water content can be obtained by a determination of the specific gravity. The most accurate method for accomplishing this is to deter- mine directly the weight of a given volume of blood and compare it with the weight of the same volume of water. Since this method re- quires a rather large amount of blood, indirect methods using smaller amounts have been devised. One of these (Hammerschlag's) uses a. solution of chloroform and benzol of a specific gravity of about 1.050, in which a drop of blood is suspended by delivering it cautiously from a pipette bent at right angles near its tip. If the drop sinks, chloroform is added; if it rises, benzol is added until the drop remains suspended. BLOOD: ITS GENERAL PROPERTIES 87 The specific gravity of the benzol-chloroform mixture is then determined, and this value is supposed to give the specific gravity of the blood. The specific gravity of the blood determined in this way varies be- tween 1.040 and 1.065. It is somewhat less after eating and increases after exercise; it is slightly lower during the day than at night, and the variation in .individuals. is considerable. The changes which occur in the specific gravity of the blood in disease are chiefly due to variation in the percentage of protein, since the salt content of the blood is rela- tively fixed. It is only when great changes occur in the concentration of the noncolloidal salts that they markedly affect the specific gravity. From 90 to 92 per cent of the plasma and from 59.2 to 68.7 per cent of the corpuscles consist of water. Of the whole blood, from 60 to 70 per cent by volume or about 55 per cent by weight consists of plasma; and from 40 to 30 per cent by volume or 45 per cent by weight consists of cor- puscles. THE PROTEINS OF THE BLOOD The plasma obtained by centrifuging the blood rendered noncoagula- ble by oxalates, hirudin or other means (see page 99), contains 5 to 8 per cent of coagulable proteins. These proteins are serum albumin, serum globulin, and fibrinogen. They can be separated from each other by the use of acids and neutral salts. Their proportion varies under dif- ferent conditions, but is approximately as follows: Fibrinogen 0.15-0.6% Serum globulin : 3.8% Serum albumin 2.5% The amount of fibrinogen is subject to the greatest variation (Mathews). Fibrinogen The least soluble of the blood proteins is fibrinogen. The plasma is almost freed of it by half-saturation with sodium chloride, or with a small amount of acetic acid. It is precipitated as fibrin in the process of blood coagulation (see page 99), and is estimated by weighing the amount of fibrin which it produces. Serum Globulin and Serum Albumin Globulins are ordinarily defined as being insoluble in distilled water, and albumins as being soluble. It is, however, impossible to separate serum globulin and albumin satisfactorily in this manner. The globu- lin obtained by dialysis can be returned to solution by the addition of 88 . THE BLOOD AND THE LYMPH a suitable amount of water, Avhich makes the salt adherent to the pre- cipitate a weak saline solution. In neutral or acid solutions it is coag- ulated by heat at about 75° C. But it does not act as an individual pro- tein, since a portion of it is precipitated by dialysis or by carbon diox- ide. Probably serum globulin really consists of two or more proteins. The serum albumin remaining in solution after saturation with am- monium sulphate likewise does not represent a chemical entity. It is possible by carefully heating the solution of serum albumin to distin- guish three separate coagulation temperatures. This fact has been in- terpreted as meaning that the serum albumin consists of at least three closely related proteins. Since the refractive index of the Mood depends primarily upon the amount of protein present, it has been taken as a means of determining variations in the concentration of the proteins. It has been found that the concentration of the blood proteins varies somewhat; during ex- ercise it is increased probably because of the taking up of water by the tissues, and during profuse bleeding it is diminished because large amounts of fluid are being added to the blood from the lymph, which is relatively poor in proteins. The ingestion of considerable amounts of salts has been found to reduce the concentration of the blood proteins for a short time. In pathologic conditions, as in diabetes, when rapid changes in the body weight due to alterations in the diet are oc- curring, changes in the fluid content of the blood are often observed. Likewise in edema caused by faulty renal function, there may be a re- tention of fluid in the blood before there is any indication of edema. The hydremic condition of the blood can therefore be considered as a useful diagnostic aid in determining the water metabolism. The relative concentration of the proteins of the blood is also of some interest, especially since in some diseases a considerable amount of blood protein is lost. By refractrometric methods it is possible to sep- arate the globulin and albumin fractions. Normally the total proteins range between 6.7 and 8.7 per cent, of which the albumins lie between 4.95 and 7.7 per cent, and the globulins between 1 and 2.54 per cent. In some diseases, as in chronic nephritis, pneumonia, and syphilis, the total proteins of the blood are decreased and the relative amount of serum globulin is increased On the other hand, in many mild infections and chronic septic conditions the globulin fraction may be increased with no change occurring in the total protein content.3 Our knowledge of the origin and the function of the Hood proteins is quite unsatisfactory. Previous to the discovery of amino acids, the building stones of the proteins, in the blood it was thought that the nitrogenous nutrients were converted somehow into blood proteins dur- BLOOD: ITS GENERAL PROPERTIES 89 ing or immediately following their absorption from the alimentary canal, and that the tissue cells were nourished from this common pro- tein. It is now known that the amino acids are not immediately syn- thetized into blood proteins after their absorption from the digestive system. The blood proteins are radically different from the tissue pro- teins. Substances which retard or accelerate nitrogen metabolism do not alter the relationship existing between the protein bodies of the blood. This fact indicates that the. serum proteins have a function quite independent of the nitrogenous metabolism of the body. They un- doubtedly maintain the viscosity of the blood and assist in preserving its neutrality. Attempts to localize the site of formation of the blood proteins have not been successful. There is some evidence that fibrin- ogen is formed for the most part in the tissues of the splanchnic area (liver). It is quite possible that the blood forms its own protejns, just as do other tissues, from the amino acids it contains. THE FERMENTS AND ANTIFERMENTS OF THE BLOOD The blood plasma contains many of the ferments present in the tissues. The nature of these ferments has been the subject of many investiga- tions in recent years, primarily because it has been found that they are intimately connected with the problems of immunity. Among the ferments the following have been demonstrated in the blood: Proteases are probably present normally in the human blood serum in small amounts, but they are found in large amounts in the white blood corpuscles. A protein foreign to the body if injected into the blood ordinarily produces no untoward symptoms, but a second injec- tion following the first by some days will produce symptoms of poison- ing known as anaphylaxis. This fact has led to the assumption that the injection of any foreign protein into the blood promptly leads to the appearance therein of specific proteolytic enzymes which will digest the strange protein into its derivatives, which are poisonous. This power of the body to produce specific proteases has been the subject of much research and debate, and Aberhalden proposed a test for preg- nancy, for cancer, and for other conditions in which he made use of this phenomenon. He believes the presence of placenta or tumor tissue to cause the presence of proteins that bring about the production of specific ferments whose duty it is to rid the system of these substances. Other investigators fail to find the specificity in proteolytic action claimed by Abderhalden, and believe that proteolytic ferments which are" capable of digesting foreign proteins are absorbed from the alimentary canal 9Q THE BLOOD AND THE LYMPH from the digestive juices (Boldyreff). Some investigators fail to confirm the claim that the proteolytic activity of the blood serum is increased under the above conditions. Blood contains an antiferment known as antitrypsin. This can be removed from the blood serum by several substances, among which are kaolin, colloidal iron and starch. Serum thus treated shows strong pro- teolytic activity and autodigestion will occur. In this case there can be no question of the specific origin of proteases. Abderhalden believes that the ferments of the blood of the pregnant woman are able to digest the placental tissue. Human placental tissue has the ability of absorb- ing antitrypsin and it is very questionable as to whether the test pro- posed by Abderhalden is due to the new formation of ferments or to the removal of the antitrypsin and the action of the protease normally present in the blood. Nuclein ferments are capable of decomposing nucleic acid and purins into the. simpler bodies. Lipases have been demonstrated in the blood. Amylase. — The presence of starch-splitting ferments in the blood was first shown by Magendie in 1841, and later Bernard showed that gly- cogen or starch injected into a vein produced glycosuria. Since then it has been proved conclusively that diastatic enzymes are normally present in the blood and lymph. The source of these enzymes has given rise to much speculation. Some observers believe that they are derived from the amylopsin of the pancreatic secretion, while others believe that they are manufactured by the liver. Ligature of the pancreatic ducts is said to inccease the amount of amylase, while removal of the pan- creas may (Carlson and Luckhart) or may not (Schlesinger) increase the amylase of the blood. In some forms of experimental diabetes the amylase of the blood has been found increased, and this is the case in human diabetes (Myers and Killian). If this is true, a cause for the inability of the diabetic to store up glycogen is found. In impairment of renal function, there is usually an increase in the blood amylase and a decrease in the urine amylase. This has been suggested as being of diagnostic value. The blood contains a feeble glycolytic enzyme capable of destroying glucose. It is claimed that this power is reduced in diabetics (Lepine). Catalase is found in the blood and tissues generally. It has the power of liberating oxygen from hydrogen peroxide without any accompany- ing oxidation process. Its physiologic significance is not known. It is said that the amount of catalase is increased during excitement and exercise,* and is decreased in conditions where the body's activity is lowered. Its determination is clinically unimportant at present. CHAPTER XI BLOOD: THE BLOOD CELL BY R. G. PEARCE, -B.A., M.D. THE RED BLOOD CORPUSCLES, OR ERYTHROCYTES The most prominent function of the blood is to carry oxygen to the tissues. It owes this property chiefly to the red blood cells which are present in large numbers (5,000,000 per c.mm. of blood). These cells are biconcave discs, having a diameter of about 7.7 /*. They are con- structed out of a framework composed largely of lipoidal material, in the meshes of which is deposited a substance called hemoglobin, to which the remarkable oxygen-carrying power of the blood is due. Nei- ther the manner by which the red cell carries its hemoglobin nor the intimate structure of the cell itself is accurately known. It is com- monly believed that the hemoglobin is held enmeshed in a framework or stroma, or encased in the cell membrane. One thing is certain, how- ever, that the union of hemoglobin with the stroma of the red cell is a fairly strong one, since mere fragmentation of the corpuscle fails to liberate the hemoglobin.. The fact that the framework contains a large amount of lipoidal substances enables the corpuscles to maintain their shape and is responsible for their characteristic permeability. Hemoglobin is a very complex substance belonging to the group of conjugated proteins. By chemical means it can be broken up into a simple globulin and a pigment hematin, containing iron. "When com- pletely saturated, oxygen is present in hemoglobin in the proportion of two atoms of oxygen to one atom of iron (Peters) ; or 401 c.c. of oxygen can be carried by hemoglobin containing one gram of iron, the molecular weight of the molecule being about 16.669, or some multiple thereof (Barcroft and Peters) (see also p. 397). At this figure the iron in the molecule would represent 0.34 per cent of the total weight of the molecule. The corpuscular surface area has been estimated to be 3200 square meters. There is therefore a very large surface avail- able for the absorption of oxygen from the alveolar air, as the blood corpuscles pass in single file through the capillaries of the lungs. Since the amount of oxygen which the blood can carry depends upon its hemoglobin content, it is of some importance clinically to have 91 92 THE BLOOD AND THE LYMPH' methods of determining the approximate amount present. The amount of hemoglobin present in a quantity of blood is usually determined colorimetrically by comparing the color of the blood with standard col- ors which correspond to known strengths of hemoglobin. In normal persons the amount of hemoglobin varies greatly at different ages, and in order to determine whether or not a given blood contains more or less hemoglobin than normal, it is imperative to consider the age. The greatest variations occur between birth and the sixteenth year. After the sixteenth year the blood in males usually contains a larger amount than that in females (Williamson4). Instruments used in determining the amount of hemoglobin should be standardized to give the value in grams hemoglobin per 100 c.c. of fluid. The amount of hemoglobin which is present in each corpuscle in terms of normal is therefore of some clinical interest. This relation of the number of red cells to the amount of hemoglobin is known as the color index and is computed as follows: The average red count in man is 5,000,000 to the c.mm., and the average minimal amount of hemo- globin is taken as 13.88 grams in 100 c.c. of blood (=80, Sahli; =90, Miescher; =86, Plesch; and 110, Tallquist methods). These relative values give a color index of one. The percentage of normal red cells divided by the percentage of normal hemoglobin present gives the color index. The Origin of the Red Blood Cells In fetal life the spleen and the liver are generally believed to be re- sponsible for the formation of the red blood cells. In extrauterine life this function is taken over by the -red bone marrow. In the primitive condition all red blood cells are supposed to be nucleated. In extra- uterine life the nuclei of the red cells are lost, and nonnucleated forms are alone present in the blood stream. In fetal life and in certain path- ologic conditions, the rate of blood formation is so rapid that some nucleated cells appear in the blood. The normal response of the body to a loss of red blood corpuscles consists in an increased activity of the blood-forming cells of the red bone marrow. It is not easy to follow the course of the regeneration of the red corpuscles or to discover the mechanism of their formation in the bone marrow, since this tissue pre- sents a mixture of cells which are precursors of the varied corpuscles found in the blood and the identity of which can not be determined. Recently new methods of staining blood for microscopic examina- tion have allowed more detailed study to be made on the site and method of blood cell formation. When fresh unfixed blood is treated with solutions of various dyes, such as brilliant cresyl blue, polychrome THE BLOOD CELL 93 methylene blue or neutral red, an otherwise invisible structure appears in some cells in the form of coarse granular particles or threads, which give a reticulated appearance to the corpuscles. These reticulated cells are more abundant in infants' blood and in patients suffering with se- vere anemia or hemolytic jaundice than in normal blood, and may be taken as evidence of the youth of the red cell and not as a degenera- tive process. Since the number of the reticulated cells that are present in the blood is more or less directly proportional to the hemopoietic activities of the bone marrow, enumeration of the reticulated cells is of clinical importance in anemias. In conditions in which animals have been made plethoric by the transfusion of blood,- it has been found that the number of reticulated cells is decreased; the bone marrow of these animals also shows a marked reduction in reticulated erythroblasts. The diminished rate of blood cell formation sometimes noted after blood transfusions may be explained by assuming that the stimulus which awakens the formation of red cells in the bone marrow is absent or made subnormal on the injection of red cells into the blood, and thus the formation of red cells is depressed. Small transfusions are there- fore preferable to large ones in cases in which the rate of blood forma- tion is greatly impaired. By means of living cultures of red bone mar- row the different stages of the development of the normoblasts into true red corpuscles may be studied (Tower and Herm5). Some evidence has been gathered from such studies which points to the conclusion that in place of the red cells being cells which have lost their nucleus, as is the current teaching, they are rather cells which develop as a nuclear bud and escape into the circulation as true red cells.. The nucleated red cell and the red nucleated corpuscle of the bird are the product of intranuclear activity and are morphologically identical. Rates of Regeneration of Erythrocytes Microscopic examination of the blood during rapid regeneration of red cells shows the presence of nucleated forms. Nucleated red cells in the blood have therefore been taken as an inevitable feature of rapid blood regeneration. The evidence upon which this belief depends, however, is hardly complete, since changes in the manner of red blood cell formation may be responsible for the nucleated forms. The red bone marrow is considered the seat of red cell formation, and it is true that an abnormal increase in the red bone marrow usually accompanies increased red cell formation. The nature of the stimulus which brings about the new. formation of red cells is not understood. Oxygen wrant may be an important factor, since we find the presence of an abnormally large number of red cells in conditions where there is a scarcity of 94 THE BLOOD AND THE LYMPH oxygen in the inspired air, as in life at high, altitudes, or a difficulty in its absorption through the lungs, as in congenital heart disease. The red cells produced following hemorrhage and in simple anemia contain less than the normal amount of hemoglobin, but their shape and size are approximately normal, and few nucleated cells are present. In the regeneration of red cells which is found in pernicious anemia, we find the cells containing an unusually large amount of hemoglobin. The red cells in this disease have abnormal forms, many being large, with or without a nucleus, and containing basic staining granules. This type of blood cell formation is due to degenerative changes. The Fate of the Erythrocytes The length of life of the red blood cell is unknown. Estimates based upon the daily excretion of bile pigments are not reliable, since Hooper and Whipple have shown that the pigments, in part at least, arise from pigments which the liver has made in excess of its needs for the manu- facture of hemoglobin, and which, not being needed, are excreted.5 There is no question however that every erythrocyte sooner or later undergoes disintegration, a process formerly thought to be ushered in by the ingestion of the red blood cell by a phagocyte in the spleen or in a hemolymph gland, the hemoglobin of the disintegrated cell being set free and carried to the liver, where it is broken up into hematin, which the body stores for future use, and into bile pigments, which are ex- creted. Rous and Robertson6 fail to find evidence that this process occurs in man to an extent sufficient to account for the normal destruc- tion of the blood cells. However they have recently found another and unsuspected method for blood destruction in all animals thus far studied — namely, the disintegration of the blood cells by fragmentation while they are circulating, without loss of their hemoglobin. These fragmented cells are found most frequently in the spleen. They believe that the small ill-formed cells, known as microcytes and poikilocytes, observed in severe experimental anemias, are due not to the fact that they are produced by the bone marrow, but rather to the fact that the marrow in its anemic condition is not able to produce a resistant ery- throcyte, and fragmentation therefore takes place too readily.- A sim- ilar condition may exist in the severe anemias of man and account for the general high resistance of the red cells found in the blood of these patients, inasmuch as the weak cells are generally fragmented very soon after they are formed. Long ago Ehrlich stated that the microcytes and poikilocytes of anemia are the result of fragmentation of the cells in the circulating blood, but he believed that this fragmentation was a THE BLOOD CELL 95 purposeful division in order to increase the total surface of the red cells. The ultimate fate of the red cell fragments is not known. It is reasonable to suppose that the fragmented bits containing hemoglobin are carried to the liver, where the hemoglobin is transformed into hematin and bile pigments. Hemolysis Another method of red blood cell destruction, which, however, does not take place normally, is by hemolysis. The nature of the combina- tion of the hemoglobin with the stroma of the red cell, as already re- marked, is not definitely known. That it is not merely contained in a sac is shown by the fact that the cell may be cut into bits without the hemoglobin being set free. In some manner the hemoglobin is chem- ically bound with the stroma of the red cell, from which it can be freed by a number of physicochemical and chemical agents. This proc- ess is known as hemolysis, and the substances which bring it about are known as hemolytic agents. The manner in which these agents effect the release of hemoglobin from the blood is quite varied. If the osmotic pressure of the plasma is lowered by dilution, the pres- sure within the corpuscle remains high, and water is absorbed by the cell. If this absorption is sufficient, the cell ruptures and the hemoglobin is discharged. For this reason it is necessary in diluting the blood to use solutions of salt having an osmotic pressure equal to that of the blood to protect the red cell from hemolysis. This is obtained by using a 0.9 per cent solution of sodium chloride. Better results are had, however, by using either Ringer's solution (0.9 per cent NaCl, 0.026 per cent CaCl2, and 0.03 per cent KC1) or Locke's solution (0.9 per cent NaCl, 0.024 per cent CaCl2, 0.042 per cent KC1, 0.01-0.03 per cent NaHC02 and 0.1 per cent glucose). In normal corpuscles hemolysis occurs to a small extent in solu- tions containing about 0.42 per cent of sodium chloride. In certain diseases the fragility of the corpuscles may be increased (Butler7). The membrane and stroma of the erythrocyte contain lipoidal ma- terial which is soluble in alcohol, ether, fatty acids, and bile salts. Addition of these agents to the blood brings about hemolysis, presum- ably by dissolving the lipoidal material present. The hemolysis which occurs with saponin is similar in type, since saponins combine with lipoids, the compound being soluble in water. The hemolytic properties of serum, whether they are found to be normally present when the bloods of certain animals are mixed or to be produced artificially by the injection of foreign red cells, furnish a subject of great interest from the standpoint both of immunology and 96 THE BLOOD AND THE LYMPH of clinical medicine. The hemolytic serum produced by the injection of foreign corpuscles owes its activity to two substances. The one called the amboceptor, or immune body, is specific against the type of cell injected and is increased during immunization. The second body is the complement; it is nonspecific, and is not increased dur- ing immunization. Complement is destroyed by heating the serum for one hour at 55° C., leaving the amboceptor alone present. Corpuscles placed in such serum are not hemolyzed until complement either from fresh immune or from nonimmune serum is added. The serum of animals possessing natural hemolytic properties towards the corpuscles of other animals likewise owes its effect to the joint action of amboceptors and complement. Ordinarily the serum from animals of one species does not exhibit hemolytic properties to blood from another animal of the same species. In unusual cases, however, the serum of an animal will produce hemol- ysis of the corpuscles of an animal of the same species. Such sera are said to possess isohemolysins. The fact is of great importance in the transfusion of blood from one individual to another. The cause of the acute hemolysis which occurs in the disease parox- ysmal hemoglobinuria is not known. It is probably due to the presence of a hemolytic substance Avhich unites with the blood corpuscles at temperatures below the normal body temperature, since the attack fol- lows exposure to cold, and blood from patients subject to the condition may be hemolyzed in vitro by cooling and subsequently heating it. LEUCOCYTES There are a number of varieties of white cells in the blood. These are differentiated from one another by their shape, staining properties, and the granules in their protoplasm. We may divide them into two main groups — nongranular mononuclear cells and granular polynuclear cells. The nongranular mononuclear cells are termed lymphocytes. Two va- rieties are differentiated, the small and the large. The small mononuclear leucocyte makes up from 23 to 28 per cent of the total leucocytes and the large mononuclear, from 2 to 4 per cent. The polynuclear leucocytes are divided into three groups according to whether their granules stain with basic, neutral or acid stains. The leucocytes that stain with basic dyes, or the basophile cells, are very few, making up less than one per cent of the total count. Likewise the acid-staining granular cells, acidophile, are few, comprising from 2 to 4 per cent of the total count. The most numerous are the neutrophiles, THE BLOOD CELL 97 or the polynuclear leucocytes, with neutral-staining granules. These comprise from 65 to 75 per cent of the total count. Another type of white cell is known as the transitional cell, because it was supposed to represent an intermediate form between the mono- and polynuclear cells. Probably such transitions do not occur, and the transitional leucocyte is related to the mononuclear cells. The polynuclear cells originate in the bone marrow, and for this reason .have been termed myeloid cells. They develop from cells in the bone marrow termed myeloblasts, wrhich are nongranular and con- tain a large nucleus. In the course of development the characteristic granules appear, and the nucleus remains round and later becomes lobulated. These intermediate forms are called myelocytes. The mono- nuclear cells originate in the lymphatic tissues of the body. The leucocytes possess the ability to make ameboid movement and to ingest foreign particles which may be presented to them. On ac- count of this latter ability they are commonly called phagocytes. In the process of inflammation the leucocytes assemble at the spot which is the seat of the injury or infection, and remove the foreign organism or necrotic tissue by ingesting and digesting it. It is not definitely known whether or not the lymphocytes func- tion as phagocytes. Other functions besides those as phagocytes have been ascribed to the white cells, but they are not universally ac- cepted. The number of leucocytes in the blood is subject to con- siderable variation. They normally number between 6,000 and 8,000 per c.mm. At the height of digestion and after strenuous exercise there is usually a small increase, and under pathologic conditions, especially in infectious diseases, this becomes quite marked. Some infections increase the polymorphonuclear cells, while others add to the lymphocytes. The factors governing the type of increase are not fully known, nor are the functions of the various forms differentiated. The Blood Platelets These are small oval particles about 3 p in diameter, which are found in large numbers (250,000 to the c.mm.) in the blood. They are sup- posed to be formed from particles of protoplasm which are pinched off from the large blood cells in the bone marrow. Their biological and chemical properties are not understood. They probably play a very important role in the coagulation of the blood (see page 103). CHAPTER XII BLOOD: BLOOD CLOTTING On leaving the blood vessels, the blood clots so as to form "a plug, which assists in preventing further hemorrhage. The clotting must therefore be considered as a protective mechanism against excessive draining of blood out of the organism. When the wounded vessels are small, the clotting, along with constriction of the damaged vessels and the formation in them of thrombi containing large numbers of platelets, serves to effect complete stoppage of the hemorrhage even though the blood pressure may not have become materially reduced. The greater loss of blood from larger vessels causes the arterial pressure to fall, and this enables the clot to stiffen and seal the wound before the pressure again rises. When the clotting power of the blood is subnormal, life is endangered by even trivial wounds; under these conditions the smallest surface scratch may continue to bleed exces- sively in spite of whatever local treatment is applied. The most ex- treme degree of this condition occurs in hemophilia, a disease which is characterized by a most interesting family history — namely, that although it affects only certain of -the male members of a family, yet it is transmitted from generation to generation by the female side alone. The disease has existed in certain of the royal families of Europe for many generations, which has made it possible by con- sulting the genealogic trees to demonstrate the infallibility of this law of inheritance. The clotting of the blood is also either depressed or increased in a variety of physiologic and pathologic conditions. We shall, however, defer further consideration of these until we have learned something of the nature of the factors which are responsible for the process itself. The Visible Changes in the Blood During Clotting In a few minutes after it leaves the blood vessels, the blood forms a jelly-like clot, which adheres to the walls of the container in which the blood is collected and soon becomes so solid that the vessel may be inverted without spilling any of the blood. Clotting is now said to be complete. The clot soon begins to contract, and as it does so, drops of clear fluid or serum become expressed and float on the surface of the 98 BLOOD CLOTTING 99 clot or collect between it and the walls of the container, so that after some time the clot breaks away from the container and comes to float in the serum. The latter may be perfectly clear, but usually is more or less opalescent, partly because of the presence of fat, and partly be- cause of leucocytes which have migrated out of the clot on account of their power of diapedesis. If a drop of freshly shed blood is examined under the microscope, it will be observed that the first step in clotting consists in the formation of fine threads radiating from foci, which are undoubtedly the blood platelets. The fine threads are called fibrin. They multiply rapidly, so as to form an interlacing meshrwork which entangles the red blood corpuscles and leucocytes. By the use of the ultramicroscope (page 52), How ell1 and others have observed that the fibrin (produced by adding thrombin to oxalated plasma) is really deposited in the form of fine crystalline needles — "fibrin needles" — which become packed together as they increase rapidly in numbers. Although the process of clotting consists therefore in the conversion of a hydrosol into a hydrogel (see page 60), it is a unique process; a solution of the blood protein which is responsible for the formation of the fibrin (fibrinogen) may, like other colloidal solutions, be precipitated in a variety of ways, but it is only when the conditions are favorable for blood clotting that fibrin needles, and therefore fibrin threads, are formed. The blood of invertebrates forms a structureless gel when it clots (Ho well). Methods of Retarding Clotting of Drawn Blood To understand the nature of the clotting process and the factors that are responsible for its occurrence, it is advantageous to simplify the conditions somewhat by getting rid of the red corpuscles and most of the other formed elements of the blood and then using the fluid in which these are suspended in living blood — namely, the plasma. This separation of blood into corpuscles and plasma is readily effected either by sedimentation or by centrifuging after measures have been taken to inhibit or greatly delay the clotting process. The methods used for this purpose are numerous. A few of the most important are as follows: (1) Keeping the blood at a temperature very slightly above freezing point. This method is, however, not very effective unless the blood is immediately received into narrow vessels placed in ice and the tempera- ture kept most strictly at the low level. In the case of horses' blood and other slo\vly clotting bloods, the method succeeds without these precau- tions. (2) Receiving the blood through a strictly clean and smooth can- nula, coated with a layer of paraffin or vaseline, into a vessel similarly coated. This method is of practical importance when it is necessary to 100 THE BLOOD AND THE LYMPH transfuse blood without making a vessel-to-vessel anastomosis. (3) Mix- ing the blood with chemicals that are capable of removing the calcium from solution. Such reagents are potassium or sodium oxalate (in a con- centration of 0.1 per cent after mixing), and sodium fluoride and sodium citrate (2 per cent solution, with one part of the solution to four parts of blood). (4) Mixing the blood with certain neutral salts, particularly the sulphates of sodium and magnesium (one part of 27 per cent solution of magnesium sulphate mixed with four parts of blood). Blood thus treated is known as "salted blood," and the plasma separated by centri- fuging, as "salted plasma." Clotting is readily induced by adding water to the salted blood or plasma, and in this way diminishing the concen- tration of the salts. (5) The addition to blood of one of a class of sub- stances known as antithrombins. Leech extract or the purified substance separated from it, knowrn under the trade name of "hirudin," and sub- stances present in blood removed from animals after they have been injected with peptone solutions, are examples. The methods which have just been described are those applied to blood after it has left the blood vessels. Another interesting group of anti- coagulants prevent clotting only when injected into the blood vessels of the living animal. The most powerful example of this group is snake venom, certain varieties of which can prevent clotting in the dosage of %QO °f a milligram for each kilogram of body weight. Similar but much less potent effects are produced by the injection of several proteolytic enzymes, but most attention has been paid to the effect of commercial peptone injected in solution intravenously in the proportion of 0.3 gram to each kilogram of body weight. Blood subsequently removed up to about half an hour or more does not clot, and as we have already seen, if added to blood from another animal, materially retards clotting. This group of intra vitam anticoagulants is particularly interesting, since none of the substances belonging to it is capable of preventing clotting of blood when mixed with this after it has been shed. Their action therefore obviously depends on the production of some substance in the body, probably, as we shall see later, in the liver, since they fail to act after the removal of this organ from the circulation (see page 111). The time of clotting varies greatly according to the conditions under which the blood is collected and the animal from which it is derived. Human blood, for example, received into a test tube from a puncture through the skin may clot at any time within three or ten minutes, five minutes being taken as an average time for blood kept at a temperature of about 20° C. This time may be considerably shortened by increasing the extent of foreign material with which the blood comes into contact, and more particularly by whipping the blood with a bunch of twigs or BLOOD CLOTTING 101 wires. In this latter case, however, the clot does not form in the usual manner, but the fine threads of fibrin collect on the twigs or wires, leav- ing behind the blood serum with the corpuscles still suspended in it. The fibrin removed in this way may then be washed free of. adherent serum. The serum and corpuscles now form defibrinated blood, which is used for many physiologic purposes. Clotting is also greatly acceler- ated by allowing the blood to flow over exposed tissues. Something is evidently added to it from the tissues which accelerates the clotting process, this influence being particularly marked in the case of blood of the lower vertebrates. When the blood of the bird, for example, is received through a cannula inserted directly into a vessel with as little injury to the walls as possible, it very slowly clots if at all, but soon does so if the blood is allowed to come into contact with excoriated tissues, or if it is mixed with tissue extract, such as that of muscle. Clotting is considerably accelerated by warming the blood. The ap- plication of a cloth or tampon well wrung out with hot physiological saline to a wounded surface is a most efficient means of allaying hem- orrhage from vessels too small to ligate. The Nature of the Clotting Process Plasma obtained by centrifuging blood that has been prevented from clotting by one of the foregoing methods can be made to clot by removing the inhibiting influence ; for example, in cooled plasma by warming the blood to room temperature, in salted plasma by diluting it with at least an equal volume of water, and in decalcified plasma by adding a suffi- cient amount of soluble calcium salts to combine with all the added oxalate and leave a small trace of calcium salts in excess. The first question concerns the source of the fibrin, and the answer to it is furnished by comparing the composition of bood plasma with that of serum. Though both of these fluids contain the proteins, albumin and globulin, in approximately the same concentrations, the plasma also contains another protein not unlike globulin in most of its reactions, but distinguished from typical globulin in that it is precipitated by half-saturation writh sodium chloride, in which typical globulin is solu- ble, and is more readily coagulated by heat. To produce half-saturation of the plasma with sodium chloride, equal volumes of plasma and satu- rated sodium-chloride solution are mixed together. The precipitate of fibrinogen, as the substance is called, is then collected at the bottom of the tube by centrifuging and is washed several times by decantation with half-saturated sodium-chloride solution. The washed precipitate, dis- solved in weak saline solution (preferably containing a trace of bicar- bonate), will then be found to clot under certain conditions. 102 THE BLOOD AND THE LYMPH The next question concerns the nature of the conditions that cause the fibrinogen to clot. When a fibrinogen solution is mixed with a few drops of blood serum, a clot usually forms, which however is not the case when plasma is added or when the serum is heated before adding it. Because a small quantity of serum is capable of causing the clotting of a large quantity of fibrinogen solution or plasma, it is supposed that the active substance present in it is of the nature of a ferment — fibrin ferment or thrombin. It must be pointed out, however, that there is considerable doubt whether this active body is really of the nature of a ferment or enzyme. For example, although heated serum does not cause clotting, thrombin, prepared from serum by the method about to be described, in the absence of inorganic salts can withstand even a boiling temperature. Moreover, true enzymes are characterized by the fact that, like other catalytic agents, a very minute quantity can effect a change in an indef- inite amount of substance without the enzyme becoming used up in the process (page 72). When thrombin is allowed to act upon a fibrinogen solution, on the other hand, it is said that only a fixed amount of fibrin can be formed when a small amount of thrombin is added. Neither does this amount increase when the time of reaction is prolonged. Whatever may be the significance of the foregoing facts, it is impor- tant to know that the clotting substance, thrombin, can be isolated from blood serum in a tolerably pure condition. For this purpose blood serum is allowed to stand under a large volume of alcohol for a week or two ; the precipitate is then collected and rubbed up with water, which extracts the thrombin from it, leaving the serum protein in a coagulated state. The resulting watery solution of thrombin may be further pre- cipitated by alcohol, the precipitate washed in alcohol and redissolved in water, yielding ultimately a solution which exhibits very marked co- agulating powers Avhen added to plasma or fibrinogen solution. Throm- bin shows most of the protein reactions but it is not coagulated by heat. As would be expected, a considerable quantity of thrombin remains adherent to the fibrin formed in the process of clotting, and Howell8 describes a very useful method by which it can be separated from fibrin and preserved in a dry condition. Briefly stated, this method consists in allowing washed fibrin to stand overnight under eight per cent sodium-chloride solution, which dissolves the thrombin. The resulting extract is then mixed with an equal volume of acetone, which throws down a precipitate containing the thrombin. To preserve it, the precip- itate is collected on a number of small filter papers, which are subse- quently opened out and dried by exposure to a current of cold air before an electric fan. When the thrombin solutions are desired, the dried pre- cipitates are extracted with a little water. BLOOD CLOTTING 103 Thrombin does not exist in blood plasma, for if a clean and paraffined glass tube is inserted into an artery and the blood collected under al- cohol, the precipitate after standing a few weeks will yield no thrombin when triturated with water. Quite clearly, therefore, the thrombin is produced at the time the blood clots, and the question arises, What is it produced from? It will be remembered that, when the blood is ex- amined under the microscope during the clotting process, the fibrin threads are seen to- start from foci which correspond to the blood plate- lets. It would appear therefore that the thrombin must be derived from some substance that is shed forth from the platelets during the disin- tegration which they undergo shortly after the blood is shed. The sub- stance is called prothrombin. The platelets or their precursors, the megacaryocytes of red bone marrow, are probably not its only source, for clotting may occur in the complete absence of platelets, when it appears to come from the leucocytes. Prothrombin appears plentifully in the fluid used to perfuse red bone marrow outside the body (Drinker and Drinker9). To sum up what we have so far learned, it may be stated that the process of clotting starts with the disintegration of blood platelets and probably of leucocytes, as a result of which there is shed forth into the plasma a substance called prothrombin, which immediately afterward becomes activated or converted into thrombin. The thrombin then at- tacks a protein present in plasma called fibrinogen, producing from it in thread-like form the insoluble protein, fibrin. But this does not com- plete the history, for at least two other important factors come into play; the one is the presence of soluble calcium salts, and the other that of peculiar substances derived from the tissues outside the blood vessels and called thromboplastic substances or thromboplastin (Howell). We must now consider the action of these two factors. The Influence of Calcium Salts. — As already explained, the proof that soluble calcium salts are necessary for clotting is furnished by the ob- servation that the process is entirely prevented when the freshly drawn blood is mixed with soluble oxalate. To this proof, however, objection might be made on the score that the oxalate per se inhibited the clotting. That such is not the case is indicated by the fact that, if the oxalated blood or plasma is dialyzed against physiologic saline solution till all the soluble oxalate has been removed from it, clotting is still absent but immediately supervenes if some soluble calcium salts are added. The question arises as to how the calcium ion acts. Two possibilities exist: (1) that it 'is concerned in the conversion of fibrinogen to fibrin, and (2) that it is necessary for converting prothrombin into thrombin. It can quite readily be shown that it is by the second of these processes 104 THE BLOOD AND THE LYMPH that the calcium acts; for example, clotting occurs when purified throm- bin is added to dialyzed oxalate blood or plasma or to a pure solution of fibrinogen. Citrates prevent clotting by forming calcium citrate, which although soluble does not ionize in solution. It is the free calcium ions that are important. The action of the fluoride is somewhat mysterious, for it has been found that to produce clotting in fluoride plasma the sim- ple addition of calcium chloride will not suffice; thrombin itself must be added as well. Some authors assert, however, that if the calcium chlo- ride is added cautiously to " fluoride" blood, it will induce clotting (Rettger).- . In any case it appears that the fluoride does something more than precipitate the calcium; possibly it prevents the breaking up of platelets and leucocytes. The Influence of the Tissues. — As already stated, when slowly clotting blood, like that of a bird, is collected through a sterile glass tube into a thoroughly clean vessel and immediately centrifuged, the plasma will often remain indefinitely unclotted. If an extract of some tissue, such as muscle, is added, however, the plasma immediately clots. To a much less degree, the same phenomenon is exhibited by mammalian plasma when it is collected in a similar manner. From these observations the conclusions may be drawn that the tissues furnish some substance as- sisting in the clotting process, and that this substance is also formed from certain elements present in mammalian but not present in avian blood. The absence of platelets from the latter blood suggests that these must be the source of the activating substance in mammalian blood. It is plain that this tissue factor in clotting is of importance in hasten- ing the process when an animal is wounded. Before attempting to formulate an hypothesis that will explain the process of clotting, it is necessary to call attention to one other impor- tant fact. This refers to the presence in blood of a substance that pre- vents clotting and is hence called antithromfoin. Antithrombin.is pres- ent in normal blood, for a given specimen of pure fibrinogen will clot less rapidly when mixed with serum to which some oxalated plasma has been added than with an equal amount of the same serum correspond- ingly diluted with a solution of soluble oxalate. A striking increase in the concentration of antithrombin in blood can be brought about by rapidly injecting a solution of commercial peptone into the blood ves- sels fifteen to thirty minutes before bleeding. The peptonized blood or plasma will remain fluid for many hours, if not indefinitely. That the failure of this blood to clot depends on the presence of some anticlotting substance, and not upon the absence of one of the necessary clotting sub- stances (fibrinogen, thrombin, etc.), is evidenced by the fact that the addition of some of it to a mixture of thrombin and fibrinogen inhibits BLOOD CLOTTING 105 the coagulation, which it does not do, however, if it is first of all heated to 80° C. and filtered free of the coagulated protein. Moreover, the antagonistic action is quantitative in the sense that a fixed amount of the peptone-plasma inhibits the action of a fixed amount of thrombin. The source of antithrombin in the body appears to be mainly at least the liver, for it has been found: (1) that peptone injection into an animal from which the liver has been removed does not cause antithrombin to be formed (Denney and Minot) ;10 (2) that peptone injections into the portal vein cause antithrombin to appear in the blood much more rap- idly than when the injection is made into a systemic vessel; and (3) that, when the liver is perfused outside the body with a perfusion fluid con- taining peptone, antithrombin accumulates in the perfusion fluid. A fluid containing a high concentration of antithrombin is secreted by the so-called salivary gland at the head end of the leech. The func- tion of the fluid is to prevent clotting of the blood, so that the animal may continue to suck it without interference by clotting. After apply- ing leeches for medicinal purposes it is therefore necessary to wash the wound thoroughly with water so that all traces of .the antithrombin may be removed ; otherwise the bleeding may continue for a considerable time. Practical use is made of this effect of the leech to prevent clotting of blood outside the body, or it may be used to inhibit coagulation intra vitam in experiments where clotting would otherwise interfere with 'their prog- ress; for example, in crossed circulation experiments (page 365) and in experiments in vividiffusion (page 607). For such purposes the leech head is cut off and extracted either with saline or by treatment with chloroform, which removes other proteins from the saline solution leav- ing a strong antithrombin, known under the trade name of "hirudin." At temperatures about that of the body the action of antithrombin is greatly augmented. In animals like the mammals in which the content of antithrombin is small, this may be important in maintaining the flu- idity of the blood (Howell). Blood containing antithrombin can be made to clot by the addition of thrombin, and therefore of blood serum. CHAPTER XIII BLOOD: BLOOD CLOTTING (Cont'd) THEORIES OF BLOOD CLOTTING Attempts to link all the foregoing facts together in the form of a simple theory have not so far been entirely successful. All agree that the fibrin is derived from fibrinogen by the action of thrombin, the points in dispute being those which concern the origin of the thrombin and the mode of action of the calcium and thromboplastic substances. The theory most widely accepted in Europe is that of Morawitz, according to which the thrombin exists in living blood in an inactive state called thrombogen (prothrombin), which becomes converted into thrombin by the simultaneous action on it of soluble calcium salts and of thrombo- plastic substances furnished by the tissue cells in general and by the cellular elements of the blood platelets and leucocytes. According to this view the thromboplastic substance, aided by the presence of calcium ions, converts thrombogen (prothrombin) to thrombin. It acts there- fore as a kinase and is called thrombokinase. The fundamental fact of this theory, then, is that kinase is necessary for the union of the cal- cium Avith prothrombin — a fact, however, which is challenged by Howell, who states that prothrombin may be converted to thrombin by the action of calcium ions alone. This investigator believes that the thrombo- plastic substance acts not as a kinase but because it neutralizes anti- thrombin, which is constantly present in the blood, and the function of which is to prevent the calcium from uniting with the prothrombin to form thrombin. Howell's theory in his own words is as follows: "In the circulating blood we find as constant constituents fibrinogen, pro- thrombin, calcium salts and antithrombin. The last named substance holds the prothrombin in combination and thus prevents its conversion or activation to thrombin. When the blood is shed, the disintegration of the corpuscles (platelets) furnishes material (thromboplastin) which combines with the antithrombin and" at the same time liberates more " prothrombin; the latter is then activated by the calcium and acts on the fibrinogen." Antithrombin can also prevent the action of thrombin on fibrinogen. As already pointed out, the thromboplastin can be de- rived from the blood itself in the mammals, but only from the tissues jn the lower vertebrates. It is interesting to note that the thromboplastin 100 BLOOD CLOTTING 107 can be extracted from the tissues by fat-solvents, and that it appears to belong to the class of phosphatids, being indeed closely related to, if not identical with, kephalin (Howell). Intravascular Clotting The practical application of the theory of blood clotting concerns the manner in which the blood is maintained in a fluid condition in the blood vessels, and the disturbance of this function causing intravascular clot- ting. According to the one theory, the blood is maintained fluid by the absence from it of any considerable quantity of kinase, and according to .the other, by the presence in it of an amount of antithrombin suffi- cient to prevent the union of calcium with prothrombin. The fluidity is maintained even when large amounts of thrombin or of blood serum, which contains this substance, are injected into the living animal. We can best explain the immunity of the blood to the action of thrombin un- der these circumstances as being due to the instantaneous appearance in it of antithrombin in amounts sufficient to prevent the action of thrombin on fibrinogen, for, as stated above, it is claimed by Howell that anti- thrombin has this influence as well as that of preventing the conversion of prothrombin into thrombin. Intravascular clotting may be brought about by a variety of means: (1) Mechanical damage to the lining of the blood vessels; after the ap- plication of a ligature, for example, the damaged endothelium is soon covered by a clot, which gradually becomes firmer and firmer, and may spread up the vessel to the next branch. (2) The presence of foreign substances in the blood. Emboli, for example, are apt to cause clots to form at the places where they stick, namely, in the smaller vessels. Clotting is also a frequent occurrence when there are local dilatations of the cardiovascular tube, and it may occur under imperfectly understood conditions causing the condition known as thrombosis. (3) An inter- esting variety of intravascular clotting is that caused by the intrave- nous injection of saline extracts of cell-rich tissues, such as the thymus, lymph glands or testes (Wooldridge). By precipitation with acetic acid and digestion with peptone, a residue can be obtained from these extracts which, when dissolved in alkali, has a very pronounced intra- vascular clotting effect. Since these precipitates are very rich in phos- phorus, it is probable that they are of the nature of phosphoprotein (nucleoalbumin). Their action must depend on neutralization of anti- thrombin, according to Howell 's theory, or because they serve as throm- bokinases (according to Morawitz' theory). As a matter of fact, however, the foregoing observation is not com- pletely explained by either theory. If in place of making one injection 108 THE BLOOD AND THE LYMPH frequent injections of small amounts of the above material are made, instead of intravascular clotting, a delay in the coagulation time is likely to occur. Indeed, repeated injections of small amounts may en- tirely remove the clotting power of the blood. The readiness with which this so-called "negative phase" appears, seems to depend on the nutri- tive condition of the animal at the time of injection. If a large dose is injected into a fasting dog, for example, thrombosis is confined to the portal area, w-hereas if it is injected into a recently fed animal, the thrombosis is universal throughout the vascular system. The develop- ment of the negative phase is undoubtedly dependent upon some reac- tion on the part of the living cells of the organism, since it does not occur on the addition of similar substances to blood outside the body. The reaction is, indeed, akin to that by which immune bodies in general are produced. For example, a toxin injected in large amount has a cer- tain toxic effect, but in repeated small doses with intervening intervals it leads to the production of an antitoxin. So with the substance in question ; a large dose injected at one time causes a positive effect — clot- ting— but smaller doses frequently injected, the opposite effect — want of clotting. It is probable, as suggested by Starling, that more intensive study of the conditions causing intravascular clotting will throw con- siderable light on the general question of the production of immunity. Measurement of the Clotting Time To measure the clotting time of drawn samples of blood, several con- ditions must be observed. These have been tabulated by Addis11 as follows: 1. The specimens of blood must always be obtained by exactly the same technic. It would introduce serious errors to compare the clot- ting time of one specimen of blood received from an incision of the skin (ear lobe) with that of another collected in a syringe by veni- puncture. 2. The temperature conditions must always be the same. Probably 25° C. is the best temperature to use. Higher temperatures are unsuit- able for two reasons: first, because during its collection the blood will have become cooled to about or below this point, and time would be con- sumed in raising it higher; and second, because the time of coagulation is more and more shortened for each degree that the temperature is raised, this acceleration becoming especially pronounced for tempera- tures above 25° C. Quite apart from the liability to incur errors inci- dent to measurement of shorter periods of time, observations at higher temperatures necessitate most rigorous adherence to a fixed temperature of the water-bath. Temperatures much below 25° C. are unsuitable, be-. BLOOD CLOTTING 109 cause the clotting sets in gradually and it is difficult to tell precisely when it occurs. 3. The blood must always be collected in the same sort of vessel and come in contact with the same kind and amount of foreign material. To this it may be added that the receiving vessel must be scrupulously clean ; any trace of old blood clot or of serum is especially to be guarded against. 4. The end point must be sharp. It is here that the greatest technical difficulties are met with in making precise measurements, and it is greatly to be desired that different investigators should adopt some uni- form method. For experimental purposes the method of Cannon and Mendenhall12 is no doubt the best, and it has the added advantage of giving a graphic record of the observations. The accompanying figure (Fig. 19) shows the principle of the method. The blood is received through a standard cannula (C) into a tube (T) 5 cm. long and of 5 mm. t R1 Fig. 19. — Diagram of the graphic coagulometer. The cannula at the right rested in a water bath not shown in this diagram. For further description see text. (From Cannon and Men- denhall.)* internal diameter; and a loop (of 2 mm. diameter) at the end of a copper wire (Z>), which is 8 cm. long and 0.6 mm. in diameter, is al- lowed to fall gently into the blood at. regular intervals. The upper end of the wire is articulated with the short arm of a light lever so counter- poised that when the stop (R), which ordinarily holds it in a horizontal position, is released, the wire, now having a net weight of 30 mg., falls on the blood in the tube. The long arm of the lever is provided with a 'writing point, which is made to inscribe its movements on a drum. So long as the blood is unclotted the loop sinks into it when the lever is released and a vertical line is traced, but whenever clotting occurs the loop sticks on the blood and the writing point does not rise. For clinical purposes where blood collected in a syringe by venipuncture is used, the method of Howell13 is most accurate. It consists in placing, *Am. Jour. Physiol., May 1, 1914, xxxiv, No. 2. 110 THE BLOOD AND THE LYMPH 2 or 4 c.c. of the blood in a wide tube (of 21 mm. diameter) that has been cleaned by a bichromate-acid mixture. The period that elapses between the moment of the entry of fluid into the syringe and that at which the clot has become firm enough so that the tube can be inverted without spilling any blood, is taken as the clotting time. Since the blood does not come in contact with exposed tissues, it takes from 20 to 60 minutes to clot by this method. For routine clinical examination of blood taken from a skin wound Brodie and Russel's method14 is most satisfactory. This consists in principle in observing a drop of blood, under the low power of the microscope, while a fine current of air is gently blown against it at regular intervals in a tangential direction. Until clotting sets in, the individual corpuscles move freely in a circular direction, but as soon as clotting begins they move in masses which soon tend to become fixed so that, although they move somewhat when the air impinges on them, they immediately return to their original position when the current is discontinued. When clotting is complete, the air current merely Fig. 20. — Coaguloniett- r. The drop of blood is placed on the lower end of the glass cone and the air stream is directed against it from the side tube shown by the black dot. The apparatus is placed on the stage of the microscope and the drop observed by the low power. presses on the corpuscles at one point. By this method the clotting time averages five minutes. A convenient apparatus for this method is that of Boggs, which is shown in Fig. 20. It consists of a truncated cone of glass, projecting into a moist chamber provided with a tube on the side so arranged that when air is blown into the chamber, it strikes the drop of blood placed on the end of the cone tangentially. Blood Clotting1 in Certain Physiologic Conditions Besides the experimental conditions already enumerated as changing the clotting time in the blood of laboratory animals, special mention must be made of the influence of epinephrine injections, of conditions supposed to cause a hypersecretion of this hormone, of the emotions, and of hemorrhage. Epinephrine added to drawn blood does not affect the clotting time, but if small amounts are injected intravenously or even subcutaneously, a marked decrease occurs (Cannon and Gray; cf. Cannon, loc. cit.). Larger injections may have the opposite effect, and intermediate amounts BLOOD CLOTTING 111 may cause at first a prolongation and later a shortening of the time. These results with larger doses are related to Howell's observation that repeated doses of relatively large amounts of epinephrine in dogs may so greatly retard coagulation as to make the animals practically hemo- philic. It was further found by Cannon and his coworkers that epineph- rine does not influence the clotting time when injected into animals from which the abdominal viscera have been removed from the circulation by ligation of the inferior vena cava and the abdominal aorta. In the light of the influence which destruction of liver cells (by phosphorus, chloro- form, etc.) is known to have in lengthening clotting time, it would seem reasonable to conclude that it must be through this organ that epineph- rine develops its clotting effects. Stimulation of the splanchnic nerves also shortens the clotting time, and it would appear that this action depends on the resulting hyperse- cretion of epinephrine (page 746), for it is not observed following stimula- tion of the nerves in animals from which the adrenal glands have been excised (Cannon and Mendenhall). The interesting suggestion is made by Cannon that the shorter clotting time observed in animals showing strong emotions of fright or fear may also be due to the hypersecretion of epinephrine which this worker believes accompanies such states. Blood Clotting in Disease With the factors concerned in the process so wrapped in mystery, it is not surprising that the underlying causes responsible for delayed or de- ficient clotting of blood in diseased conditions or for the formation of intravascular clots (thrombi) are little understood. According to How- ell 's theory of the nature of the process, which is the most satisfactory at the present time, abnormal clotting might be due to the following causes: (1) A diminished amount of fibrinogen. This occurs when the hepatic cells are greatly damaged, as in poisoning by chloroform or phosphorus and in such diseases as acute yellow atrophy and yellow fever. In many cases of chronic cirrhosis of the liver, as Whipple, etc.,15 have shown, the blood also clots feebly because of deficient fibrinogen. It should be pointed out that it is not so much the clotting time that is increased in such cases, as the firmness or consistency of the clot that is affected. 2. A deficiency in prothrombin. In the condition known as "melena neonatorum," undoubted benefit is derived from intravenous injections of blood serum or by direct blood transfusions, probably because throm- bin or prothrombin is thus furnished. 3. A deficiency of thromboplastin. Since this substance is derived from both blood cells and tissue cells, it does not seem likely that a deficiency 112 THE BLOOD AND THE LYMPH could ever occur. Certain observers, however — Morawitz, for example — lay great stress on this as an important factor in hemorrhagic diseases. 4. An excess of cmtithrombin. The undoubted increase in this substance that can be brought about experimentally by injecting hirudin or pep- tone into animals, has stimulated careful search for a similar increase in the blood in clinical conditions in which abnormal blood clotting is one of the symptoms (Whipple10). Antithrombin is said to be increased in septicemia, pneumonia, miliary tuberculosis, etc. 5. A deficiency of calcium ions. Although at one time it was supposed that this might be responsible for the feeble clotting in hemophilia, it has not been found, after very extensive trials, that the exhibition of Ca salts in any way relieves the condition. It is said, however, that the slow coagulation seen in obstructive jaundice is decidedly shortened by treatment with calcium salts. One thing stands out prominently in connection with the whole problem, and that is the close relationship of the blood platelets to the clotting process. From these cells are derived, according to Howell, not only the prothrombiri but also, as from other cells, thromboplastin. It is not sur- prising therefore to find that decided alterations in the platelet count occur in cases of faulty blood clotting, and that local accumulations of these elements within the blood vessels, produced by their clumping to- gether or agglutinating, is followed by a formation of local clots, as in thrombosis. Hemorrhagic Diseases In many of the so-called hemorrhagic diseases (acute leucemia and aspastic anemia) and in the hemorrhagic varieties of diphtheria and smallpox, the platelet count drops from its normal of between 200,000 and 800,000 per cubic millimeter to well below 100,000, and indeed in these conditions it is frequently difficult to find any platelets. Samples of blood clot outside the body within the normal time, but the clot is soft and usually fails to retract in the normal manner. It is on account of this, rather than slow clotting that the hemorrhage continues, so that in appraising the gravity of the symptom it is best to measure not the clot- ting time but the time that it takes for bleeding to cease from a small skin wound, as in the lobe of the ear. This can be very accurately done by applying blotting paper at regular intervals to the puncture (Duke17). The most interesting and at the same time the most mysterious of all conditions in which blood clotting is interfered with is hemophilia. The clotting time is longer than normal, but even after the clot forms, bleed- ing is likely to continue because the clots are very readily displaced. Both clotting time and bleeding time are increased. So far no change in the BLOOD CLOTTING 113 clotting factors of the blood has been demonstrated in this disease ; the corpuscles and the platelets are normal in numbers, fibrinogen and cal- cium salts are normal, and, as Howell has shown, there is no excess of antithrombin. One significant fact, however, is that the addition of thromboplastin or of its active ingredient, kephalin, greatly shortens the clotting time of the blood when it is removed by venipuncture. In agree- ment with this observation it has been found that hemophilic blood clots much more rapidly, indeed sometimes in the usual time, if it is allowed to flow over cut or damaged tissue and so become mixed with thromboplas- tin. These facts taken together would seem to indicate that the fault must lie in a deficiency in prothrombin, and since this is derived mainly from the platelets, which however are not decreased in number, we must further assume that these elements have undergone some qualitative change preventing their disintegration. An accompanying defect in their agglutinating properties would at the same time explain their fail- ure in hemophilia to clump together at the' site of the hemorrhage so as to block the smaller vessels with thrombi; hence the prolonged bleeding time even after clotting has occurred. Thrombus Formation The first formed portion of a thrombus is paler than those formed later, because it contains excessive numbers of platelets; and it seems clear that it is by agglutination of these into m&sses, which then stick in the blood vessels and by disintegrating shed forth prothrombin and thrombo- plastin, that the clotting starts. This platelet agglutination may result from stagnation in the bloodflow, or from roughening and damage to the vessel walls. Stagnation may be due either to failure of the circulation as a whole as in heart disease, or to local physical alterations in the vas- cular tube, setting up conditions in which eddy currents with stagnant pools of blood are formed, such as will occur at places where the vessels suddenly become wider, as in varicose veins, in aneurisms and at the sudden bend of large veins. The first formed (platelet) thrombus is fol- lowed by one of a darker color, which fills the vessel up to the next anastomotic branch. Similar stagnation may also follow the obstruc- . tion caused by lodgment of emboli in the smaller vessels (air, foreign bodies in fine suspension, bacteria, etc.). The thrombi in such cases are very small and occur particularly in the capillaries of the liver, spleen, and lungs. The small thrombi often serve as foci from which clotting spreads into the larger vessels, this being often encouraged by an increase in the coagulability of the blood. When the intima is inflamed, it is pos- sible that excessive amounts of thromboplastin are produced and that this neutralizes the antithrombin in blood moving so slowly that it is not 114 THE BLOOD AND THE LYMPH replaced by fresh blood before clotting ensues, or it may be that sub- stances derived from the inflamed tissue cause the platelets to aggluti- nate. The increased clotting often observed after the injection of hemo- lytic agencies (foreign sera, snake venom, etc.) may also be due to platelet agglutination. Like the thrombosis following embolism, the clotting occurs at first in the capillaries, the initial thrombi containing masses of platelets along Avith skeletons of blood corpuscles and cells from the blood-forming organs. CHAPTER XIV LYMPH FORMATION AND CIRCULATION GENERAL CONSIDERATIONS Lymphatics are modified veins. They grow from the veins in embry- onic life as buds of endothelium, which are first visible in the human embryo in the sixth week of development. The earliest outgrowth oc- curs from the internal jugular vein, and the endothelial buds soon be- come hollow and join together, forming first a plexus and subsequently a sac, from which again lymphatic vessels made of endothelium grow out to invade the skin of the head, neck, thorax and arm, and partly the deep structures of the head. The sac is ultimately transformed into groups of lymph glands. At a later stage similar nodes develop from certain of the abdominal veins, forming a retroperitoneal sac, from which grow out the lymphatics of the abdominal and, to a certain extent, of the thoracic viscera. A similar pair of sacs also develops from the iliac veins supplying the lymphatics for the skin of the legs and abdominal walls. The retroperitoneal and iliac sacs then become connected with the jugular sac by means of the thoracic duct. In the embryo there are no valves in the lymphatic vessels, so that the whole system can be in- jected either from the thoracic duct or from the skin, showing clearly that the superficial and deep lymphatics are parts of one closed system of vessels. Anatomists have succeeded in tracing the course of the lymphatics in many parts of the body. This knowledge is of great importance in connection with the spread of infections, etc. Lymphatics are abun- dant in the skin, the intestine, and connective tissues, but are absent from the muscle bundles, from the hepatic lobules (though present in the connective tissue between them), from the substance of the spleen, and from the central nervous system. The lymphatics have the same functions as blood capillaries, namely, to absorb substances from the tissue spaces. There is some evidence to show that this absorption may be selective. When injections are made into the peritoneal cavity, the pathway of absorption may be either the blood vessels or the lymphatics, according to the nature of the sub- stance injected. True solutions are absorbed by the blood, but granules 115 116 THE BLOOD AND THE LYMPH are taken up by special large cells showing phagocytic powers, and trans- ferred to the lymphatics — for example, those of the diaphragm. A sim- ilar selective absorption is well known in the case of the villi of the in- testine, where fat passes into the lacteals and carbohydrates into the blood. It appears as if lymphatic adsorption, both of solid materials and of solutions, requires the cooperation of phagocytic cells. The newer conception of the lymphatics as a closed system is at vari- ance with the older one, in which they were supposed to get smaller and smaller, and their walls less and less complete until ultimately they faded off into the tissue spaces. These, however, bear no closer relation- ship to lymphatics than they do to blood capillaries. The tissue spaces include all the minute spaces between the fibers and cells of the con- nective tissues and between the parenchyma of the organs and the great serous cavities of the body (pleural, peritoneal), as well as specially developed tissue spaces, forming the subarachnoid spaces of the brain, the scala vestibuli and tympani of the cochlea and the anterior chamber of the eye. The fluids in these spaces — the tissue fluids — are quite dif- ferent from the lymph in the lymphatics both in composition and in function. Indeed, the tissue fluids are among the most varied of all the fluids of the body. The spaces may themselves become linked to- gether so as to form a circulatory system, M'hich is quite independent of the lymphatics. This is particularly the case in the brain, where the tis- sue spaces surrounding every individual nerve cell extend into the sub- arachnoid area, where they drain into the cerebral sinuses through the arachnoidal villi, which exist as lace-like projections of the arachnoid into the dural sinuses, being covered by a layer of mesothelial cells spe- cially abundant at the tips of the villi, where they form cell nests. Ob- servations of the passage of substances in solution by these pathways have been made by injecting potassium ferrocyanide and citrate. of iron into the subarachnoid and subdural spaces and afterwards detecting the presence of the salts by mounting sections in acid media, so as to permit prussian blue to develop. Ordinarily the precipitate is found in or near the villi, but after cerebral anemia it forms in the tissue spaces that surround the nerve cells. There are therefore three fluids concerned in the transference of food materials and gases between the gastrointestinal apparatus and lungs and the tissue cells — namely, the blood plasma, the tissue fluids, and the lymph. The tissue fluid, being in contact with the tissue elements, serves as their immediate nutritive fluid, and it is the function of the blood and lymph to maintain it of proper composition. Everything must be trans- ferred to and from the tissue cells through the tissue, fluid, making it LYMPH FORMATION AND CIRCULATION 117 therefore in many ways the most important of the fluids of the body. In the tissue cells themselves there is also the fluid in which the various colloids and crystalloids that enter into the composition of protoplasm are dissolved. This can be removed from cells only by mechanical means, such as grinding with fine sand in a mortar and subjecting the mass to a pressure of several thousand atmospheres in a hydraulic (Buchiier) press. This is known as the tissue juice. The ultimate exchange of foodstuffs occurs between the tissue fluids and the tissue juices across the cell membrane. The extent and character of this exchange depend on many circumstances, some affecting the cell wall, others, the osmotic and other properties of the two fluids. Obviously, the function of the circulation is to maintain the tissue fluids of correct composition, the blood plasma serving to carry food materials and dissolved oxygen to them (see page 380), but being assisted in the opposite function of re- moval of effete products by the lymph. The lymph is purely a scavenger ; the blood is both purveyor and scavenger. The above description of the lymphatics is not universally accepted by anatomists, certain of whom believe that the lymphatics are developed from tissue spaces and are consequently much more extensive than they appear to be from injected specimens. The above conclusions are based on reconstruction models, made from serial sections of embryonic tissues, in which the lymphatics frequently appear as isolated vesicles without visible connections. The failure of injections to penetrate into the re- moter parts of such a lymphatic system in the embryo is attributed to the discontinuity of spaces, which is, however, removed at later stages of development. The manner of absorption of injected fluids does not, however, sup- port the view that the lymphatics are directly connected with the tis- sue spaces. When all the structures of a part are ligated except the main artery or vein, injected poisons which affect central structures, such as the nerve centers, develop their action as quickly as in the in- tact animal (e.g., strychnine). Similarly, when pigments such as meth- ylene blue are injected into the pleural cavity or subcutaneously, they appear in the urine long before the lymph of the thoracic duct. Such results indicate the pathway of absorption to be the blood rather than the lymph vessels. Through this latter channel absorption proceeds more slowly, but can be greatly assisted by massaging the site of injec- tion. When colored solutions, such as India ink or carmine, are injected subcutaneously, however, a very perfect injection of the neighboring lymphatics may ultimately occur, and through the same pathways mi- croorganisms spread from an infected area. 118 THE BLOOD AND THE LYMPH EXPERIMENTAL INVESTIGATIONS It has proved a most difficult problem to gain any exact knowledge of the production of lymph by experimental means. Starling, some years ago, in repeating many of the experiments of older physiologists in the light of the newer facts of physical chemistry, added much that is of interest, and it is chiefly with his work that we will concern ourselves here. The unequal lymph supply of different regions of the body is strik- ingly demonstrated by comparing the lymph flow from the lymphatics of the leg with that from the thoracic duct. No lymph flows from the former unless the muscles are thrown into activity or the blood is pre- vented from, leaving the limb by ligaturing all the veins. Changes in the arterial blood pressure do not affect the flow. On the other hand, a great increase in the flow from the thoracic duct can readily be induced by disturbances in the blood supply. Obstruction of the portal vein, for example, immediately increases the lymph flow four or five times because of venous congestion in the intestinal capillaries, whilst a still greater increase — perhaps tenfold — is induced by obstruction to the inferior vena cava, which raises the capillary pressure in both the liver and the intes- tines. After ligation of the hepatic lymphatics (at the hepatic pedicle), obstruction of the vena cava no longer causes the outflow of lymph to increase, indicating that the lymph in the last mentioned experiment must have come from the hepatic lymphatics. These results, so far as they go, could be satisfactorily explained on the basis that lymph formation is a filtration process, that is, a process dependent upon difference in mechanical pressure between the blood capillaries and the tissue spaces. The lymphatics would then serve as channels to return this fluid to the blood vessels through the thoracic duct. The difference in the magnitude of the increased lymph flow from increase in capillary pressure in different regions would be dependent on the permeability of the filter, the capillaries of the limbs being much less permeable than those of the intestine, and particularly of the liver. Another fact in conformity with this view concerns the composition of the lymph from the two regions, that from the limb lymphatics being poor in protein, whereas that from the thoracic duct does not fall far behind the blood plasma in this regard. Although filtration may explain the considerable increase in lymph flow produced by extreme changes in capillary pressure, it by no means suffices to explain lymph formation under less abnormal conditions. When a muscle or a gland is at rest, it produces practically no lymph, LYMPH FORMATION AND CIRCULATION 119 but during activity the flow becomes marked. This can not be explained by filtration, but may be accounted for by a physico-chemical process — namely, osmosis. The energy required for the activity of the tissue cell is produced by chemical changes, whereby large molecules become broken down into numerous smaller ones. These smaller molecules are then discharged into the surrounding tissue fluids, the osmotic pressure of which they increase, with the consequence that water is attracted by osmosas from the plasma in the blood capillaries (see page 4). This increases the volume of tissue fluid, which is then drained away by the lymphatics. The increase in molar concentration will also affect the tissue juices, tending to make the cell swell up by absorbing water. In gland cells this extra water is immediately extruded to form the water of the secretion (see page 421). An analogous method of lymph formation is not confined to situations where the capillaries are relatively impermeable, for it also occurs in the liver, the lymph flow from which is greatly increased by the injec- tion of bile salts. A similar process no doubt results from muscular activity, although in this case the tissue spaces must form a continuous system of their own, there being, according to most authorities, no lymphatics. Considerable interest has been taken in the stimulating effect which certain chemical substances have on the secretion of lymph from the thoracic duct. These so-called lymphagogues belong to two classes — crystalline and colloidal. Of the former, glucose, urea, and sodium chloride in hypertonic solution, are the best known. Starling explains their action as dependent upon an increase in the osmotic pressure of the blood. This attracts water into the blood from the tissue juices, and leads to an hydremic plethora, with a consequent increase in capil- lary pressure. If the blood pressure is lowered by hemorrhage before the hypertonic solution is injected, very little stimulation of lymph flow occurs, because there is no available fluid in the tissue to produce the plethora. This observation does not, however, very strongly support the explanation, because so many other disturbances may result from hemorrhage. The colloidal lymphagogues include watery extracts of the dried tis- sues of leeches, crayfish, and mussels, as well as commercial peptone. They probably act by damaging the endothelium of the capillaries, so that filtration occurs more readily. Although their action is displayed more particularly on the lymphatics of the liver and intestines, it is also apparent on the skin capillaries, producing cutaneous edema and the formation of blisters (nettle rash). THE BLOOD AND THE LYMPH EDEMA With such an imperfect knowledge concerning the physiology of lymph formation, it is not surprising that the causes of excessive accu- mulation of fluid in and between the tissue elements should be little un- derstood. All of the conditions which have been mentioned as capable of causing an increased secretion of lymph — such as increase in capillary pressure, hydremic plethora, action of poisons on the endothelium — are likely to cause edema if the lymphatics of the part are simultaneously obstructed. To produce in animals edema of the subcutaneous tissues like that observed clinically, it is, however, necessary that the vascular disturbance be accompanied either by local damage to the capillary endothelium, such as is produced by arsenic or uranium; or by a gen- eral toxemic condition, such as is set up by nephritis. When large amounts of saline solution are injected intravenously, extensive ex- travasation of fluid may occur into the liver, peritoneum and intestinal lumen, without any subcutaneous edema. Clinical edemas are of at least three types: 1. The inflammatory edemas, in which the fluid permeates the cells of the inflamed area and does not shift to other parts of the body under the influence of gravity. 2. The nephritic edemas, in which the fluid is more or -less loose in the subcutaneous tissues and readily changes its position, and which is accompanied by excess of water in the blood with a corresponding in- crease of sodium chloride ; the percentage concentration of sodium chlo- ride in the blood remains unchanged, but that the other constituents diminished. 3. Cardiac edemas, which are also hypostatic, but are unaccompanied by changes in the relative amount of water and sodium chloride in the blood. The second and third varieties of edema may of course be more or less present together, for the kidneys are likely to become secondarily affected during venous stasis. The salt retention in nephritic edema is very significant. As ex- plained elsewhere, it is revealed by comparing the daily output of so- dium chloride by the urine with the concentration of this salt in the blood. Less salt is eliminated than would be the case in a normal in- dividual with the same percentage of salt in the blood. In many cases also edema can be diminished by withholding salt from the food. Widal and Javal have conclusively shown the relationship of retention of water in the body, as judged by variations in body weight, to the hydremic condition, as judged by the refractive index of the blood serum, and LYMPH FORMATION AND CIRCULATION 121 to the amount of salt in the diet. A very considerable retention of water usually occurs before there is any evidence of edema; indeed, as a result of giving salt, the body weight may increase from five to seven kilograms (10 to 15 pounds) within a day or two without the appear- ance of puffiness. The cause of the edema during salt retention is no doubt closely re- lated to the action of lymphagogues. In a normal person excessive in- gestion of salt is immediately followed by excretion of the excess through the kidney. Where the kidneys are diseased, this excess of salt is re- tained in the blood, raising its osmotic pressure and attracting water from the tissue fluids. This leads to excessive thirst, the imbibed water being used to replace that lost from the tissues. But all the crystalline lymphagogues do not, when present in excess in the blood of nephritic patients, necessarily cause edema; urea, for example, may accumulate considerably without any such effect. The different action is usually attributed to inequality in the diffusibility of the two crystalloids through animal membranes, sodium chloride diffusing much less readily than urea. It is most important to note that the fluid in edema is loose in the tissues and can be drained away by the insertion of tubes. There is absolutely no evidence in support of the claim of Martin Fischer that edema is due to imbibition of water by the colloids of the tissues. This question has been fully discussed elsewhere (page 62). (Monographs) iHowell, W. H.: The Harvey Lectures, J. B. . Lippincott Co., xii, 272. Starling, E. H.: Human Physiology, Lea & Febiger, 1915. sKowe, A. H.: Arch. Int. Med., 1917, xix, 354, ^Williamson, C. S.: Arch. Int. Med., 1916, xviii, 505. sTower and Herm: Proc. Soc. Biol. and Med., 1916, xviii, 505. eRoua and Eobertson: Jour. Exp. Med., 1916, xxiii, 219, 239, 549. TButler, G. G.: Quart. Jour. Med., 1912, vi, 145. SHowell, W. H.: cf. Harvey Lecture; also Am. Jour. Physiol., 1913, xxxii, 264. '^Drinker, C. K., and K. E.: Am. Jour. Physiol;, 1916, xli, 5. loDenny and Minot: Arch, Int. Med., 1916, xvii, 101; Am. Jour. Physiol., 1915, xxxviii, 233. "Addis, T.: Quart. Jour. Med., 1910, iv, 14. 12Cannon and Mendenhall: Am. Jour. Physiol., 1914, xxxiv, 225. "Howell, W. H.: Arch Int. Med., 1914, xiii, 80. "Brodie, T. G.: Jour. Physiol., 1897, xxi, 403. isWhipple, G. H.: Arch. Int. Med, 1912, ix, 365; Jour. Exp. Med., 1911, xiii, 136. leWhipple, G. H.:' Arch. Int. Med., 1913, xii, 637. . i?Duke, W. W.: Arch. Int. Med., 1912, ix, 258. CHAPTER XV BLOOD PRESSURE The object of the circulation is to maintain through the tissues a sup- ply of blood that is adequate to meet their demands for nutriment and oxygen and to remove the effete products of their metabolism. The de- mands vary according to the activities of the tissue, being particularly variable in the case of such tissues as the muscular and the glandular. In studying the physiology of the circulation we have therefore to bear in mind two aspects of the problem: (1) the cause for the continuous bloodflow, and (2) the mechanism by wrhich alterations in this bloodflow are brought about. If we open an artery we shall find that the blood escapes from it under such a pressure that it is thrown to a height of about six feet, that its outflow is proportional to the size of the artery, and that it pul- sates. If, on the other hand, we open a vein, we shall find that the blood wells out without any very evident pressure, and that it flows in a continuous stream, its outflow being the same in a unit of time as that of the artery, provided the two vessels are the only ones supplying the particular area. The general conditions governing the bloodflow are the same as those governing the flow of fluid through any system of tubes. For example, in the city water mains it is known to every one that the rate of outflow from any part of the system depends finally on two factors: (1) the difference in pressure at the beginning and end of the system, and (2) the caliber of the tube at the outlet. We may in- crease the outflow by raising the pressure at the beginning of the system, the caliber of the outlet meanwhile remaining constant, or we may maintain the pressure constant but increase the caliber of the outlet. In the circulation of the bjood, the difference in pressure at the be- ginning and end of the circulation is furnished by the pumping action of the heart, and the alteration of the caliber of the outlet is provided for by the constriction or dilatation of the blood vessels. These simple physical principles indicate the direction which a study of the circulation 122 BLOOD PRESSURE 123 should take. They indicate that our first consideration should be of the mean blood pressure, how it is maintained, and how it can be made to vary. After we have learned this, we may then proceed to a more particular examination of the mechanism of the pump — that is, of the heartbeat; then finally we may proceed to examine the nature of the processes by which the caliber of the arteries is controlled. THE MEAN ARTERIAL BLOOD PRESSURE The first prerequisite to the investigation of the blood pressure, as of any other physical problem, is that we should possess some means by which it can be quantitatively measured. The earliest attempt to accom- plish this was made by the English scientist, the Rev. Stephen Hales, a little over a century after Harvey published his account of the circula- tion of the blood. Hales connected a glass tube nine feet in length with a severed artery of a horse, the connection between the two being made by means of a piece of brass pipe joined to the windpipe of a goose as a substitute for rubber tubing. He found on untying the ligature on the artery that the blood rose in the tube to a height of eight feet and three inches above the level of the left ventricle of the heart, 'and that when at full height it rose and fell with each pulse through a distance of two, three or four inches. Mercury Manometer Tracings The somewhat crude but very significant experiment of Hales clearly established the existence of the enormous pressure at which the blood is made to circulate through the arteries. To render possible a further investigation of the factors on which this pressure depends, it became necessary to invent some more convenient means for its measurement, but this was not accomplished until a century later, when Poiseuille ap- plied the mercury manometer, which Ludwig subsequently adapted so that tracings might be taken (Fig. 21). Having before us such a tracing as shown in Fig. 22, let us consider how it may be used in the study of blood pressure. The first thing we must do is to measure the average height of the tracing above the line of zero pressure; the mean arterial blood pressure is then equal to this distance multiplied by two, because the distance through which the mer- cury has moved up in the limb of the manometer carrying the writ- ing point is only one-half of its total displacement. Since mercury is about 13.5 times heavier than an equal volume of blood, the above measurement must be multiplied by this figure if we desire to express 124 THE CIRCULATION OP THE BLOOD our result in terms of the height to which the blood pressure could raise a column of blood. In arteries of approximately the same size, the mean arterial blood pressure does not markedly vary in different mammals. Thus, in the carotid artery of the dog it averages about 110 to 120 mm. Hg, in that of the cat about 105 to 115 mm., in the rabbit from 90 to 105 mm., in the sheep about 150 mm., in the horse about 200 mm., and in man some- Fig. 21. — Mercury manometer and signal magnet, arranged for recording the mean arterial blood pressure in a laboratory experiment. The pressure bottle (R) is filled with anticoagulating fluid and is connected by tubing with the manometer (M), the cannula for the artery (U) being connected with the T-piece (J). By this arrangement it is possible to flush out the tubing when clotting interferes with the experiment. (From Jackson— Experimental Pharmacology.) where between 120 and 140 mm. The pressure varies in different parts of the vascular system, being greatest in the aorta and least in the small- est arterioles but the fall in pressure — the pressure gradient — does not become very pronounced until the arterioles have become so small that it is no longer possible to insert, a cannula into them; thus, the mean 125 blood pressure in the renal or femoral artery is very little less than that in the aorta. If Ave examine the contour of the tracing which the pressure draws, we shall find that it exhibits two types of wave, small and large ; and if we observe the animal while the tracing is being taken, we shall find that Fig. 22. — The arterial blood pressure recorded with a mercury manometer (lower tracing), along with a tracing of the respiratory movement of the thorax. Note that the beginning of respiration occurs distinctly before the rise in blood pressure. the former are caused by the heartbeats and the latter by the respira- tions— an observation which immediately raises the question as to the trustworthiness of the method, for it will be asked, How can it be that the heartbeat produces an effect on blood pressure which is less than that of the respirations? Obviously the tracing must be faulty in re- gard to the relative significance of the waves. 126 Spring Manometer Tracings The cause of this inaccuracy depends on the inertia of the mercury, an inertia which is so great that the sudden changes of pressure produced by each heartbeat are not able to overcome it, whereas the much less significant but more prolonged pressure changes produced by each respi- ration develop their full effect on the mercury. These facts led investi- gators to seek for instruments in which the inertia error is eliminated, with the result that they invented what are known as spring manometers. Fig. 23. — Hurthle's spring manometer. Many varieties of this instrument have been produced, but for our pur- pose it is necessary to describe the principle of only the simplest and most efficient — the Hiirthle manometer. As shown in Fig. 23, it consists of a variety of tambour, which differs from the ordinary tambour in two important particulars: (1) its chamber is made as small as possible, and (2) it is covered not with an elastic membrane but with one of leather or of thin fluted metal. These two precautions are taken in order to avoid spuri- ous waves set up on account o€ elastic recoil. Such errors are further reduced by filling the tubing and chamber of the tambour with an anti- coagulating fluid. Fig. 24. — Arterial pressure recorded by a spring manometer. The effect of weak excitation of the vagus is seen during the period marked by the signal m. (From Dubois.) Before the tracing taken with the spring manometer can be em- ployed for quantitative measurements, it must obviously be graduated according to some scale. This is accomplished immediately before or after the experiment by connecting the manometer through a T-piece with a pressure bottle, which can be raised or lowered to a specified height, and with a mercury manometer. The displacement of the writing point of the spring manometer corresponding to each 10 mm. Hg of pressure is then written on the tracing. BLOOD PRESSURE 127 The tracings taken with such a manometer, as shown in Fig. 24, are quite different from those with the mercury manometer. It will be seen that now the cardiac waves are decidedly the more pronounced, the respira- tory, being comparatively inconspicuous. Instead of there being a fairly steady pressure in the arteries, this undergoes very considerable altera- tion during each heartbeat.* Examination of this tracing gives us accurate information regarding the blood pressure both between the heartbeats — diastolic, as it is called — and during them — systolic. It gives us a means of telling what must be the dead load of the circulation — that is, the pressure that is constantly present — as well as the live load that is superadded to this by each heart- 160 120 L ine of SYSTOLIC PRESS U BE L/r>e of MEAN PRESSURE Lt'nt D/ASTOLIC Pressure D/ASTOL/C PRESSURE Fig. 25. — Diagram based on experiments on dogs to show the magnitude of the systolic, diastolic and mean blood pressures at different parts of the circulatory system. O is the line of zero pressure, and the letters below it indicate the parts of the system to which the curves refer. (From Brubaker.) beat. This difference is often called the pressure pulse, and in man it amounts to somewhere about 35 mm. Hg. If we take tracings with a spring manometer from different parts of the arterial tree, we shall find that, as we travel towards the periphery, the pressure pulse becomes less and less marked, until finally by the time the capillaries are reached it has almost entirely disappeared. This decline in the pressure pulse can moreover be seen to be dependent more largely on a fall in systolic than in diastolic pressure. In other words, the dead load of the circulation — the diastolic pressure — remains practically constant all along the arte- rial tree, whereas the systolic pressure falls relatively quickly (Fig. 25). *The tracings shown in Figs. 22 and 24 are not typical, (he pulse pressure being too small in the latter and too large in the former. 128 THE CIRCULATION OP THE BLOOD Clinical Measurements The methods of blood-pressure measurement in man have recently become so perfected that the results are almost as accurate as those ob- tained in laboratory animals by direct measurement through the use of cannulae inserted into the vessels. Both the systolic and the diastolic' pressure can be measured with equal facility and 'accuracy! Since the technic for making the systolic measurements was described at a much earlier date than that for the diastolic, it has until recently been the habit with a great part of the medical profession to be satisfied with systolic readings alone. This is most unfortunate, because the knoAvle'dge* which such information gives us is incomparably inferior to that which can be obtained by gauging the diastolic pressure. Until we have learned more about the dynamics of circulation, it would be profitless to go into any details as to the reasons for this statement, but it will soon become self-evident. Suffice it for the present to- state that the diastolic pressure is the more important because it gives us the load which the ves- sels and aortic valves must constantly bear, and the resistance which must be overcome prior to the opening of these valves at the beginning of systole. Moreover, it helps us to gauge the peripheral resistance. The first step in the technic of blood-pressure measurements in man is the placing of an armlet or cuff around the arm or leg. This armlet consists . of a rubber bag at least 12 cm. broad and covered on its outer surface by cloth or leather. The bag is connected by tubing with! a pres- sure gauge and a pump. The pressure gauge may be either an ordinary mercury manometer or one of the numerous gauges built on the aneroid principle that are now on the market (Fig. 26). For measuring the blood pressure in the vessels of the upper extremities, the armlet should be applied around the fleshy part of the upper arm and for the lower limbs around the thigh. For accurate reading of both pressures the following procedure should be followed. Having applied the armlet, the pulse should be palpated at the radial artery, and the pressure in the armlet then raised until the pulse can no longer be felt, at which moment the pressure in the manometer should be noted. The cuff should then be slowly decompressed and the pressure noted at which the pulse reappears. These two readings of systolic pressure should be close together, but they will not usually agree exactly for reasons which will be explained immediately. 'They give us the palpatofy systolic index, as it is called. The pressure is now lowered about 15 mm. Hg, and a stethoscope is placed in front of the bend of the elbow over the artery and as close up to :the cuff as possible. With each heartbeat a distinct sound like a pistol shot will be heard. The decompression is now continued slowly, and as the pressure falls the sounds will be heard to become louder and prob- BLOOD PRESSURE 129 ably somewhat murnmrish in quality. At a certain pressure this loud character of the sound will suddenly become much less marked, and the murmurish quality if present will suddenly disappear. This point cor- responds to the diastolic pressure, which is now read off from the manometer. It must be remembered that below this point, as the pressure in the cuff is further lowered, a sound is still heard in the artery; indeed it does not entirely disappear until the pressure has become quite low. This point of final disappearance is, however, of no significance. The cuff is the Fig. 26. — Apparatus for measuring the arterial blood pressure in man. The pressure in thi :uff is raised by means of the syringe until the pulse can no longer be felt at the wrist. Thi pressure is read off on the mercury manometer (systolic pressure). now entirely decompressed, and should be left so for a moment or more, so that the circulation in the part of the arm below it may return to the normal. The above readings should then be controlled by a second observa- tion, in which the methods employed are slightly modified. With the stethoscope at the bend of the elbow the pressure in the cuff is run up to a little above the previously determined diastolic pressure, so that the sound is clearly heard. The pressure is then further raised till the sound disappears. This point indicates the systolic pressure; it is called 130 THE CIRCULATION OF THE BLOOD the auditory systolic index. It will be found to give a systolic pressure a little higher than that obtained by palpation of the artery at the wrist. The sound being now absent, the pressure in the cuff is lowered until the sound reappears, and the point at which this occurs should almost exactly correspond to that at which the sound was found to disappear. If the palpatory systolic index is not below the auditory, it indicates that some error has been made in the application of the apparatus, and that the reading of the diastolic pressure will be unreliable. The usual source of error is in the position of the stethoscope. If readjustment of this does not bring the two indices into proper relationship, the auscul- tatory method can not be relied upon for either systolic or diastolic readings. In case of the failure of the auscultatory method, we have to fall back upon the palpatory method for systolic pressure; and for the measure- ment of diastolic, we have to use the method known as the oscillator]/, which until recent years was the only one used for gauging the dias- tolic pressure. This consists in observing the oscillation of the indicator of the pressure gauge; as the pressure in the cuff falls gradually from below the systolic pressure, these oscillations will be observed to increase in amplitude, until they reach a maximum beyond which with lower pressure they rapidly decline. The pressure in the cuff at the moment when the oscillations are at the maximum represents the diastolic pres- sure. With a mercury instrument it is obviously difficult to employ this method, but with a modern spring instrument it can with a little practice be used with great accuracy and will serve as a valuable check on the diastolic reading as taken by the auscultatory method. The procedure may be altered in various ways, there being only one pre- caution to bear in mind ; namely, that the pressure in the cuff should not be applied continuously for more than a few moments of time, for if this is done for long periods, not only will it interfere with the accuracy of the reading, but it may cause considerable discomfort to the patient. There are several conditions affecting the accuracy of the readings by each method Avhich it is Avell to bear in mind. These have been investi- gated by MacWilliam,1 Leonard Hill,2 and Erlanger.3 With regard to the systolic pressure the most important of these are as follows: (1) The compression cuff should be a wide one (12 cm.), and it should never be applied so that there is any chance of its compressing the artery against a bony surface. This precaution is necessary, since it has been found that much less pressure is required to obliterate any perceptible pulse below the armlet when the artery is flattened against some hard structure than when it is uniformly compressed in the tissues in which it lies. (2) Dis- crepancies are often noted between the systolic readings on compres- BLOOD PRESSURE 131 sion and decompression of the artery ; that is, the pulse may reappear on decompression at a lower pressure than that at which it disappeared on compression, the difference being most marked when the decompression is done quickly. This difference is owing to the fact that the full force of the pulse does not reach the forearm until all the vessels have become distended with blood. (3) There are often discrepancies in the systolic readings taken from different limbs; thus, it is not uncommon to find that the systolic pressure in the leg is higher than that in the arm even when the observed person is in the horizontal position. These differences are most commonly observed in patients suffering from aortic regurgi- tation or thickened arteries. In aortic regurgitation the pulse is of the water-hammer variety, and the greater systolic pressure observed in the leg vessels in such cases seems to depend on differences in the phys- ical conditions concerned in the transmission of this exaggerated pulse wave to the vessels of the two extremities. The. reason for the discrepancies in cases of hardened arteries is no doubt that the hardening is likely to be more pronounced in the ves- sels of the thigh than in those of the arms. When a hardened vessel is compressed it does not collapse uniformly- — that is, it does not become completely closed — but its walls come together at the middle part while chinks still -remain at the sides. The blood continues to pass through these chinks, and a very considerably higher pressure in the cuff is re- quired to obliterate them. That this is probably the correct explanation is supported by the observation that, although in such patients the pulse does not disappear in the vessels of -the foot at the same pressure as it does at the wrist, a distinct change is nevertheless perceptible in the pulse of the foot at a cuff pressure equal to that producing obliteration in the wrist. In a patient showing a systolic pressure of 115 mm. for the upper arm and 198 mm. for the leg, at 116 mm. the pulse in the leg, although not obliterated, became notably cut down in volume. It there- after persisted at a small volume with little alteration until the pressure became sufficient to obliterate it. It is said that repeated compression and decompression of the hardened arteries greatly reduces the dis- crepancy in the systolic readings. Differences in systolic readings are also sometimes observed in normal individuals, particularly after mus- cular exercise, but for these no satisfactory explanation can be given. While palpating the radial artery, it will often be noticed, as the pressure in the cuff is gradually raised from zero, that the force of the pulse increases perceptibly until a pressure of about 50 mm. is reached. This paradoxical behavior of the pulse can also be demonstrated by the sphygmograph (sec page 201). Its cause is not understood, but it is of significance that the greatest augmentations occur at a cuff pressure 132 THE CIRCULATION OF THE BLOOD at which a sound first conies to be heard by listening over the artery at the elbow. "With regard to the diastolic pressure, there has been some controversy as to whether it is more accurately gauged by the oscillatory or the aus- cultatory method. If both methods are employed it will usually be found that the oscillatory gives a higher reading than the auscultatory. The concensus of opinion seems to be that the latter method is the more accu- rate, and certainly it is the easier to apply, for with the oscillatory there is often great difficulty in deciding just exactly when the maximum oscillation occurs. The strongest evidence supporting the conclusion that the auscultatory readings are more -reliable than the oscillatory has been gained by ex- periments with an artificial schema, consisting of a wide glass tube rep- resenting the armlet, filled with Ringer's solution and closed by rubber stoppers pierced by tubes, which are connected with a fresh artery, which therefore runs from end to end inside the tube. Through tubing connected with the artery a pulsatile flow of oxygenated Ringer's solution is made to flow at varying pressures, which are indicated by valved manometers (see page 152) connected with the artery tubing just beyond the com- pression tube. The pressure in the latter is also measured by a manom- eter, and it is caused to vary by. a suitable compressor. By comparing the behavior of the artery with the pulsating movement of a spring manometer connected with the compression chamber, under different degrees of pressure inside and outside the artery, it has been observed that the maximal oscillation occurs when the artery is actually some- what flattened between the pulse beats; that is, it occurs at an outside pressure above the diastolic pressure, at which of course the vessel should retain its circular shape. When a stethoscope is applied to the tube leading from the artery just beyond the compression chamber, in the above de- scribed model sounds similar to those in the arm are heard with each pul- sation. While the pressure is being gradually lowered from above the ob- literation point, these sounds will be found to become first audible as soon as a certain amount of fluid is forced through the compressed area at each pulse (the systolic index), and to become louder and often murmurish in quality as the decompression is proceeded with, until a pressure is reached at which they suddenly become less intense and change in character. At this moment it will be observed by watching the artery that the external pressure is no longer capable of producing any flattening of the vessel between pulses. Evidently, therefore, the change of sound corresponds exactly to the diastolic pressure. With regard to the cause of the sound, it should be clearly understood that it is the systolic wave that produces it, but its occurrence and its BLOOD PRESSURE 133 character when present are dependent upon the intra-arterial pressure existing during the diastolic phase. The cause of the sound has been shown to depend on the production of a water-hammer in the blood ves- sels below the compression cuff (Erlanger3). By a water-hammer is meant the pressure changes which are caused by suddenly stopping the flow of water in a pipe. When a sudden pressure occurs in vessels with elastic walls, these walls are thrown into vibration and so produce a sound. In the taking of blood-pressure measurements, as above de- scribed, when the pressure in the cuff is between systolic and diastolic, the volume of the compressed artery will increase abruptly with each heartbeat and thus permit a considerable volume of swift-flowing blood to enter the rest of the artery underneath the cuff. /When this quickly moving column of blood comes into contact with the stationary blood filling the uncompressed artery ~below the cuff, it will become immedi- ately checked, and thus distend the arterial wall with unusual violence and set it into vibration. CHAPTER XVI THE FACTORS CONCERNED IN MAINTAINING THE BLOOD PRESSURE Having become familiar with the principles of the methods by which blood-pressure measurements are made, the next problem is to examine into the causes which operate to maintain the pressure. Two of these causes may be considered as fundamental, since without them no such pressure could exist. These are: (1) the pumping action of the heart, and (2) the peripheral resistance — that is, the resistance to outflow of blood from the ends of the arterial system. Less essential though im- portant factors are: (3) the volume of blood in the blood vessels, (4) the viscidity or viscosity of the blood, and (5) the elasticity of the walls of the vessels. We shall now proceed to examine the experimental evidence which indicates the relative importance of each of these factors. 1. The Pumping' Action of the Heart Changes produced in the mean arterial blood pressure by alteration in the pumping action of the heart are most strikingly demonstrated by observing this pressure after cutting or during stimulation of the vagus nerves. As will be explained later (page 217), impulses- conveyed through these nerves to the heart make the beats slower and weaker. These impulses are constantly acting in the heart, so that when both vagus nerves are cut, the beats become more frequent and stronger, with the result that the mean arterial pressure rises considerably. A lesser degree of this effect can usually be obtained by cutting the vagus nerve on one side (Fig. 27). If now the peripheral end of a cut vagus nerve is stimulated, as by applying an electric current to it, the heart will either stop beating altogether or become very much slowed, with the result that the mean arterial blood pressure will fall, in the former case almost to zero and in the latter, to a level Corresponding to the degree of slowing of the heart (Fig. 28). 2. The Peripheral Resistance To demonstrate the influence of peripheral resistance on mean arte- rial blood pressure, the most striking experiment is performed by cut- ting or stimulating the great splanchnic nerve. In this nerve impulses, 134 BLOOD PRESSURE 135 which are called vasoconstrictor because they constrict the lumen of the blood vessels, are transmitted to the blood vessels in the abdomen. The vessels are under the constant influence of these impulses so that, when the nerves that transmit them are sevei-ed, the vessels dilate and thus offer less resistance to the movement of blood along them. The result produced on the mean arterial blood pressure by cutting the two splanchnic nerves is therefore a marked and sudden fall, which is im- mediately recovered from if the peripheral end of one of the cut nerves is stimulated artificially (Fig. 29). In choosing this experiment to prove the relationship between peripheral resistance .and the. mean arterial blood Fig. 27. — Effect of cutting the vagus nerve on the arterial blood pressure. pressure, it must be remembered that it is not entirely conclusive, since the results observed on the mean arterial blood pressure from cutting or stimulating the nerve may be in part explained as due to variation in the total capacity of the circulation ; more room is created by cutting the nerves, less room by stimulating them. 3. The Amount of Blood in the Body This can be altered by hemorrhage or transfusion, and the results of such procedures are of interest not only on account of their physi- ologic bearing, but also because of their great practical importance. 136 THE CIRCULATION OF THE BLOOD To appreciate the significance of the results, it is important to bear in mind that the total volume of the blood constitutes from 5 to 7 per cent of the weight of the animal. This fact has been determined partly by postmortem, and partly by antemortem measurements. In the post- mortem method, the total amount of blood is determined by collecting the blood while bleeding the animal to death and then washing out the vessels with saline solution until the escaping fluid is no longer tinged with red. The blood contained in the saline solution is estimated by colorimetric methods (see page 92), and is added to that directly col- TVmt C. OttS. Fig. 28. — Effect of stimulating the peripheral end of the right vagus on the arterial blood pressure. lected. In the antemortem method some substance that does not dif- fuse through vessel walls or become quickly destroyed is added to the blood. By determining the concentration of this substance in a speci- men of blood, the volume with which it has become mixed can readily be calculated. Acacia has recently been found suitable for this purpose (Meek), but the best known work (of Haldane) was done by causing the animal to inspire a known amount of carbon monoxide. This combines with the hemoglobin of the blood (see page 401) to displace an equal quantity of oxygen. By determining the difference between the volume BLOOD PRESSURE 137 of carbon monoxide in the blood before and following its administration we can calculate with how much blood the known inspired quantity of carbon monoxide must have combined. The results vary somewhat in different animals; in the dog, the blood constitutes about 7.7 per cent of the body weight, and in man, about 5 per cent. The immediate effect of hemorrhage on the blood pressure depends on the rate of bleeding. If a large artery, such as the femoral, is cut across, Fig. 29.— -Effect of stimulation of the left splanchnic nerve on the arterial blood pressure. Note the primary and secondary rises. the pressure will show an immediate but moderate fall, due largely to the fact that we have suddenly decreased the" peripheral resistance. If on the other hand only a small artery or a vein is opened, the bleeding will at first produce no effect on the blood pressure, and it is only after some considerable amount of blood has been removed that it begins to fall (Fig. 30). To be more exact, we may state that the removal of 5 c.c. of blood per kilogram of body weight does not influence the blood pressure. The re- moval of a second portion of 5 c.c. per kilogram causes the blood pres- sure to begin to fall, the fall of pressure for each subsequent 5 c.c. of 138 THE CIRCULATION OF THE BLOOD blood per kilogram removed averaging about 6 mm. Hg, until after 20 to 25 e.c. of blood per kilogram have been removed, when a more rapid fall in pressure sets in (Downs'4). "When the pressure reaches the level of from 20 to 30 mm. Hg, the danger limit is reached, for there now supervenes a train of symptoms known as ' ' shock, ' ' and the chances for the animal's recovery become uncertain. That the removal of the first por- tion of blood, if this removal is slow enough, does not influence the blood pressure, indicates that some adjustment has occurred in the vascular system to hold up the pressure in spite of the loss of blood. This adjust- ment is believed to consist in vasoconstriction. Starr* |n Endof ifowRerrwBKemov/al Time in Sees 8t Abscissa Fig. 30. — The effect of rapid and slow hemorrhage on the arterial blood pressure. Between the second and third pieces of tracing an interval of two minutes elapsed. Recovery from hemorrhage is remarkably rapid, the original volume of blood being restored within a few hours. The chances of recovery de- pend upon the amount of blood lost. A loss equal to 2 or 3 per cent of the body weight can almost always be recovered from in laboratory ani- mals, and in the case of man there is reason to believe that recovery may occur after as much as 3 per cent of the body weight has been lost. The recovery of blood pressure is brought about partly by a transfer of fluid from the tissues to the blood. This abstraction causes a drying out of the tissues, which soon excites an extreme degree of thirst. The dilution of blood by fluid derived from the tissues occurs very rapidly, as can be shown by comparison of the hemoglobin content, or the number of blood corpuscles, in samples of blood removed immediately before BLOOD PRESSURE 139 and immediately after a hemorrhage. The specific gravity of the post- hemorrhagic blood is also decidedly below normal, indicating that the diluting fluid contains a lower concentration of dissolved substances than the blood plasma. The dilution of the blood is indeed often so great that hemolysis occurs, the plasma being distinctly tinted red. Hemorrhage also slightly raises the hydrogen-ion concentration of the blood plasma, and diminishes the store of reserve alkali, so that the ad- dition of a certain amount of acid to the blood (e.g., carbon dioxide) causes a greater rise in the hydrogen-ion concentration. The deficiency in the blood elements produced by the dilution is recti- fied by the manufacture of new corpuscles in the bone marrow, etc., but this process in a liberally fed animal takes several days for accomplish- ment, and while it is going on microscopic examination o'f the blood will reveal the presence of immature corpuscles. Careful studies of blood regeneration following the removal on two successive days, of 25 per cent of the blood, by Whipple and Hooper, have shown that even in starving animals the total amount of hemo- globin (percentage of hemoglobin multiplied by the volume of blood) slowly recovers. Eecovery is greatly hastened by feeding with flesh or even with gelatin. Removal of the spleen or the establishment of a bili- ary fistula does not interfere with the recovery. Incidentally it will be advantageous to consider here the effects of transfusion! These are very different according to the nature of the fluid used for transfusion. Three transfusion fluids have been investigated: (1) blood itself, (2) physiologic saline solution (see page 95), and (3) physiologic saline solution containing viscid substances such as gelatin. The effects are also very different according to whether the solutions are injected into animals with normal blood pressure or into those whose blood pressure has been lowered by preceding hemorrhage. When blood is injected into animals with normal blood pressure, it will very soon cause the pressure to rise, and as the injection is main- tained the rise may continue until the pressure is perhaps 50 per cent or more above its normal level. If the injection is long continued, how- ever, a sudden fall of pressure occurs, on account of engorgement of the right side of the heart. If the injection is not pushed so far, the increased blood pressure after being maintained for a short time returns to its old level. Injection of saline into a normal animal, if made slowly, has no effect at all on the blood pressure; if more rapidly injected, the pressure will rise slightly, but to a much less extent than that observed when blood itself is injected. Much larger quantities of the saline than of the blood can be tolerated before cardiac embarrassment ensues. After the dis- 140 THE CIRCULATION OF THE BLOOD continuance of the saline injection, the blood pressure returns very rapidly to its old level. The most striking result of such experiments is the enormous volume of saline solution which can be slowly injected without perceptibly affecting the pressure. The question is, Where does the fluid go ? If the urinary outflow is examined, a certain increase will usually be observed, but never by any means sufficient to account for the disappearance of the injected saline. If we open the abdominal cav- ity, we shall find that a considerable transudation of the saline into the peritoneal cavity has occurred, and that the liver is conspicuously edem- atous. A certain degree of edema is also usually evident in the tissues of the extremities. Still more interesting and important, from a practical standpoint, are the results obtained by injecting the above solutions into the animals whose blood pressure has been lowered by a previous hemorrhage. If the blood removed during the hemorrhage is defibrinated (see page 101), and then reinjected into the animal, it will bring the blood pressure al- most but not quite back to its original level, which will then be 'fairly well maintained. If, on the other hand, saline solution instead of blood is injected, the restoration of blood pressure (with an amount of saline equal to that of the removed blood) will amount to only about three- quarters of the extent to which it had fallen. This partial recovery is, moreover, maintained for only a short time, after which the pressure rapidly falls nearly to the level to which it was reduced by the hem- orrhage. These observations raise two important practical questions: (1) Why is saline relatively ineffective in the restoration of pressure? and (2) Why does. the pressure thus restored so quickly fall again? The answers to these questions brings us to a consideration of the next of the factors concerned in the maintenance of the blood pressure, namely, the viscosity of the blood. 4. The Viscosity, of the Blood The importance of this factor arises from the fact that facility of flow in a tube is inversely proportional to the viscosity of the fluid and directly proportional to the driving pressure to which it is subjected — that is, to the difference in pressure between two points in the tube. If therefore the output of the heart should remain constant, but the viscosity of the blood be decreased by a saline injection, the facility of flow would be increased and the pressure decreased. This fact can easily be shown experimentally in a model by causing gum solutions of various concentrations to be driven through a glass tube by means of a small piston pump delivering a constant amount of fluid into the tube with BLOOD PRESSURE 141 each movement* Although the outflow from the narrow end of the tube must remain constant, the pressure in the tubing will vary in proportion to the viscosity of the gum solution (Bayliss5.) Transferring these results to an animal whose blood pressure has been lowered by hemorrhage, it has been found that if saline solutions con- taining a sufficient amount of gum acacia or gelatin to make the viscos- ity about equal to that of blood, are injected, the original level of blood pressure is recovered as well as it would be had blood itself been in- jected. A 7 per cent solution of gum acacia almost fulfills these require- ments, but unfortunately this solution contains a slightly greater amount of calcium than it is safe to inject into an animal. The excess of calcium may, however, be removed by exactly neutralizing the gum solution with sodium hydroxide, neutral red being used as an indicator. Most of the calcium becomes precipitated as phosphate. The mucilage of the British Pharmacopeia, diluted five times with water, makes a 7 per cent solu- tion of gum acacia. A 6 per cent solution of gelatin, after being heated to 100° C., gives a viscosity similar to that of blood, but on account of the possible presence of tetanus spores such solutions must be very care- fully sterilized before injection, and the process of sterilization causes a decrease in viscosity. The injection of a quantity of one of the above solutions equal to that of blood lost by a hemorrhage will usually bring the blood pressure back to its original height and hold it there for an hour or so. Viscosity is, however, not. the only property of such solutions upon which their desirable effect depends. The osmotic pressure of the colloids also comes into play. By injecting saline solution containing a sufficient amount of a colloid such as soluble starch, which gives it the correct viscosity but has no osmotic pressure, the blood pressure, although it temporarily recovers after transfusion, does not maintain its recovery in the same way as with solutions containing gum or gelatin. The difference between a starch solution and one of gum or gelatin is that the former has no osmotic pressure, the effect of which is developed mainly on the excretion of urine, as can be shown by observing the outflow from the ureters during the injection into animals of equal quantities of saline alone or of saline containing starch or gelatin (Knowlton6.) With the first two fluids diuresis is produced, but not with gelatinous solutions. The reason that the osmotic pressure of certain colloids prevents passage of water from the blood into the urinif erous tubules is that the develop- ment of -this pressure on the blood side of the renal epithejium tends to counteract the filtration pressure by which the urine is formed (see page 514.) Although the urinary factor will not in itself explain the efficiency of 142 THE CIRCULATION OP THE BLOOD the colloids in recovering the blood pressure, the conditions controlling it reveal the mechanism by which the passage of fluid from the blood vessels into the tissues is prevented when solutions of correct composi- tion are injected. Normally the protein content of the blood plasma is higher than that of the tissue lymph, so that there is a continual attrac- tion of water from the tissues to the blood — an attraction Avhich is nor- mally balanced by filtration going in the opposite direction. When the filtration pressure in the blood vessels exceeds the difference existing between the osmotic pressure of their contents and that of the tissue fluids, water will pass into the tissue spaces. When the blood is diluted, as by the injection of saline solution, the osmotic pressure of the colloids in a given volume becomes lowered and, the filtration pressure remaining constant, fluid passes into the tissue spaces. Of course these explanations rest on the assumption that the wall of the blood vessels consists of a membrane which is permeable to crystalloids but impermeable or nearly so to colloids. Another important property of the transfused saline solution to con- sider is its hydrogen-ion concentration. This value increases in the blood left in the body after hemorrhage, and injection of sodium chloride solu- tion aggravates the acidosis; addition of NaHC03 so as to make a 0.2 M solution restores the correct PH, and at the same time restores the lost buffer influence (Milroy7.) These observations are of interest in the light of the recent discovery of Cannon that a condition of acidosis, as judged by the C02-combining power of the blood, is present in shock, and that the development of this condition can often be guarded against by bicarbonate injections. 5. Elasticity of Vessel Walls The elasticity of the vessel walls is essential to the maintenance of the diastolic pressure. If the walls presented no elasticity but were rigid, blood pressure would fall to zero between the heartbeats. This fact can very readily be shown by a simple physical model consisting of a pump to represent the heart, connected through a T-piece with two tubes, one of \vhich is elastic, the other rigid. The free end of each tube is con- tracted to a narrow aperture representing the peripheral resistance, and either tube may be shut off from the pump by means of a stopcock (see Fig. 30). Each tube should also be connected with a mercury manom- eter. If now the stopcocks are arranged so that the fluid passes into the rigid tube while the pump is in action, it will be found that with each stroke of the pump the pressure in the tube rises considerably, but that it falls to zero between the strokes. If now the stopcocks are turned so that the flow is through the elastic tube, the action of the pump being BLOOD PRESSURE 143 meanwhile kept up, it will be found that the pressure between the strokes is maintained at a height which is dependent on : ( 1 ) the rate at which the pump is operating, and (2) the resistance to outflow from the tube. The quicker the action of the pump and the higher the resistance, the lower the fall of pressure between the beats. The physical explanation of this result is clearly that the fluid within the elastic tube when the wave of pressure travels into it from the pump distends the Avails of the tube, so that when the pressure from the pump ceases to act, the stretched elastic Avails recoil on the column of fluid and maintain the pressure. We may say that the elastic fibers in the vessel Avails store up some of the systolic pressure and then transmit it to the blood during diastole. Fig. 31. — Diagram of experiment to show that the diastolic pressure depends on the elasticity of the vessel wall. The pulse (produced by compressing the bulb B) disappears when fluid flows through an elastic tube (F) when there is resistance (g) to the outflow. A, basin of water; B, bulb syringe; C and E, stopcocks; D, rigid tube; F, elastic tube; G, bulb filled with sponge. These considerations would lead us to expect that patients with hard- ened arteries should exhibit a lower diastolic pressure than normal per- sons, which, however, is not usually the case, since such patients also suffer from an increase in the resistance to the floAV of blood in the periph- ery. The pressure pulse in these patients is, hoAvever, very marked. On the other hand, Avhen the Aressel Avails become more extensible and elastic, as in certain cases of aneurism, the pressure pulse in the vessels beloAV the aneurism is distinctly less than that observed in normal ves- sels of the same patient. CHAPTER XVII THE ACTION OF THE HEART Having studied the methods for measurement and the main factors con- cerned in the maintenance of the arterial blood pressure, we may now pro- ceed to study in greater detail the two most important of these ; namely, the action of the heart, and the peripheral resistance. The heart action has to be studied from two viewpoints, the physical and the physiological. From the physical viewpoint we have to study the heart as the pump of the circulation. We must see how it acts so as to raise the pressure of the blood within it, and how the valves operate so as to direct the bloodflow always in one direction. We must also ex- plain the causes of certain secondary physical phenomena, such as the heart sounds which accompany the heart action, and of certain secondary changes in pressure produced in the other thoracic viscera by each heart- beat. From the physiologic Viewpoint we must investigate the conditions responsible for the constant rhythmic activity of the heart and the con- trol to which this is subjected through the nervous system. THE PUMPING ACTION OF THE HEART When the heart is viewed in the opened thorax of an animal kept alive by artificial respiration and lying in the prone position, it can be noted that with each contraction the ventricles become smaller and harder, that the apex tends to rise up a little, so 'that if the thorax were intact it would press more firmly against the walls, and that it rotates slightly from left to right, but does not move nearer the base of the heart. If the auriculoventricular groove is carefully observed, it will often be noted that it moves slightly toward the apex with each systole, whereas th'e base of the heart itself, where it is attached to the large vessels, re- mains fixed. The auricles can often be seen to contract and relax before the ventricles. The most noteworthy results of this inspection are that during sys- tole the apex of the heart does not move toward the base, but that the auriculoventricular groove moves slightly toward the apex. That these same movements occur in the intact animal can be shown by the very simple experiment of pushing two long steel knitting needles 144 THE ACTION OF THE HEART 145 through the thoracic walls into the heart walls, one of them so placed that it pierces the apex of the ventricle, the other so that it pierces the base. The needles then act as levers with their fulcra at the. chest wall, and if the movements of their outer free ends, produced by the movements of the heart, are observed, they will be found to confirm the observations made on the exposed heart. More particular investigations of the changes occurring in the shape of the heart cavity during systole and diastole have been undertaken by making measurements of sections across the heart in one or other of these conditions. For such purposes the heart in diastole is easily ob- tained, but for the heart in systole it is necessary to use the somewhat artificial means of injecting the heart with hot chromic acid solution just before the death of the animal. The chromic acid causes the cardiac muscle to contract and maintains it in this condition. The outcome of these investigations is, however, not of much practical importance. Although it is now common knowledge that the direction of the flow of the blood is from the veins to the arteries, yet it may be of interest to consider for a moment the general principle of the methods by which William Harvey succeeded in making this discovery. His evidence was partly anatomic, partly experimental. He pointed out that the walls of the veins, arid of the auricles to which they lead, are very thin, whereas those of the arteries and ventricles are very thick, and he concluded that in the veins the blood must flow gently from the tissues toward the heart, to which the valves in the veins direct it, and that in the arteries it must be propelled by pulses with each systole through the arteries towards the tissues by the contraction of the walls of the ventricles. The experimental support for this hypothesis he furnished partly by clamping the large vessels, veins and arteries leading to or from the heart, and observing the resulting distension or collapse of the vessel ; and partly by calculation of the amount of blood which must be expelled from the ventricles in a given period of time. Harvey's discoveries concerning the events of the cardiac cycle were not much added to until experimental methods were devised by which the pressure changes occurring in the various cavities could be measured and compared. Until such measurements were elaborated, it was impos- sible to investigate the mechanism by which the various valves between the heart cavities and the vessels connected with them perform their function, or to describe with any degree of accuracy the events occurring in the heart chambers during the various phases of the cardiac cycle. It is for the purpose of ascertaining the exact time relationship of these changes that intracardiac pressure curves are studied. 146 THE CIRCULATION OF THE BLOOD Intracardiac Pressure Curves The earliest method for taking such curves consisted in introducing into the cardiac chambers and the blood vessels of the horse, so-called cardiac sounds. These consisted of a more or less rigid tube furnished at one end with a little elastic bag or ampulla and connected at the other with a tambour, by means of rubber tubing. One of these little bags was placed in one of the ventricles, another in the auricle or aorta, the tube being inserted in the former case through one of the large veins at the root of the neck ; in the latter case through the carotid artery. The intracardiac pressure curves obtained in this way marked a great ad- vance over the methods that had previously been used to study the events of the cardiac cycle, but they were so faulty in comparison with tracings C-- Fig. 32. — Diagram of Wiggers' optical manometer. The wide glass tube (A) (connected with the ventricle, etc.) is connected with a brass cylinder (B) provided with a stopcock (C), the lumen of which comes in apposition with a plate (a) having a small opening in it. The freedom of communication between B and a is regulated by the position of the tap. Above a is a segment capsule (fc) 3 mm. in diameter and covered by rubber dam. This carries a small mirror (C) fastened so that it pivots on the chord side of the capsule. Above the capsule is arranged an inclined mirror, from which a strong beam of light is reflected on to the mirror (c) on the capsule. This beam then travels back and the mirror (£) is adjusted so that it impinges on a moving photographic plate. The slightest movements of the small mirror (C) are thus greatly magnified. taken by more modern methods that it is not worth while considering them any further here. The physical errors involved in the use of the older instruments were due mainly to the elastic recoil of the membranes, etc., used in their construction. A great improvement in technic was afforded by the use of the spring manometer of Hiirthle (see page 126), which was connected with one of the heart cavities by a cannula filled before insertion with some anticoagulant fluid. The cavity of the tambour was made as small as possible, and either left empty or filled with the anticoagulating fluid. THE ACTION OF THE HEART 147 A searching investigation into the physical principles involved in tak- ing records of sudden changes in pressure by such instruments has, how- ever, shown that considerable errors are incurred, the inertia of the moving mass of fluid in the tubing and the necessity of using levers in order to secure records being responsible for most of them (cf. Wig- gers). Their elimination has recently been achieved by using a so-called optical manometer, one of which (Wiggers') is shown in the accom- panying figure. It consists of a wide glass tube A, connected above with a hollow brass cylinder B, provided with a stopcock C, the lumen of which tapers from below upward till it assumes the same diameter as an aper- ture in the segment capsule 6, above it — that is, a capsule cut away at one end — which is 3 mm. in diameter and covered with rubber dam. By ad- justment of this stopcock the pulsations of the fluid in A and B can be damped to a greater or less extent before they are transmitted into the Fig. 33. — Optical records of intraventricuiar pressure; a-l, auricular systole; b-d, presphygmic period; d-f, sphygmic period; after /, diastole. Instruments of varying degrees of sensitiveness were employed in taking the curves. ('From Wiggers.) segment capsule. A small piece of celluloid carrying a tiny mirror rests on the rubber dam, being pivoted on the chord side of the capsule. A mirror is attached to the capsule with its plane so adjusted that the image of a strong light placed at some distance from it is focused on the little mirror carried by the celluloid. The ray reflected from the little mirror and again reflected from the larger mirror is adjusted so as to impinge upon a moving photographic plane travelling at a uniform rate in a suitably constructed photographic apparatus. By the use of such an apparatus the chief errors encountered by the use of the older in- struments are eliminated, because there is no moving mass of fluid and there are no levers to set up spurious vibrations. Curves secured by the use of this instrument are shown in Fig. 33. Two objects must be kept in view in analyzing the curves: (1) Curves obtained from the different cavities may be compared in order to de- termine the exact moment during the cardiac cycle at which such pres- •148 THE CIRCULATION OF THE BLOOD sure changes occur as must serve to produce opening or closing- of the various valves; and (2) the contour of the curves obtained from each cavity may be examined in order to find out exactly how the pressure in that particular cavity is behaving. Comparison of the Curves Before using the curves for ascertaining the relative pressure in the different cavities, they must be graduated according to some scale, for it is clear that by the use of instruments like those we have been describ- ing, the absolute pressure value of each curve will vary according to the construction of the instrument (thickness of membrane, etc.), and in- deed instruments of varying degrees of resistance must be employed in taking curves from places having such different pressures as exist in the auricles and ventricles. The graduation is, however, a very easy matter, and consists, as already explained (page 126), in connecting the instrument by means of a T-piece with a mercury manometer and a pres- sure bottle and then marking on the tracing, the points corresponding to each 10, 20 or 50 millimeters of increase of pressure, as the case may be. To ascertain the time relationship between the opening and the closing of the auriculoventricular valve, the tracings should be taken from the right auricle and the right ventricle, and to ascertain the same with re- gard to the semilunar valve, from the left ventricle and the aorta.* By comparing the curves it is now an easy matter to ascertain the exact moment at which the pressure in the one cavity comes to equal that in the other. This moment, read on the accompanying time tracing, will obviously indicate that at which the particular valve is just about to open or close. From the results of such experiments, the curves may be superimposed as in Fig. 34. , In the first place let us compare the curves from the right auricle and ventricle. The curves begin at the very end of diastole, and they show that a distinct increase in pressure is occurring in both auricle and ven- tricle and lasting about 0.05 second. This is of course caused by auric- ular systole, and since it occurs in both cavities, it indicates that the passage between them, the auriculoventricular orifice, must be open. The ventricular curve then suddenly shoots away beyond the auricular because of the onset of systole in the ventricle, and the point at which the two curves begin to separate indicates the moment at which the auriculoventricular valves close. From this time on until ventricular systole has given place to diastole, (about 0.2 second), the auricle is *The connections with the heart may be made by pushing long cannulae down the large veins or arteries, or in the case of the ventricles by inserting a cannula with a sharp point directly through the wall of the ventricle. THE ACTION OF THE HEART 149 therefore shut off from the ventricle. The exact moment in diastole at which the two cavities are again brought into communication — i.e., the ventricular valves open — is indicated by the curves coming together. Having thus determined the exact moments of opening and closing of the auriculoventricular valve, we may now proceed to compare the intraventricular pressure curve with that taken from the aorta. After the necessary calibration corrections, this curve has been placed in Fig. 34 in its true relationship to the ventricular curve. Beginning again at the end of diastole, we find that the aortic pressure is very considerably above that of the ventricles, indicating that the semilunar valves must be closed; and it will be observed that the intraventricular pressure at Fig. 34. — Pressure curves after being graduated have been superimposed. The presphygmic, sphygmic and postsphygmic periods of ventricular systole are shown by the vertical lines. The A-V valves close at the first line. The aortic valves open at the second line and close again at the third line. The A-V valves open at the fourth line. The position of the two main heart sounds is also indicated. the beginning of systole does not rise sufficiently to open them until an appreciable interval (0.02 to 0.04 second) after the closure of the auric- uloventricular valves; that is to say, there is a period at the beginning of ventricular systole during which the ventricle is a closed cavity. It is a period during which the ventricle by its contraction is getting up a sufficient amount of pressure in the fluid contained in it to force open the semilunar valves against the resistance of the pressure in the aorta, and it has been popularly called ''the period of getting up steam," or, in physiologic language, the isometric, or the presphygmic, period. We shall use the last-mentioned term in our further discussion here. 150 . THE CIRCULATION OF THE BLOOD After the aortic valves have been opened, it will be observed that the pressure in the ventricles is just a little above that in the aorta, and that it continues so during the whole of ventricular systole. When diastole sets in, the pressure in the ventricles quickly falls, and a point is soon reached at which equality of pressure in ventricle and aorta is again attained. This corresponds to the moment of the closure of the semi- lunar valves. The pressure in the ventricle, although now rapidly fall- ing, takes a little time before it has fallen low enough to permit the auricular valves to open. Here again, then, the ventricle is a closed cavity, and we have what is known as the postspliygmic period. CHAPTER XVIII THE PUMPING ACTION OF THE HEART (Cont'd) THE CONTOUR OF THE INTRACARDIAC CURVES The Ventricular Curve From an analysis of the contour of each curve, further interesting points are brought to light. The ventricular curve in the diagram alluded .to above (Fig. 34) is shown as having a flat top or plateau. By the use of the more modern, optically recording, instruments it has been shown that this plateau becomes displaced by a peak if every precaution is taken to prevent dulling down of the pressure changes in the instrument, as by opening wide the stopcock in the instrument (Fig. 33). The peak is, however, by no means a sharp one, so that we may fitly describe the contour of the ventricular curve during the sphygmic period as consist- ing of a rising portion, almost continuous with the curve during the pre- sphygmic period, a summit and then a declining portion, which is usually slower than the ascending. The practical value arising from a study of the curves lies in the insight which they give us into the nature of the stroke of the cardiac pump. They show us that the impulse which the ventricle gives to the moving mass of blood in the aorta is rather a sudden than a sustained one. The column of blood in the aorta is a mighty thing to move, and it would appear as if a sustained pressure brought to bear on it during the sphygmic period would be far more efficient in bringing about an adequate movement of the blood than a sudden jerk. In closing a heavy gate a slow sustained pressure is far more effective than a short push. It is further of interest to note on the intraventricular pressure curye that there is very little indication of any secondary waves or vibrations at the moment during which the semilunar valves are opened or closed. Nevertheless, by close scrutiny it can usually be, seen that a slight change in the direction of the ascending curve is evident when the valves open (see Fig. 33), and similarly that the moment of closing is indicated by a sharper bend in the curve. As a matter of fact, Wiggers has shown that the exact contour of the curve during the sphygmic period depends partly on the degree of sensitiveness of the optical manometer used and partly on the tension existing in the ventricle just before contraction. 151 152 THE CIRCULATION OP THE BLOOD In the case of the right ventricle the contour of the curve also depends on the degree* of resistance to the bloodflow through the pulmonary circuit. The top of the curve becomes broader when the initial tension is high, and more rounded when there is a high pulmonary resistance. Another point of interest in connection with the ventricular curve is that early in diastole it descends below the line of zero pressure, indicating that a negative or suction pressure must exist in the ventricle at this time. It will be further observed, however, that this subatmospheric pressure exists for only a very short time. The auriculoventricular valves being opened, a similar negative pressure is also present in the auricular tracing. Were we to depend on such records alone for evidence of the actual existence of this negative pressure in the heart, objection might be taken to the conclusion on the ground that it was due to the to manometer max valve to heart Fig. 35. — Von Frank's maximal and minimal valve, which is placed in the course of the tube between heart and mercury manometer. By turning the stopcocks, it may be used as a maximum, minimum, or ordinary manometer (central tubes open). (From Starling.) sudden recoil to which the instrument is subjected at the beginning of diastole. It is necessary therefore to control these observations by the use of an entirely different method. This consists in connecting the heart with a valved mercury manometer (see Fig. 35). This instru- ment does not of course record any sudden changes of pressure in the cardiac cavity, but in obedience to changes in pressure the mercury slowly moves in the direction in which the valve permits it to move. Such an instrument, with the valve opening towards the heart, is called a minimal manometer, and after it has been connected with the ventricle, it will te found that a negative pressure of perhaps 40 or 60 mm. Hg is recorded. Evidently, then, the negative pressure does actually exist in the ventricle during some phase of the cycle, and the question arises as to whether it is of importance in connection with the pumping action of the heart. At first sight, considering the heart as an elastic structure, we might con- THE PUMPING ACTION OF THE HEART 153 ceive that the negative pressure would serve to suck blood into the heart, just as it sucks water in an ordinary ball syringe. Closer consideration will, however, show that this conclusion is untenable, partly because the negative pressure exists in the ventricle for so short a period of time, and partly because it would have to operate on the slowly moving column of blood in the thin-walled veins, with the result that it would cause the walls of these vessels to come together rather than produce a movement of the blood contained in them. The negative pressure of the heart can not therefore be of much consequence in attracting the venous blood into the ventricle. Several factors may cooperate to produce this negative pressure, among them being the sudden opening out of the base of the ventricles at the beginning of diastole, the elastic recoil of the tissue which becomes compressed in the heart walls during systole, the turgescence of the walls of the ventricles produced by the sudden inrush of blood into the coro- nary vessels at the beginning of diastole, all of which factors tend to1 cause an opening out of the walls of the ventricles with a consequent increase in the capacity of their cavities. The Auricular Curve Examination of the intraauricular pressure curve is of particular in- terest because of the relationship which it has to a tracing taken of the movements in the jugular vein at the root of the neck (see page 274). This jugular pulse curve, as it is called, is produced mainly by the changes of pressure occurring in the auricle, from which it differs only in the relative height of the various waves. By graduating the. intra- auricular pressure curve by the method described above, we can tell exactly the magnitude in the changes of pressure occurring during each cardiac cycle. This obviously can not be done with a tracing taken from the jugular vein, although qualitatively the tracings reflect exactly the changes that are occurring in the auricle. On examining the auricular pressure curve (consult Figs. 34 and 97), we find that after the wave of presystole, which of course coincides exactly with that on the intraventricular curve, a second wave occurs culminating in a peak almost exactly at the beginning of the sphygmic period. The curve then rapidly descends, usually indeed below the line of zero pres- sure, and slowly rises throughout the rest of ventricular systole, until the moment of opening of the auriculoventricular valve, when it descends again and thereafter runs parallel with the ventricular curve. The let- ters used to designate the waves are the same as those employed for similar waves shown on the jugular pulse tracing, and although the 154 THE CIRCULATION OF THE BLOOD lettering is more or less arbitrary, we must accept it because of its gen- eral usage in all Avork of this kind. As to the causes of the waves A is of course caused by auricular systole or presystole; C, occurring as it does at the beginning of the period of ventricular systole, is caused by the bulging into the auricle of the closed auriculoventricular valve. The floor of the auricle, in other words, at this moment becomes somewhat elevated and imparts to the blood which is resting upon it a slight wave of pressure, which is transmitted along the veins for a considerable distance. The succeeding depression is marked x, and the negative pressure which it indicates is- probably due to the co-operation of three forces, all tending to increase the auricular capacity: (1) the diastole of the walls of the auricle; (2) the descent of the auriculoventricular groove, thus tending to open out somewhat the folds in the walls of the auricle; and (3), probably most important of all, the tendency of the thin-walled auricles to become dilated as a result of the sudden diminution in intrathoracic pressure produced at each heart- beat by the discharge of blood from the heart and intrathoracic blood vessels into those of the rest of the body. All thin-walled structures in the thoracic cavity, the auricles included, will expand to take up the extra room created in the thoracic cavity. Similar negative heart pulses, as they are called, can be observed with each systole in the lungs and in the esophagus. THE MECHANISM OF OPENING AND CLOSING OF THE VALVES When physical valves open and close as a result of the changes in pres- sure on their two surfaces, a certain amount of fluid must succeed in passing the valve flaps before these become perfectly closed. But there is every reason to believe that such is not the case in the heart, the flaps of both the auriculoventricular and the s'emilunar valves being already completely closed before pressure conditions entailing a possible regur- gitation of blood through them become established. Auriculoventricular Valves During diastole the flaps of the auriculoventricular valves are hanging down into the ventricle and floating in a half-open position in the blood, which is meanwhile accumulating in the chamber. This position is de- pendent upon the operation of tAVo opposing forces on the valve flaps: the pressure of the blood floAving from the auricle on their upper aspects, and reflected waves of pressure from the Avails of the ventricle on their under aspects (centripetal reflux). When presystole occurs, the pres- sure of the auricular stream momentarily increases, thus slightly dis- tending the Avail of the meanAvhile relaxed ventricle and after a moment 's THE PUMPING ACTION OF THE HEART 155 delay causing the reflected wave to become more pronounced. At the same time the muscular fibers in the valve flaps (Kurschner's fibers) contract and make the flaps shorter, the total effect of the two factors being that the valve takes up a position nearer that of closure. When presystole suddenly stops, the reflexed waves will persist for an instant of time longer than the auricular wave which causes them, because of the elastic nature of the ventricular wall, so that the valve flaps close with perfect opposition not merely at their edges but also for a con- siderable distance along their upper surfaces. When ventricular systole starts, the only effect of the high pressure which is brought suddenly to bear on the under surfaces of the already closed valves is to cause them to vibrate and to bulge into the auricles, being meanwhile anchored down and prevented from flapping into the auricle by the chordae tendineae. Although there is reason to believe that the musculi papillares to which these are attached begin to contract at the 1 23 Fig. 36. — Diagram to show the positions of the cardiac valves: 1, during diastole; 2, during the presphygmic period; 3, during the sphygmic period. very outset of ventricular systole — indeed slightly to precede it (see page 263), and thus keep the chordae taut, yet as systole continues the contraction of these muscles becomes more and more pronounced, and the resulting tightening of the chordae serves to draw down the valve flaps, so that progressively larger proportions of their upper aspects tend to become opposed. Meanwhile the auriculoventricular orifice is also be- coming narrowed down on account of the contraction of the musculature of the auriculoventricular groove. Semilunar Valves The mechanism involved in the operation of the semilunar valves is somewhat different. It has been shown that, when fluid is flowing in a tube, the pressure and velocity are not equal in the axial and peripheral parts of the stream. In the axis the velocity is greater than in the layers of fluid next to the walls, but the pressure is less. These facts can be 156 THE CIRCULATION OF THE BLOOD demonstrated by observing the flow through a wide tube of water in which are suspended lycopodium spores. By placing in the tube small bent tubes so arranged that one open end lies near the periphery and the other near the center, it can be seen that the differences in pressure are such as to cause the fluid to flow from periphery to axis (centripetal eddies). If the bent tubes are used to study the conditions of flow in a tube which suddenly becomes wider, it will be found that Avhere the wide portion starts centripetal eddies are set up, which tend to carry the seeds into the axis of the' stream, where their velocity is greatly increased. Now these are the conditions obtaining at the beginning of the large arteries S.a.-D.v. D.a.-S.v. Fig. 37. — Diagram showing the position of the cardiac chambers and valves during presystole (S.a. — D.v.) and during the sphygmic period. (From Landois.) of the heart, the orifice into the ventricles being constricted, while at the sinus valsalvae the vessels are dilated. A centripetal vortex must be set up in the sinus, tending to throw the valve flaps into a closed posi- tion, which, however is prevented by the blood rushing between them from the ventricles. They thus take up a mid-position and vibrate in the stream. When the efflux from the ventricle stops at the end of sys- tole, the reflux, lasting for a moment longer and being now unopposed, immediately closes the valves, in which position they are then maintained by the greater pressure on their upper surfaces. The position of the valves relative to the events of the cardiac cycle is shown in Figs. 36 and 37. THE PUMPING ACTION OF THE HEART 157 THE HEAET SOUNDS During certain phases of the cycle distinct sounds, the heart sounds, can be heard by applying a stethoscope to the thoracic wall. The first occurs at the beginning of ventricular systole and is best heard over the apex beat; the second occurs at the beginning of diastole and is heard best at the second right costal cartilage or in the second left intercostal space. A third sound, much less distinct, is sometimes heard in diastole a short time after the second. To study the exact time relationship of the sounds the vibrations which they set up can be recorded graphically alongside cardiac tracings by means of a microphone attachment to the electrocardiograph (see page 259). Causes of Sounds It has been found that the first sound consists of two distinct elements, one high pitched and the other of a dull character. The former element is believed to be the result of vibrations set up in the flaps of the auric- uloventricular valves, and therefore in the blood in the heart, by the sudden rise in systolic pressure. The dull element on the other hand is undoubtedly of muscular origin. The evidence for these conclusions is as follows: (1) When the auriculoventricular valves are prevented from closing properly either by disease or by pushing a loop of wire down the large veins, the high pitched quality disappears, and nothing but a rush- ing sound accompanies the dull bruit produced by the contracting muscle. (2) In a heart that has been rendered bloodless by an incision near the apex, or even in an excised but still beating heart, the dull element of the first sound still continues to be heard for a short time. That con- tracting muscle produces a sound is a well-established fact. There are, however, many obscure phenomena connected with the causation of the first sound, but we can not go into such controversial matters here. A close inspection of the electrophonographic tracing shows that the sound starts at the beginning of the presphygmic period, and that it lasts with gradually declining variation in intensity until well into the sphygmic period (Fig. 38). The second sound occurs accurately at the beginning of diastole and can readily be shown to be caused by the sudden shutting and stretching of the semilunar valves, which throws them, the blood in contact with them, and the neighboring walls of the aorta into vibration. Proof of this conclusion is furnished by the following facts: The second sound immediately disappears if the blood is let out of the heart by opening the apex, and it is replaced by a rushing "bruit" if the flaps are pre- vented from closing as a result of disease or of hooking them back by 158 THE CIRCULATION OF THE BLOOD passing a wire down the carotid artery. The third sound, although audi- ble only in some individuals, can nevertheless be shown to exist by the electrophonograph, and since it occurs at the time when the auriculo- ventricular valves open, it is believed to depend upon the sudden inrush of blood from auricles to ventricles. The greatest importance of the sounds is in the clinical diagnosis of val- vular and other lesions of the heart. When a valve leaks, for example, the blood escapes past it under great pressure, and is ejected into a mass of blood at low pressure, these being conditions which are well known to create sounds or 'bruits. By examining the exact relationship of such bruits to the normal heart sounds, deductions can be drawn concerning the condition of the various valves. Record of Heart Sounds The heart sounds have been graphically recorded by transmitting them through a stethoscope to a microphone placed in circuit Avith a string galvanometer (electrophonograms,). Through this circuit passes a cur- rent the strength of which depends on the resistance offered by the microphone, this resistance being proportional to the number and ampli- tude of the vibrations of the sounds transmitted to it through the stetho- scope. There are several objections to this method. One of these is de- pendent on the varying distance of the heart from the chest wall, which causes many of the sound vibrations to be lost before they reach the stethoscope ;_ another, on adventitious sounds arising from contracting muscles, the impact of the heart against the chest wall, etc., and still another on unequal resonation by the air in the neighboring portions of lungs. To investigate the problem more thoroughly, Wiggers,37 using anesthetized animals, has recorded the sounds by carefully stitching to the heart (exposed through a small opening in the pericardium) a lever, the end of which was attached to a "transmitter" consisting of a small capsule covered with rubber dam. The transmitter was connected by rubber tubing to a ' ' recorder ' ' consisting of another small capsule carry- ing on its membrane (made of rubber cement) an eccentrically placed small mirror, on to which a beam of light Avas throAvn. The movements of the beam of light reflected from the mirror, and caused by the sound vibra- tions, were photographed. Mechanical vibrations set up in the apparatus itself Avere largely eliminated by a side opening on the recorder, and the effect of outside sounds minimized by surrounding the recorder by a ventilated glass housing. Although this apparatus is not free from faults due to inherent vibra- tion frequency and resonance, the records secured by it are valuable in shoAving the exact relationship of the sounds to the events of the cardiac THE PUMPING ACTION OP THE HEART 159 cycle. The vibrations from the two ventricles are alike, but differ from those taken from the aorta. The first ventricular sound consists of from five to thirteen irregular vibrations, usually in three groups, the first composed of two small vibrations, the middle one of several large vibra- tions, and the third of a varying number of small vibrations. The 3 r It II m pi •-.•>»<•. ; A. ^^ff ^| ' , ^^^K B. C. Fig. 38.— Electrophonograms along with intraventricular pressure curves from three dif- ferent experiments. In A the uppermost curve shows the pressure, the middle one the sounds o_f the right ventricle, and the lowermost one those of the aorta. P indicates the relative posi- tion of the curves. M is due to mechanical oscillations. Sz indicates the second sound, and /, 2, 3, and 4 the corrected time relations of the first sounds. In B, the pressure and sound curves are both from the left ventricle (letters same as in A). In C, the aortic and pulmonary arterial sounds are shown (letters same as in A). (From Wiggers and Dean.) duration of the sound is from 0.05 to 0.152 seconds, and the periodicity from 0.004 to 0.054 per second. When compared with an intraventricu- 160 THE CIRCULATION OF THE BLOOD lar pressure curve, the initial vibrations occur 0.01 second prior to the rise in pressure, the main vibrations reaching their greatest amplitude before the sphygmic period begins, and the final vibrations occurring during the early part of the sphygmic period and therefore just before the aortic pressure has reached its height. The main vibrations therefore occur during the descending limb of the R wave of the electrocardiogram (be- ginning 0.01 second before its completion), the small preliminary vibra- tions occurring during the ascending limb. When taken from the aorta, the record of the first sound is someAvhat different, there being no initial vibrations and the main ones being of greater frequency and reaching their maximum earlier than those taken from the ventricle. The sub- sequent vibrations are also larger, especially when the aortic pressure is high (Fig. 38). The record of the second sound at the ventricle is much simpler and usually of less amplitude than the first, consisting of two to six vibrations lasting 0.015 to 0.056 second. They begin a short time after the ventricu- lar pressure begins to fall, approximately at the dicrotic notch of the aortic curve, being completed in from 0.015 to 0.025 second after the bottom of the notch. Their relationship to the T wave is variable. Taken from the aorta, the record of the second sound shows vibrations of greater amplitude and of a greater frequency than that from the ventricle. CHAPTER XIX THE NUTRITION OF THE HEART THE BLOOD SUPPLY In cold-blooded animals, such as the frog, the heart muscle is nourished by blood soaking into it from the heart chambers, which indeed do not form definite cavities as in the mammalian heart, but exist as an inter- lacement of muscular tissue. In the hearts of higher animals, the muscu- lature is supplied by special arteries (the coronary), although a certain amount of blood may still pass directly from the cardiac cavities into the musculature through the veins of Thebesius. The relative importance of the various branches of the coronary artery in maintaining an adequate nutrition of the heart has been studied by observing the effect of occlusion of one or more of them (W. T. Porter9.) Occlusion of the circumflex branch of the left coronary artery caused arrest of the heartbeat in about 80 per cent of cases, the arrest being usually accompanied by fibrillary contraction. Occlusion of the right coronary arrested the ventricular contraction in about 20 per cent of the cases. Smaller branches may be occluded without any evident change in the heartbeat. These results indicate that the capillary areas supplied by the branches of the coronary artery do not freely anastomose with one another. They are more or less terminal arteries ; that is, each branch supplies a distinct region of the cardiac muscle. If one of the smaller branches of the coro- nary is occluded, although there is 110 immediate stoppage of the heart- beat, yet after some time the area supplied by that branch usually under- goes necrosis, again indicating that collateral circulation can not have become established. It is interesting, however, to note in this connection that anatomic studies have shown that a certain amount of anastomosis does occur between capillaries of different branches, although it is evi- dent, from the above observations, that no adequate collateral circulation becomes established through this anastomosis. PERFUSION OF HEART OUTSIDE THE BODY In order that the blood supply through the coronary arteries may adequately maintain the normal nutrition of the cardiac muscle, certain 161 162 THE. CIRCULATION OF THE BLOOD conditions must be fulfilled. The recognition of these conditions has been accomplished by observations on the excised heart, for it has been found that if they are fulfilled the mammalian heart can be made to beat in perfectly normal fashion for several hours after its removal from the animal's body. Indeed certain mammalian hearts, such as that of the rabbit, may be made to beat for several days outside the body. We may consider the essential conditions of the blood supply under four headings: (1) the temperature; (2) the oxygen supply; (3) the pressure; and (4) the chemical composition. Successful perfusion may be performed with artificial saline solutions (e. g., Locke's), but it is simplest in investigating the relative importance of the above conditions to start the heart per- fusion with defibrinated blood. After bleeding an anesthetized animal, such as a dog or a cat, until no more blood can be removed, the blood is defibrinated and filtered through gauze to remove the fibrin. The thorax of the dead animal is then quickly opened, ligatures placed around the main arteries springing from the arch of the aorta, a cannula with its end pointing toward the heart inserted into the descending thoracic aorta, and the latter cut across below the point of insertion of the cannula. The heart is then quickly removed from the thorax and an artificial saline solution (Locke's) allowed to run into the aortic cannula through a side tube, until all the blood has been washed out -from the coronary vessels. Dur- ing this operation the heart may develop a few beats even though the solution is quite cool. The aortic cannula is now connected with a bottle containing the defibrinated blood diluted with Locke's solution and brought to body temperature by immersion in a water-bath. By means of a suitably regulated air pressure exerted on the surface of the diluted blood in the bottle, this is forced through an outlet at the foot of the bottle into tubing which runs to the aortic cannula. The fluid thus finds its way into the coronary vessels; for in passing toward the heart in the aorta it will close the semilunar valves and force its way under pressure into the coronary vessels, subsequently escaping by the coronary sinus into the right auricle. Very soon after the perfusion is started the heart begins to beat vigorously and regularly, thus offering a suitable prepara- tion upon which to test the first three mentioned conditions necessary for the nutrition of the cardiac musculature (Fig. 39). If the temperature of the solution is allowed to fall considerably, the beat becomes much slower, and if the cooling is proceeded with, the heart will after a while cease beating altogether. If the pressure is lowered, the beat will not necessarily become slower but very much feebler, and will soon cease. In general it may be said that the temperature of the solution affects the rate of the beat, and the pressure affects its strength. THE NUTRITION OF THE HEART 163 Funnel (refilling & air vent) Stock solution (Diluted blood + a salt solution) • Metal pan Hot water bath "fiver Tube soldered between palls Fig. 39. — One form of apparatus for recording tracings from an excised heart (Langendorff method). The heart is kept warm by a water bath (heart warmer), and the perfusion fluid is also warmed. The driving pressure in this apparatus is supplied by gravity. (From Jackson.) 164 THE CIRCULATION OF THE BLOOD It is, however, obvious that in perfused preparations changes in pres- sure are likely to cause alterations in rate as well as in force, unless great care is taken to keep the heart itself as warm as the perfusion fluid. The importance of an adequate pressure in the coronary vessels has been clearly brought out in certain experiments in which the beat has been maintained for a short time by establishing a pressure in the cor- onary vessels by means of indifferent fluids or gases. Thus, if oxygen gas is alloAved to pass through the vessels under pressure, the heart will beat for a short time, and the same result has been observed even when mineral oil or mercury has been perfused under pressure (Sollmann). The necessity for an adequate oxygen supply is very readily demon- strated. If the darker blood ejected from the right auricle with each heartbeat is transferred immediately to the perfusion bottle, the heart- beat will soon become feeble and irregular, to be readily restored to normal when this dark blood is shaken up with air or oxygen. By artificial perfusion in the manner above described, the automatism of the heart may be restored many hours after death. Partial restora- tion, confined to the auricles or to that part of the ventricles lying im- mediately adjacent to the large blood vessels, can also be accomplished in the heart of man several days after death, provided death has not been caused by some acute toxic infection such as diphtheria or septice- mia. The Russian physiologist Kuliabko, has succeeded in restoring for over an hour the normal beat of the heart of a three-months-old boy twenty hours after death from double pneumonia, but here again the pulsation returns only in certain parts of the heart. As will be pointed out, the remarkable resistance of the heart muscle displayed in these experiments has been taken as an argument in favor of the myogenic hypothesis for automatic rhythmic power of cardiac muscle, the argu- ment being that nervous structures could not live so long a time after death. The fallacies in this argument are discussed elsewhere. RESUSCITATION OF THE HEART IN SITU A suitable intracoronary pressure is a sine qua non for the mainte- nance of the heartbeat, and this is a fact of great clinical significance, for it indicates that any attempts to resuscitate a dead animal are cer- tain of failure unless the method is such as will bring a nutrient fluid under a certain pressure to bear on the coronary arteries. Injection of fluid, even of defibrinated blood, into a vein will obviously fail to ful- fill this condition, for the perfusion must be made into an artery so that the fluid is carried down the aorta and thence into the coronary arteries. THE NUTRITION OF THE HEART 165 The practical question, in so far as resuscitation of the heartbeat is concerned, is therefore, How can we get the necessary fluid under pres- sure into the beginning of the aorta f Even if we were to transfuse fluid under considerable pressure into the aorta through the carotid artery, it would mainly follow the large vessels leading away from the heart, only a fraction of it reaching the beginning of the aorta. To compel the fluid to pass towards the heart AVC must introduce some obstruction to its passage peripherally. This can be done by the injection of a consid- erable dose of epinephrine (adrenaline) in normal saline solution through the needle of a hypodermic syringe inserted into the tubing leading from the burette or pressure bottle to the cannula in the carotid artery. As the perfusion fluid is running in, the epinephrine injection is quickly made, artificial respiration and cardiac massage being meanwhile prac- ticed. In the majority of animals it will be found that complete res- toration of the normal blood pressure can be effected by this method. Indeed by performing the resuscitation under aseptic conditions, some animals may be permanently resuscitated so far as the circulation is concerned, although the nervous structures, even after a few minutes of "death," never reacquire their normal condition. The epinephrine acts mainly by constricting the small arterioles and thus directing the bloodflow towards the heart, but partly also by a direct stimulating action on the cardiac muscle. It does not, however, con- tract the coronary vessels; on the contrary, it is said to cause these slightly to dilate. THE RELATIVE IMPORTANCE OF THE VARIOUS CONSTITUENTS OF THE PERFUSION FLUID We can study the chemical conditions necessary for resuscitation of the heartbeat by observing the beat of an artificially perfused heart while solutions of different chemical composition are being perfused through the coronary vessels. At the outset we are impressed with the fact that for successful resuscitation the organic constituents of the nutrient fluid are of trivial importance compared with the inorganic constituents. With a solution containing the proper proportion of in- organic salts, and of course an adequate supply of oxygen, the heart of a rabbit, for example, may be made to continue beating for several days.. It is true that it will beat longer if some of the organic con- stituents of the blood plasma, particularly carbohydrate, are present, but on the inorganic constituents alone its ability to beat is truly remarkable. 166 THE CIRCULATION OF THE BLOOD Observations on Cold-Blooded Heart The earlier experiments for the investigation of the chemical condi- tions necessary for the maintenance of the heartbeat were performed on the heart of the frog or turtle. By perfusing either of these hearts with physiologic sodium-chloride solution, it was observed that though the beat might continue for some time, yet it gradually grew feebler and feebler, until at last it ceased altogether with the heart muscle in a condition of extreme relaxation or diastole. If small proportions of potassium and calcium salts (as chloride) were added to the sodium- chloride solution, the beat was much better maintained. Doctor Sidney Ringer proved that the optimum concentration to produce efficient and prolonged contraction for the heart of the frog or terrapin is as follows: potassium chloride, 0.03 per cent; calcium chloride, 0.025 per cent. The effectiveness of the solution was also found to be increased by the addition of 0.003 per cent of sodium bicarbonate. This acts as a buf- fer substance (page 36), holding the hydrogen-ion concentration at a constant level. More recent work has shown that the hydrogen-ion con- centration of the perfusion solutions is of considerable importance in determining the efficiency of the beat, but the optimum is not the same for the hearts of different kinds of animal, and indeed it may differ for different parts of the same heart. The question naturally arises as to the relative importance of each of the above salts; or rather, we should say, cations, since the anion, chlorine, is the same for all of them. The function of the sodium chlo- ride in the solutions is twofold: (1) to endow the solution with the proper osmotic pressure (see page 4) ; and (2) to perform the special role of the sodium ion in the origination and maintenance of the auto- matic beat. The latter function of Na can be shown by observing the behav- ior of strips cut out from the ventricle of the turtle heart and placed in solutions of correct osmotic pressure but containing no sodium chlo- ride— isotonic solutions of cane sugar, for example. They soon cease to beat, but if a small amount of sodium chloride is added to the cane sugar solution, rhythmic contractions return. The role of the calcium ions is almost entirely a pharmacologic one. If a strip of turtle ven- tricle which has been made to cease beating by immersion in isotonic sugar solution is placed in a weak solution of calcium chloride before it is transferred to sodium chloride solution, the spontaneous contrac- tions will return earlier and continue for a longer time. On the other hand, if more than the correct amount of calcium salt is present in the solution, the beats will soon be found to become smaller and smaller in amplitude, because relaxation does not properly occur between them, and ultimately they will cease altogether with the ventricle in a condition THE NUTRITION OF THE HEART 167 of extreme contraction, called calcium rigor. The importance of calcium may also be shown by attempting to perfuse a turtle heart with blood serum from which the calcium has been removed by the addition of sodium oxalate (which precipitates it as insoluble calcium oxalate). The heart soon ceases to beat, but can readily be made to do so again by adding a slight excess of calcium chloride. The potassium ions do not appear, like those of calcium and sodium, to be absolutely essential for the maintenance of the heartbeat; at least the heart of the turtle will beat for a long time when perfused with a solu- tion containing only sodium and calcium salts. The explanation of this result need not, however, necessarily be that potassium is an unessential constituent of the perfusion fluid, for it may well depend on the fact that there is a sufficient store of potassium locked away in the muscle fiber to supply the requirements of the heart muscle for this ion for at least as long as the beat would continue under any circumstances. In any case, we know that potassium has a profound influence on the heart- beat, for when the proportion of it in the perfusion fluid is increased, the beat becomes very slow and the tone of the heart is greatly diminished — that is, it becomes extremely relaxed between the beats; and if the amount is further increased, will very soon come to a standstill in a greatly dilated or diastolic position. The striking antagonism displayed by these inorganic cations upon the heartbeat has led some investigators to suggest that the stimulus re- sponsible for the rhythmic activity of the heart depends on some sort of chemical union occurring between the inorganic cations and the con- tractile substance of the heart. Union of calcium with the contractile substance will lead to systole or contraction, whereas union of sodium or potassium will lead to relaxation or diastole. Observations on Mammalian Heart Investigation of the efficiency of various saline solutions on the iso- lated mammalian heart has shown that the proportion of the above salts must be somewhat different from that used for the cold-blooded heart. As might be expected, the most efficient proportions are those present in the blood serum of the particular animal whose heart is being per- fused. Basing his proportions upon the results of analyses of the inor- ganic constituents of mammalian blood serum, Locke found that an inorganic solution of the following composition is most efficient: so- dium chloride, 0.9 per cent; calcium chloride, 0.024 per cent; potassium chloride, 0.042 per cent; and sodium bicarbonate, 0.01 to 0.03 per cent. When "Locke's solution," as it is called, is perfused, with oxygen in it, under pressure through the isolated mammalian heart at body tempera- 168 THE CIRCULATION OP THE BLOOD ture, efficient beating can be maintained for many hours. More recently a solution known as Tyrode 's is commonly used. It contains a small amount of magnesium and of phosphates. Although undoubtedly superior for some perfused preparations, such as the intestine, it does not seem to be in any way superior to Locke's for the perfusion of the heart. The bicar- bonates and phosphates in these solutions endow them with a hydrogen-ion concentration near that of the blood (slightly on the alkaline side of neutrality), and at the same time they act as buffer substances. As already pointed out, the organic constituents of such perfusion fluids do not appear to be relatively of nearly so much importance as the inorganic. Nevertheless it appears that a small percentage (0.01 per cent) of glucose does materially improve the nutritive qualities of the solution, and it has moreover been shown that after a while the con- centration of glucose in the perfusion fluid distinctly decreases. This does not of itself necessarily mean that the glucose is actually utilized by the heart muscle: it might be stored away in it as glycogen. That some consumption of carbohydrate does however occur in the heart has been demonstrated by measuring the intake of oxygen and the output of carbon dioxide through the lungs of an isolated heart-lung prepara- tion perfused outside the body with defibrinated blood. By experiments of this type the attempt has been made to show that the heart of diabetic animals loses the power of burning glucose as compared Avith the hearts of normal animals. While the experiments are very suggestive, the results do not as yet justify us in claiming that in the latter disease the power of burning glucose in the tissues has been materially depressed. The concentration of hydrogen ions in the perfusion fluid has an im- portant influence on cardiac efficiency. We also know that the most convenient method for changing the hydrogen-ion concentration of such fluids is by altering their tension of carbon dioxide (see page 354). In a heart-lung preparation,* such alteration in carbon-dioxide tension can very readily be brought about by altering the percentage of this gas in the air with which the lungs are ventilated. To measure the efficiency of the heartbeat in such an experiment, it is convenient to enclose the organ in a cardioplethysmograph, the tracing of which will tell us the degree to which the heart is contracted or relaxed, as well as the output of blood per minute. By increasing the tension of carbon dioxide, it has been found in such experiments that the dilatation of the ventricle is encouraged, so that the heart with each beat discharges a larger quan- tity of blood (Fig. 40). When defibrinated blood is used the optimum *A heart-lung preparation is one in which both heart and lungs are perfused outside the body, the vessels being suitably connected to maintain a continuous circulation. THE NUTRITION OF THE HEART 169 pressure or tension of carbon dioxide has been found to lie between 5 and 10 per cent of an atmosphere. That the effect of carbon dioxide in encouraging the relaxation of the heart between beats is dependent upon the change in hydrogen-ion con- centration of the perfusion fluid has been shown by securing the same results in experiments with perfusion fluids to which different quanti- tities of weak nonvolatile acids have been added. These observations are Fig. 40. — Volume curve of ventricles of cat (lower curve) in a heart-lung perfusion prepara- of practical importance because of the light which they throw on the cause of cardiac failure following upon conditions in which there has been excessive removal of carbon dioxide from the blood, as in forced ventilation of the lungs. Yandell Henderson has suggested that sur- gical shock may be, partly at least, due to cardiac failure following the "washing out" of carbon dioxide from the blood by the dyspnea so often incident to the administration of anesthetics in surgical operations. CHAPTER XX THE PHYSIOLOGY OF THE HEARTBEAT THE ORIGIN AND PROPAGATION OF THE BEAT— THE PHYSIO- LOGIC CHARACTERISTICS OF CARDIAC MUSCLE The origin and propagation of the heartbeat are studied on the excised heart of a frog or turtle, or on the mammalian heart by perfusing it under suitable conditions, which have already been described. The results obtained on the cold-blooded heart apply more or less directly to the warm-blooded. In the first place it is clear that the rhythmic contrac- tility of the heart is not at all dependent upon the central nervous sys- tem, for if it were so, 'the excised heart could not continue beating. This fact does not, however, necessarily imply that the beating power is in- dependent of nervous structures, for in the heart itself an extended net- work of nerve cells and connecting nerve fibers can readily be demon- strated. It might quite well be the case that the rhythmic beat is de- pendent upon the transmission to the muscle fibers of the heart of impulses generated in the nerve cells and transmitted along the nerve fibers of this local nervous system. Such is the neurogenic hypothesis of the heartbeat. On the other hand, it may be that these nervous structures are not at all responsible for the origination of the beat, but serve merely as sta- tions on the pathway of the nerve impulses, transmitted to the heart from the central nervous system along the vagus and sympathetic nerves, for the purpose of altering the rate of the heartbeat so as to adjust it to the requirements of blood supply in the various parts of the body. In such a case the rhythmic power would reside in the muscular tissues of the heart — that is, each cardiac muscular cell would have the power, not merely like skeletal muscle of contracting in response to a stimulus transmitted to it, but also of originating that stimulus within itself. This is the myogenic hypothesis. Much controversy has raged around these two hypotheses and although space will not permit a detailed study of the question, it will be necessary, on account of the great importance of the subject from the physiologic standpoint, briefly to review the main arguments of each school of thought. There is no piece of evidence offered by the advocates of either the neurogenic or the myogenic hypothesis that can, taken singly, be con- 170 THE PHYSIOLOGY OF THE HEARTBEAT 171 sidered as absolutely conclusive. Although some of "the proofs" may at first sight appear to be conclusive, yet each of them breaks down when subjected to a closer scrutiny. It is only after we have collected all the evidence for and against each view that we shall be in a position to come to any conclusion, and even then it will be plain that our conclusion can be only tentative. Myogenic Hypothesis Taking first of all the evidence in support of the myogenic hypothesis, the following stands out most prominently: 1. The heart beats in the embryo chick before any nerve cells have grown into it, and not only this, but if portions of heart muscle are re- moved from the embryo and placed in blood plasma, they will continue beating for many days It has also been observed that cells may wander off from this mass of cardiac muscle and undergo multiplication and differentiation, so as to produce isolated muscle cells which exhibit rhythmic contractility. The rebuttal on the part of the neurogenists of this apparently unassailable evidence is to the effect that, although em- bryonic muscle cells may exhibit the power of rhythmic contraction, this does not mean that the fully developed muscle cells will necessarily have such power. In the eary stages of embryonic development, it is of course evident that the functions which in the fully developed animal are del- egated to various special organs and tissues should be performed by cells having several such functions in common. The muscle cells of the heart, for example, may to start with be possessed of a power not only of con- tracting but also of initiating the contraction. It may be that they are partly nervous in character and that only later, when the differentiation is consummated, does the power of rhythmic contraction become dele- gated to the nervous element and that of contraction retained by the muscle itself. 2. The nervous structure in the heart may be damaged either by me- chanical means or by drugs without apparently interfering with the power of rhythmic contraction; for example, in the heart of large tur- tles it is possible to dissect out a considerable amount of nervous tissue without any disturbance of the beat, and in all animals the administration of atropine, which paralyzes the postganglionic fibers of the autonomic nervous system (see page 226) found in the heart, does not affect it. 3. The apex of the ventricle in such hearts as that of the turtle can be shown, by careful histologic examination, to contain no nerve cells, and although a few nerve fibers may be found, these are functionless without nerve cells. This virtually nerveless piece of heart muscle can be made to contract rhythmically by perfusing it with suitable saline 172 THE CIRCULATION OF THE BLOOD solution under pressure and starting the beating by application of elec- trical stimuli. Isolated strips of ventricular muscle, in which also no nervous element can be demonstrated, may under favorable conditions be caused to beat quite regularly if supplied with proper nutrient fluid. The rebuttal of this evidence is twofold: In the first place, skeletal mus- cle itself under certain conditions, such as exposure to solutions con- taining an excess of phosphate (Biedermann's), may exhibit rhythmic contractility, especially on cooling, which indicates that exhibition of rhyth- mic power in isolated portions of cardiac muscle need not mean that under ordinary conditions such power is responsible for the normal heartbeat. In the second place, it is pointed out that although we can not reveal their presence by present-day histologic methods, this is not conclusive evidence that the heart-muscle fiber may not possess some nervous struc- tures capable of functioning as nerve cells. The heart even of mammals can be made to continue beating for sev- eral days after excision from the body. The nerve cells, as we know them in the central nervous system at least, can not, on the other hand, be made to functionate for more than a few hours after death. Therefore, it is argued, the heartbeat in surviving mammalian hearts can not de- pend on the nervous structures. The argument is hoAvever easily refuted: on the one hand, we do not know that the nerve structures situated peripherally in the heart muscles are of the same viable nature as those composing the central nervous system ; and, on the other, the survival of the heart may in itself be sufficient to maintain around the nerve cells embedded in it a nutrient environment which is much more physiologic than that which AVC can supply in artificial perfusions of surviving nervous tissues. 4. Circumstantial but nevertheless strong evidence is furnished by the fact that many other varieties of involuntary muscle are endowed with rhythmic contractility; thus, the muscle of the intestines, of the ureters, of the bladder, of the uterus, of the blood vessels of certain animals, and of the lymph vessels in the so-called lymph hearts, main- tain rhythmic contractility after isolation from the animal body. The rhythmic power seems in certain of these cases to be independent of nervous control. Neurogenic Hypothesis In favor of this hypothesis the following evidence is offered: 1. The heart of certain animals — of Limulus, the king-crab, for exam- ple, is definitely dependent for its rhythmic contractility upon neigh- boring nervous structures. The heart of this animal is a tubular sac- culated organ, and along its dorsal surface there runs longitudinally a THE PHYSIOLOGY OF THE HEARTBEAT 173 nerve cord containing ganglion cells and giving off fibers which proceed in part directly to the heart and in part to lateral cords (Fig. 41). Re- moval of this median nerve cord is followed by total abolition of the heartbeat; the heart becomes perfectly quiescent like an unstimulated skeletal muscle. In appraising the evidence at its true value, it must be noted that although by stimulation of the nerve fibers contraction of the heart can be produced, the contraction is like that of a skeletal mus- cle— it is not rhythmic ; and moreover — and this is most important — if the various physiologic properties of muscle as described below be stud- ied (page 176), it will be found that in all of them the quiescent heart muscle behaves, not like the heart muscle of other animals, but like that of skeletal muscle. This evidence, therefore, while indisputably showing that the heart of Limulus depends for its rhythmic .power upon neigh- boring nerve structures, does not justify the assumption that this will be the case in the heart of animals having different physiologic properties. 2. The disposition of the nervous structures in the heart, especially of the frog and turtle, exactly corresponds to the degree of development of Fig. 41. — Heart and cardiac nerves of Limulus polyphetmis. (Carlson.) aa, anterior ar- teries; la, lateral arteries; In, lateral nerves, mnc, median ganglionic chain; os, ostii or afferent stomata, each pair of which corresponds to one of the segments into which the Limulus heart is divided. the rhythmic power of the different parts of the heart ; thus, the greatest rhythmic power is manifested by the sinus and the least by the tip of the ventricle at the bulbus arteriosus. In the former position the nerve structures are very prominent; in the latter, no nerve cells and but few nerve fibers can be detected. This proof is, however, easily assailed. In the first place, it may merely be a coincidence that the disposition of the nerve structures and the development of rhythmic power correspond. The unequal rhythmic powers may depend primarily on a difference in structure of the muscle fibers themselves, such differences having been shown to exist between the muscle cells of the sinus and those of, say, the ventricle. The former cells, for example, have much less' developed crossed striation and their protoplasm is much more gran- ular; in short, they are much more embryonic in type than the cells from the tip of the ventricle. If a jury had to return a verdict from evidence of so conflicting a char, acter, it would no doubt be equivalent to that of the Scottish court — "not 174 THE CIRCULATION OF THE BLOOD proven." But it is likely that the majority of the jury would vote in favor of the myogenic hypothesis. Probably the safest viewpoint to take at the present time is that the power of rhythmic contraction is inherent in the cardiac muscle fibers, being most highly developed in those of the venous end of the heart, and least developed in those of the arterial end. Such a conclusion does not deny to the nervous struc- tures of the heart the power under certain conditions of also assuming rhythmic activity. In one case at least — namely, the heart of Limulus — we know that this is so. For some reason in this animal the cardiac muscle fiber has lost its inherent rhythmic power, and is now dependent for its activities upon rhythmic nervous discharges transmitted to it from the neighboring nerve cords, a condition which is paralleled in the higher animals, in the innervation of the respiratory muscles. The respiratory center rhythmically discharges impulses to the muscles, which are quiescent in the absence of these impulses. In a volume of this nature, devoted primarily to the practical appli- cation of physiology, the discussion of these problems may seem a little out of place, but that this is not the case is seen when we consider that the experiments upon which the various points of evidence depend bring to light facts of the very greatest importance in the study of the physiology of the heartbeat. One fact which stands out prominently is that the greatest rhythmic power resides in the basal portion of the heart — that is, in what corresponds, in the more primitive hearts, to the sinus venosus. Although the muscle of the entire heart possesses rhythmic power, it does not do so to an equal degree; in the sinus the rhythmic power is extraordinarily developed, while in the bulbus arteriosus it is scarcely recognizable. This observation suggests the possibility that the sinus may dominate the heartbeat — that it may be the "pacemaker" for the heart as a whole. The most natural method for demonstrating such a possibility would be to observe the effect on the heartbeat of some Hock between the sinus and the rest of the heart. Such a block can be intro- duced in the heart of cold-blooded animals by local compression around the various junctions. If a thread is tied around the sinoauricular junction, the sinus will go on beating uninterruptedly, but the auricles and ventricles — that is, the greater part of the heart below the ligatures — will cease beating, sometimes entirely (Stannius' ligature). After a while, however, the heart below the ligature will usually begin to beat, but at a rhythm which is slower than, and independent of, that of the sinus. THE PHYSIOLOGY OF THE HEARTBEAT 175 The experiment can be still better performed by using a wedge- shaped clamp. (GaskelPs clamp.) If this is applied so that the heart can be pinched either at the sinoauricular junction or at the auriculo- ventricular, it will be found that, as the cardiac tissue is gradually Fig. 42. — Heart-block produced by applying clamp at a-v junction. The clamp was tightened at a. (From Brubaker.) pinched, the portion of the heart below fails to beat as quickly as that above the clamp (Fig. 42). This is known as partial heart-block, and the degree of the block is indicated by the numerical expression 2 to 1, 3 to 1, 4 to 1, etc., meaning that the sinus is beating either twice as quickly as the ventricle, or three times, or four times as the case may Fig. 43. — Tracing of contraction of ventricle, showing the effect of the local application of heat to the auricle at z, and to the apex of the ventricle at z. Note that the rate in- creased in the former case. be. Similar conditions of heart-block may also be produced by cutting the cardiac tissue partly across at various places in the heart. Further evidence that the sinus dominates the beat in the heart of 176 THE CIRCULATION OF THE BLOOD cold-blooded animals is furnished by observing the effects of local heat- ing or cooling of the various parts of the heart. In all rhythmically acting structures it is well-known that heat increases the rate of the rhythm and cold depresses it. If we locally warm the region of the sinus, as by holding a heated wire near it the whole heart will immedi- ately beat quicker; but if we locally heat the tip of the ventricle, no alteration of rhythm will be observed to occur (Fig. 43). The establishment of the fact that the sinus dominates the heartbea* —that it is the pacemaker of the beat — raises the question as to how the impulse originated at this place is transmitted over the rest of the heart, and here again a neurogenic and a myogenic hypothesis have to be considered. Before going into this question, however, it will be well for us to consider briefly the manner of response of cardiac muscle fiber to a stimulus, because the behavior of cardiac muscle under such conditions is considerably different in many regards from that of skel- etal muscle, and it is to these differences that many of the peculiar alterations in the beat observed after interfering with the conducting structures between the sinus and the rest of the heart, are to be ex- plained. The Physiologic Characteristics of Cardiac Muscle It is necessary to bring the heart into a quiescent state in order to investigate the properties of its musculature. This is accomplished by the application of the Stannius ligature betAveen the sinus and the auri- 1....L- Fig. 44. — Frog heart showing the position of the first and second ligatures of Stannius- (Iledon): /, auricles; 2, sinus; j, ventricle. It is the first ligature which brings the heart to standstill. cles (Fig. 44). After tightening the ligature the auricles and ventricles become quiescent, and by observing the effects produced by the appli- cation of electric or other stimuli we can compare the behavior of the cardiac muscle with that of skeletal muscle similarly stimulated. This comparison is made because of the assistance which it offers in compre- hending the properties of cardiac muscle. As a matter of fact, recent investigations have shown that the differences between the two types of muscle are not fundamental, since under certain conditions the one may THE PHYSIOLOGY OF THE HEARTBEAT 177 be made to behave like the other. They are dependent upon the pres- ence or absence of anastomosis between the muscle fibers. 1. When electric stimuli of varying strengths are applied to skeletal muscle, the contraction produced by each stimulus is proportional to the strength of the latter until this has become of such a strength that the maximal response is elicited. In cardiac muscle, on the other hand, an entirely different result is obtained, for the weakest stimulus, if it produces any response at all, produces one that is maximal; that is, the height of contraction is the same as it would have been had a much stronger stimulus been applied. Expressing this result in general terms, we may say that in cardiac muscle a minimal stimulus produces a maxi- A. — Skeletal Muscle B. — Cardiac Muscle Fig. 45. — Effects of stimuli of increasing strength on skeletal and cardiac muscle to illustrate the "all or nothing" principle in the latter. (From Practical Physiology.) mal effect, whereas in skeletal, the effect, as measured by the height of contraction, is proportional to the intensity of stimulation. This is some- times known as the "all or nothing phenomenon" (Fig. 45). 2. If maximal stimuli are applied successively and at short intervals of time to skeletal muscle, a slightly higher response results from each succeeding stimulus, until about ten stimuli have been applied, after which for some considerable time the same height of contraction follows each stimulus. If each contraction is recorded, it will be seen that the first few contractions give a staircase effect; that is, if a horizontal line is drawn from the top of each contraction to the next one, the effect of a 178 THE CIRCULATION OP THE BLOOD staircase with gradually diminishing steps will be produced. If we repeat this observation with cardiac muscle, we shall find that the staircase phenomenon or treppe, as it is called, is very pronounced ; and moreover, in obedience to the all or nothing principle, the treppe is obtained in cardiac muscle whatever may be the relative strengths of the stimuli applied to the heart, provided always that all of them are effective; whereas in the case of skeletal muscle it can be demonstrated only pro- vided the stimuli are of equal strength (Fig. 46). 3. If an effective stimulus is applied to a skeletal muscle while in process Skeletal muscle Cardiac muscle Fig. 46. — -The effects of successive stimuli on skeletal and cardiac muscle to show the prominence of the staircase phenomenon, or treppe, in the latter. (From T. G. Brodie.) of contraction, as in response to a preceding stimulus, the second stimulus prolongs the contraction produced by the first one. If, however, the second stimulus is applied during the latent period* of the first one, it will have no effect — that is, the muscle during this period is refractory.! From these results it follows that stimuli succeeding each other during the contraction period will, in the case of skeletal muscle, cause a continuous contraction, or tetanus, as it is called, because the contraction produced by each stimu- lus will add itself to that of its predecessor before any trace of relax- ation has set in. If, however, the second stimulus is applied so late in the contraction period of the first that time is not available for the latent *By "latent period" is meant the period after the moment of application of a stimulus during which no effect of that stimulus is observed. fBy "refractory period" is meant the time following the application of a stimulus during which a second stimulus develops less than its full effect or no effect at all. THE PHYSIOLOGY OP THE HEARTBEAT 179 period to expend itself, then obviously a slight relaxation will have oc- curred before the effect of the second stimulus develops itself, and tet- anus will be incomplete. These facts will be evident from the accom- panying tracings (Fig. 47). Skeletal muscle Stannius' heart Fig. 47. — The effects of successive stimuli and of tetanizing stimuli on skeletal muscle and cardiac muscle. The small vertical marks show when the stimuli were introduced. (Compiled from tracings published by T. G. Brodie and Leonard Hill.) In the case of cardiac muscle the above described properties are quite different, for the refractory phase extends throughout the whole period of contraction; that is, a second stimulus applied during the contraction produced by a previous stimulus has no effect whatsoever; it does not 180 THE CIRCULATION OF THE BLOOD have one until the muscle has reached the full extent of its contraction and is about to relax. Since a latent period must supervene upon the application of this second stimulus, it follows that no complete fusion of the contractions is possible. Complete tetanus therefore, does not occur in cardiac muscle, however frequently the stimuli may be applied (Fig. 47). The refractory phase is a property of extreme importance in under- standing many of the peculiar irregularities observed in cardiac action. If we observe the effect of stimuli applied at varying periods after the Fig. 48. — Myograms of frog's ventricle, showing effect of excitation by break induction shocks at various moments of the cardiac cycle. The line O indicates the commencement of all the beats during which the shock is sent in. It will be noted that in /, 2 and J, the heart is refractory to the stimulus. The signals indicate the moments at which the stimuli were ap- plied. From 4 to 8 the heart reacts by an extrasystole, after a delay, which is progressively less the later in diastole the stimulus enters, as shown by the sections shaded obliquely to make them more conspicuous. The extrasystolcs increase in . height from 4 to 5, each being followed by a compensatory pause. (From r.uciani's Human Physiology.) termination of the refractory phase of a previous stimulus, we shall find that the height of the extra contraction is directly proportional to the time after the end of the refractory period at which it is applied. . If a stimulus is applied at the very beginning of diastole, the extra contrac- tion will be small, whereas if it is applied at the end of diastole, the extra contraction will be at least as high as that of the preceding. It may be higher because of the treppe. THE PHYSIOLOGY OF THE HEARTBEAT 181 These observations enable us to interpret the results obtained by ap- plying electric shocks (extra stimuli) to the beating heart during different phases of systole and diastole. During systole, the muscle being refrac- tory, no effect is produced by the extra stimulus, but during diastole •extra systoles which are progressively more pronounced the later in diastole they occur, follow the application of each stimulus. These re- sults are so far exactly like those obtained with a quiescent heart. But another phenomenon now becomes evident; namely, that following each extra systole there is a compensatory pause in the action of the heart, of such duration that, when the next natural beat occurs, it does so practically at the same time as it would have occurred had no artificial stimulus been applied. This Avill be apparent from the following dia- gram (Pig. 48). It should be noted that the refractory period is greatly diminished by raising the temperature of the heart. Indeed, under these conditions and with strong stimulation it may be possible to produce an almost complete tetanus. The importance of knowing the above facts is that we are thereby enabled to explain the peculiar manner in which the ventricle responds to stimuli transmitted to it from the sinus and the auricle. The muscu- lature of the auricle and ventricle of the mammalian heart is not one continuous sheet, but is separated by a space at the auriculoventricular junction, across which, in specially organized structures, the beat of the auricle is transmitted to the ventricle. Sometimes the stimuli are so frequent that the ventricular muscle is unable to respond to each stimu- lus transmitted to it, with the result that marked irregularities in con- traction occur (see page 280). In this way certain of the cardiac irregu- larities observed in man can be explained. Thus, the so-called pulsus bigeminus is due to every second beat being an extra systole. This second beat is therefore generally weaker than the preceding and succeeding nor- mal beats, and it is almost always followed by a compensatory pause. When the intervals separating the beats are of uniform length, although every second beat is diminished in size, the pulse is termed pulsus alternans. CHAPTER XXI THE PHYSIOLOGY OF THE HEARTBEAT (Cont'd) THE ORIGIN AND PROPAGATION OF THE BEAT IN THE MAMMALIAN HEART As has been shown in the preceding chapter, there is no doubt that in the cold-blooded heart the beat originates at the sinus venosus, whence it spreads to the rest of the heart. Very strong evidence has also been presented to indicate that the beating power is inherent in the muscle fiber itself and independent of nervous structure. This would suggest the further possibility that the structures through which the beat is propa- gated are the muscle fibers and not the nerve fibers — in other words, that the propagation of the heartbeat, like its origination, is myogenic rather than neurogenic. Direct proof of this hypothesis is readily fur- nished by numerous experiments, among which may be mentioned mak- ing interdigitating cuts across the heart, or excising a ribbon of ven- tricular muscle by an incision simulating the walls of Troy. In both these cases the beat will be found to travel from one end of the muscular band to the other, although it is evident that all the nerves proceeding from base to apex of the heart must have been severed. Of course this evidence is not irrefutable, for it might be argued that there are nerv- ous structures disposed in the form of a plexus continuously all over the heart, and that some branches of the plexus remain uncut in the above experiments. It is only in the heart of Limulus that undoubted evidence exists that the beat is transmitted by nerves, but as we have seen, this heart in all its properties is probably the proverbial exception which proves the rule. The balance of evidence stands in favor of the view that the propagation of the beat over the cold-blooded heart is myogenic and not neurogenic. CONDUCTING TISSUE IN MAMMALIAN HEART When we attempt to investigate the problems of the origin and propa- gation of the beat in the warm-blooded heart, many experimental diffi- culties of course face us. In overcoming these, the first thing we must do is to establish the structural relationship between cold-blooded and warm-blooded hearts. In the embryo of both classes of animals the 132 THE PHYSIOLOGY OF THE HEARTBEAT 183 heart arises as the so-called cardiac tube. As development, proceeds, diverticula grow out from the- walls of this tube to form the auricles and ventricles. In the comparatively simple heart of the turtle these dispo- sitions of the auricles and ventricles in relationship to the cardiac tube are more or less evident even in the fully developed heart, particularly in the case of the auricles (Fig. 49); but in the heart of the higher mammalia it is impossible by superficial examination alone to show any remains of the primitive cardiac tube. More careful anatomic investiga- tions during recent years have, however, shown that it exists in the form of certain definite structures composed of tissue histologically quite dif- ferent from that of the rest of the heart, and disposed in such a manner TH Fig. 49. — Heart of tortoise as suspended. B, body of tortoise; TH, threads to levers; CL, clamp holding aorta; A, auricle; C, coronary nerve; S, sinus; V, ventricle. (From Gaskell.) as would indicate not only that it is derived from the primitive cardiac tube, but also that it is the main pathway along which the beat is transmitted. This primitive cardiac tissue is much better developed in certain re- gions than in others, the first portion of it to be discovered being that known as the auriculoventricular node, or the node of Stanley Kent* (Figs. 50 and 51). This structure is found at the base of the interauricular sep- tum on the right side and near its posterior margin. It exists as a collection of peculiar small primitive cells and fibers, and is continued downward as a bundle of the same peculiar tissue to the interventricular septum, where, near the union of the posterior and median flaps of the aortic *The discovery of this node is often erroneously attributed to His, and called after his name. 184 THE CIRCULATION OF THE BLOOD valve, it bifurcates so as to send a branch down each side of the septum immediately below the endocardium. Each main branch, as it proceeds downward on the septum, divides up into an intricate system of smaller branches, which become reflected over the inner surface of the ventricles, where their existence has been known for some time as the so-called Fig. SO. — Dissection of heart to show auriculoventricular bundle (Keith); 3, the beginning of the bundle, known as the A-V node; 2, the bundle dividing into two branches; 4, the branch run- ning on the right side of the interventricular septum. (From Howcll's Physiology.) Fig. 51. — Photograph of model of the auriculoventricular bundle and its ramifications, con- structed from dissections of the heart (Miss De Witt). All of the branches in the left ventricle are not included. (From Howell.) Purkinje fibers. The fibers ultimately end in close association with the papillary- muscles. The node and main bundle and the two branches before they have begun to divide are surrounded by fibrous tissue, and they seem to have a liberal blood supply. It is of interest that they con- tain a high percentage of glycogen. In the human heart the auriculo- THE PHYSIOLOGY OF THE HEARTBEAT 185 ventricular node and bundle measure about 15 mm. in length and about 2 mm. in width. The rest of the tissue between the auricles and ventricles is fibrous in nature, although other connections like those of the auriculoventricular bundle have been described by Kent. One of these, called, the right lat- eral connection, runs between the right auricle and the external wall of the right ventricle. Another, but much smaller, mass of similar embryonic cardiac tissue has more recently been discovered by Keith and Flack in the parts of the auricle which correspond anatomically to the sinus venosus of the heart of cold-blooded animals — that is, in the area lying between the openings of the venae cavse and around the coronary sinus. To be more explicit, this tissue lies "in the sulcus terminalis just below the fork formed by the junction of the upper surface of the auricular appendix with the superior vena cava." This sinoauricular node, as it is called, is more or less club-shaped, the blunt end of the club being above, as shown in the accompanying figure (Fig. 52). It is important to note that there is no direct connection visible between the sinoauricular and auriculoventricular nodes (Fig. 53). Another anatomic fact seen also in the accompanying figure, concerns the disposition of the muscular fibers of the auricle. These radiate in bundles in a peculiar fan-shaped manner from a point which lies im- mediately below the sinoauricular node to all parts of the superficies of the right auricle. This point has been called the concentration point. At the termination of the vense cavae, the muscle fibers are arranged more or less circularly. Having become familiar with the disposition in the mammalian heart of the primitive cardiac tissue, along Avhich in the heart of the lower animals we know that the heartbeat spreads, we may now proceed to examine the evidence shoAving that this tissue is also responsible for the origination and propagation of the beat in the heart of mammals. With regard to the origin of the beat in a normally beating mammalian heart, it is of course impossible to tell where this takes place. If the heart is excised, however, it will continue to beat 'for a few moments, and as it dies it will be observed that the power of contraction remains in the au- ricular region, and particularly at the bases of the venae cavae, for a con- siderable time after the ventricles have ceased to beat. This part — the ultimum moriens — is situated in most hearts somewhat lower than the sinoauricular node. That it is the last part of the heart to cease con- tracting does not necessarily mean that it is the part of the heart in which the beat ordinarily originates; it means simply that this is the part of the auricle in which the power of contraction remains for the 186 THE CIRCULATION OF THE BLOOD longest time after death. Although the observation does not enable us to determine exactly where the heartbeat originates, yet it makes it very probable that this is somewhere in the auricles ; a conclusion which is borne out by many other pieces of evidence, such as those obtained by Fig. 52. — Diagram of an auricle showing the arrangement of the muscle bands; the concen- tration point (C.P.); and the outline of the S.A. node (S.A.N.). The diagram is to scale, and illustrates by the circles and connecting dotted lines the method of leading off by paired contacts and the subsequent orientation. (From Thomas Lewis.) i Auricular appendage ,--S/'noaurtcu/ar node -Auriculoventrlcular node -Aurluiloventricular bundle HR- • Right & left ventricular bundles -Musculi papillares Fig. 53. — Diagram to show the general ramifications of the conducting tissue in the heart of the mammal. It will be observed that there is none of this tissue between the sinoauriculo- and auriculoventricular nodes. the study of polysphygmograms (page 273), of electrocardiograms (page 266), and of observations on the heart during heart-block (page 270). Our problem therefore narrows itself down to determining the exact point of the right auricle at which the beat originates. THE PHYSIOLOGY OF THE HEARTBEAT 187 SITE OF ORIGIN OF THE BEAT The working hypothesis from which we may proceed to attack this problem is that the beat originates in the sinoauricular node, and to put this to the test, various methods have been employed: (1) Warming or cooling or injuring the node and noting the effect on the heartbeat. Such procedures greatly affect the rate of the heartbeat, whereas they produce no change when applied to other parts of the heart. (2) De- termination of the comparative rhythmic power of strips cut out from different regions of the auricular walls. It is greatest in those taken from the region of the node. (3) Determination by the use of galvan- ometric curves of the relation of the node to the seat of origin of cardiac impulse. By all these methods the results indicate clearly that the beat originates in the sinoauricular node, but on account of the great im- portance in connection with the interpretation of electrocardiograms in man, it is particularly with the result of the third group of experiments that we will concern ourselves here. Evidence Furnished by Studying the Current of Action Which Accompanies the Heartbeat To start with, it is essential that we make ourselves familiar with the principles of the methods employed. These principles are briefly as follows: When a wave of contraction passes along a muscle, it is im- mediately preceded by a change in electrical potential, which can be detected by means of a galvanometer connected with the muscle through so-called nonpolarizable electrodes. The galvanometer employed must be extremely sensitive, and must not vibrate after the current has ceased to pass. The form generally in use today is known as the string galva- nometer of Einthoven. It differs from the galvanometer ordinarily em- ployed in physical laboratories in that the current instead of passing through a coil of wire surrounding a magnetic needle, passes through a silverized quartz thread suspended in the strong magnetic field which exists between the two opposing poles of a horseshoe electromagnet. The string is thus surrounded on all sides by innumerable lines of force extending between the two poles of the magnet. When a current, how- ever small, passes along the string, it will generate lines of force of its own, and these by reacting with the stationary lines of force of the field will cause the string to move. The string is placed in the pathway of a strong beam of light, and its shadow, after being magnified by lenses, is projected on a moving photographic plate or paper arranged in a suitable holder. The nonpolarizable electrodes referred to are employed in place of ordinary electrodes in order to obviate the generation of elec- 188 THE CIRCULATION OF THE BLOOD trie currents set up by the contact of metal with the saline constituents of the muscle juices. If Ave connect a galvanometer by means of nonpolarizable electrodes with, two parts of a denervated muscle (the curarized sartorius-of the frog), it will be found that a current is set up whenever a wave of con- traction passes over the muscle from one end to the other. The part which first contracts becomes electrically negative to the rest of the muscle, but as the wave of contraction passes along it, the "negativity" de- creases at the end at which the wave started until, when the wave has reached the middle of the strip, neither end of the muscle shows any difference in potential, so that the string comes back to a position of rest. However, as the contraction wave reaches the farther end of the muscle, this lead in turn becomes negative, and the string swings in the Fig. 54. — Diagram to illustrate the development and spread of the wave of negativity in a strip of muscle (curarized sartorius) when stimulated at the end (P). The shaded portions show the position of the negativity. The portion of the curve drawn by the deflections of the galvanom- eter at each stage are shown at the right (a, b,'c, and d) . (After Lewis.) opposite direction (Fig. 54). From this comparatively simple experiment it can be seen that a muscular contraction wave arises at the electrode which is negative first, and that the movement of the string of the galvanometer is most marked — that is, the deflection is greatest — when the two electrodes are applied at the extreme ends of the muscle. When they are brought closer together, the initial deflection becomes much l«ss marked ; in other words, the amplitude of the negative wave is greatest when the time interval between the receipt of the excitation at the two contacts is greatest. The application of these facts to the study of the initiation of the beat in the auricle requires that we should consider another proposition: namely, if a pair of contacts are arranged in the center of a circular sheet of muscle and the edge of this sheet is stimulated at different THE PHYSIOLOGY OP THE HEARTBEAT 189 points, the amplitude of deflection of a galvanometer connected with the pair of contacts will be most pronounced when these are radial to the points of stimulation, for under these conditions it is evident that the greatest possible difference will exist betAveen the intervals required for the wave to reach each contact. Bearing these principles in mind, we may now proceed to examine the evidence pointing to the origin of the heartbeat at the sinoauricular node: (1) When two electrodes are applied at different points of the au- ricle, the amplitude of movement of the string of the galvanometer produced by each heartbeat is greatest when the line joining the elec- trodes converges on the sinoauricular node. To make this clear the movement of the string must be photographed in the manner above described, the resulting tracing being called an electrocardiogram. From the experiments with the circular sheet of muscle alluded to it is evident that the stimulus to produce this result must have arisen in the neigh- borhood of the node. (2) If one electrode is placed on the sinoauricular node and the other electrode is moved about from place to place on the auricle, the deflection being noted at each new position, the electrode on the node will always be found to be negative to the other electrode;* which, however, will not be the case if both electrodes are moved about on other parts of the auricle. (3) As we shall see immediately, the current of action of the beating heart may be recorded by connecting a galvanometer with various parts of the body; for example with the right fore limb and the left hind limb. On the electrocardiogram thus obtained are several waves, one of which, called the P-wave, can easily be shown to correspond to the contraction of the auricle (see Fig. 261). If now we compare such elec- trocardiograms with those obtained while contractions of the auricle are produced by applying artificial stimulation to various parts of it. it will be found that the artificial simulates the normal curve only when the stimulated part is in the neighborhood of the sinoauricular node. In other words, it is only when the stimulus is applied to the sinoauric- ular node that a characteristic P-wave is obtained. When the appendix or the superior vena cava is stimulated, the P-wave is distorted although the other waves of the electrocardiogram may be normal. (4) By taking simultaneous electrocardiograms from direct leads placed on the auricle and comparing the record with that of a standard limb lead taken simultaneously, we shall find by exact measurement that the time of onset of the excitation Avave of the auricle, as measured in •relationship to the'P-Avave on the standard electrocardiogram, is shortest *The connections between the electrodes and galvanometer are always arranged so that any upward movement of the shadow of the string above the line of equal potential at the two electrodes indicates electric negativity. 190 THE CIRCULATION OF THE BLOOD when one electrode is over the upper end of the sinoauricular node, and that in other regions of the auricle it always appears at a later interval. Further details 011 this subject will be found in the papers by Eyster and Meek8 and in Lewis, monographs. Frequently, in taking electrocardiograms from different parts of the auricle, it is found that certain of the curves show small waves of positivity below the line of equal potential preceding the main wave of negativity. These initial deflections are most marked when both the electrodes are far removed from the sinoauricular node — for ex- ample, when they arc placed on the auricular appendix; but they are never present when Fig. 55. — Simultaneous electrocardiograms to show the cause for extrinsic deflections. The upper curves are from the appendix and the lower ones from lead II. The chief or intrinsic deflection (Tn) is seen to disappear in the right-hand appendix electrocardiogram, because the base of the appendix has been crushed. The extrinsic deflection (Ex) remains, as do the ven- tricular deflections (F1 Vz). (From Lewis.) one of the electrodes is placed on the sinoauricular node itself. In other words, curves taken from leads at a distance from the sinoauricular node are more or less composite in form, being made up partly of the main deflection due to the arrival of the excitation and partly of the secondary deflections dependent upon extrinsic influences acting on the electrodes; that is, the electrode picks up electric discharges from distant areas of muscle while these are in a condition of contraction (Fig. 55). From these considera- tions it follows that the intervals between the intrinsic and extrinsic deflections should be longest in leads that are farthest from the node, and gradually become less as one of the contacts approaches the node, until over this structure the ex- trinsic deflection is no longer recorded. Such has been found to be the case. CHAPTER XXII THE PHYSIOLOGY OF THE HEARTBEAT (Cont'd) THE ORIGIN AND PROPAGATION OF THE BEAT (Cont'd)— FIBRILLATION Mode of Propagation in the Auricles From the mass of evidence we have little doubt that the heartbeat originates in the sinoauricular node, and the question now presents itself as to how the beat is propagated over the remainder of the auricles and into the ventricles. Regarding the propagation of the beat over the auricles, two possibilities exist: (1) it may spread uniformly over the muscular tissue of the auricular wall until it reaches the auriculoventric- ular node, or (2) there may be laid down between the sinoauricular and the auriculoventricular node a special strand of highly conducting tissue. It is no argument against this second possibility that we should so far have been unable by histologic methods to differentiate any such struc- tures. There is considerable practical importance attached to the solution of these questions, particularly with regard to the cause of certain types of cardiac arrhythmia, such, for example, as that known as nodal rhythm. Thus, it is evident that if the beat is transmitted uniformly over the muscular tissue of the auricle, then the whole auricle -will have con- tracted before the beat has reached the auriculoventricular bundle, by which it is then transmitted to the ventricles. On the other hand, if the beat should travel between the two nodes by special conducting tissue, then the impulse will have arrived at the auriculoventricular node be- fore the auricle has contracted. As a matter of fact, it is not quite settled yet as to which of these two views is the correct one, although the balance of evidence seems to favor the former — that is, that the wave is transmitted uniformly over the muscular tissue of the auricle. (Lewis.) The methods employed in attacking the problem have been essentially the same as those described above. One of them may be called the direct, the other the indirect. In the former, a series of pairs of contacts is placed on the auricle, each pair being in a radial direction to the sino- auricular node. The time at which the excitatory process arrives at that contact of each pair which is proximal to the sinoauricular node is accu- 191 192 THE CIRCULATION OF THE BLOOD rately determined from the galvanometric record. The exact distance be- tween the contact and the sinoauricular node is then measured and from the data the average transmission time is estimated. From his results Lewis3 concludes that the transmission rates are uniform from the node to all parts of the auricle, with the exception of the superior vena cava, in which the rate is considerably lower. One thousand millimeters per second represents very fairly the average rate at which the excitation wave travels. On the other hand, Eyster and Meek8 state that the wave is propagated throughout the sinus node, and that it spreads to the contiguous venae cavse and to the auriculoventricular node with con- siderable rapidity, reaching the mouth of the superior vena cava in 0.01 second, whereas its passage to the auricle itself takes 0.02 second. There is therefore a delay in the passage of the wave to the auricle, which indi- cates that the excitation must spread to the auriculoventricular node be- fore involving the right atrium. These authors conclude that "this leads to the inevitable conclusion that the cardiac impulse spreads to the ven- tricle and to the right auricle by different paths, and does not pass to the ventricle through the auricle, as ordinarily stated." In the second, or indirect, method, the onset of the negative wave from different leads in the auricle is compared against a standard. For the standard Eyster and Meek have used the record of the mechanical sys- tole of the auricle, but the interpretation of the result is extremely dif- ficult on account of the rate at Avhich the changes are occurring. Lewis, on the other hand, has used the standard electrocardiogram for purposes of comparison. Mode of Propagation of the Beat to the Ventricles After reaching the auriculoventricular node, the beat is transmitted to the ventricles along the auriculoventricular bundle — a fact which has been most clearly demonstrated by the experiments on heart-block. We have al- ready seen (page 174) that although each chamber of the heart of a turtle or frog has a rhythm of its own, this is much more pronounced at the venous end of the heart, and when the transmission of the beat to the ventricles from the auricles is obstructed or blocked, as by compression or partial cutting at the auriculoventricular junction, the ventricles, after coming to a standstill for a time, subsequently contract with a rhythm which is entirely independent of that of the auricles. In the mammalian heart the same results may be obtained by arrang- ing a clamp so that it compresses practically nothing but the auriculo- ventricular bundle (Erlanger.) If the compression is extreme, the rhythm of the ventricles is quite independent of that of the auricles, but if it is only partial, the ventricular systoles follow regularly every sec- THE PHYSIOLOGY OF THE HEARTBEAT 193 ond, third, or fourth auricular contraction. If after such a complete or partial heart-block has been instituted, the clamp is removed, it will usually be found that the heart-block disappears and the auricular and ventricular contractions fall back into their usual sequence. The im- portance of this discovery, apart from its physiologic interest, rests in the fact that it is exactly duplicated in clinical experience. If the pulse tracing of the radial artery is compared with that of the jugular vein in certain types of heart disease, it will be found that the auricle is beat- ing two or three times more quickly than the ventricles. In many of these cases it has been found on autopsy that definite lesions often syphi- litic in nature involve the auriculoventricular bundle. In other cases, however, such lesions have not been discovered. Sometimes the bundle is so severely diseased that the block is complete, the ventricles con- tracting quite independently of the auricle (Stokes-Adams syndrome.) In such cases it is assumed that the beat originates in the uninjured part of the bundle below the seat of the block. It should be pointed out here, however, that all cases of slow pulse in the arteries are not necessarily dependent upon heart-block, but may depend upon a slow beat of the auricle itself. This is called bradycardia. Sometimes after complete destruction of the auricuioventricular bun- dle the beat continues to be transmitted to the ventricle, and conversely this transmission has sometimes been observed to be upset by lesions not affecting the bundle. The explanation of both of these exceptional re- sults almost certainly is that the right lateral connection described above (page 184) is serving as the main pathway of transmission for the beat. The facility of conduction through the auriculoventricular bundle is subject to alteration by the impulses passing to it along the vagus nerve, particularly the left vagus. It can also be altered by certain drugs, especially digitalis and strophanthin. The clear demonstration that it is along this bundle that the beat is transmitted is strong evidence in favor of the myogenic hypothesis (page 171) concerning the transmission of the heartbeat, but it does not necessarily disprove the neurogenic hypoth- esis, for histologic investigation has shown that the bundle is closely surrounded by an intimate plexus of nerve fibers. Spread of the Beat in the Ventricle After the impulse has been transmitted by the bundle into the ven- tricles, it spreads along the many branches into which, as we have seen, the tAvo main divisions of this bundle separate. The first part of the ventricular musculature to contract is therefore located near the ter- mination of these branches, at the papillary muscles. That these should contract before the rest of the muscle of the ventricles, has an obvious 194 THE CIRCULATION OF THE BLOOD significance in connection Avith their function of tightening the chordse tendineas so as to prevent any bulging of the flaps of the auriculoven- tricular valve into the auricles when, at the beginning of the presphygmic period, the high intraventricular pressure is brought to bear on their under surfaces. After starting at this point in the ventricle, the con- traction wave seems to spread farther through the ventricular muscle at a fairly uniform rate. Investigation of this problem by means of the galvanometer has been technically a very difficult matter, and the details of the researches by Lewis and his pupils have not as yet been published in full. According to the preliminary communications at hand, however,3a it appears that, Fig. 56. — Diagram of experiment by Lewis showing the times at which the excitation wave appeared on the front of the heart relative to the upstroke of R in lead II. K.A., right appen- dix; D.B.L., descending branch of left coronary artery. (From Thomas Lewis.) when nonpolarizable electrodes are placed at various parts of the outer aspect of the ventricle, and comparison made of the moments at which the cardiac impulse arrives, as judged by the appearance of the excita- tion wave relative to R in a standard electrocardiogram, it has been found that the time of arrival bears no relationship to the anatomic ar- rangement of the muscle bundles of the ventricle. It arrives early and simultaneously over an area of the surface near the anterior attachment of the wall of the right ventricle. It arrives late at the base of the right ventricle and in the part near the posterior intraventricular groove. Histologic examination has shown that the branches of the right division of the auriculoventricular bundle are most closely connected with the THE PHYSIOLOGY OF THE HEARTBEAT 195 place where the wave arrives earliest. Somewhat different results are obtained from the left ventricle, but again they are dependent upon the relationship of the part to the Purkinje fibers (Fig. 56). FIBRILLATION OF THE HEART Ventricles The even spread of the wave of contraction over the heart depends on the uniform excitability of the muscular fibers. If certain of the muscu- lar fibers, or bundles of fibers, have a greater or less excitability than others, then, when the stimulus to contract arrives, it will not produce a uniform contraction of neighboring bundles, and coordinated action of the cardiac musculature will give place to a confused movement in which each part of the heart is contracting independently of the rest. This fibrillation, or delirium cordis, as it is called, can be produced by a large variety of experimental methods, such, for example, as by stimulating the ventricles with induced electric shocks, or by ligation of a large branch of the coronary artery, or by the injection of lycopodium spores into the coronary circulation, or by mechanical stimulation of the heart in the region of the auriculoventricular bundle. Fibrillation of the ventricles is undoubtedly a common cause of death in man, for of course the confused movements make the ventricles in- capable of contracting on the contents of the heart. It is a condition which can probably never be recovered from in the higher animals, but it is of interest that the ease with which it is set up as the result of the application of an electric stimulus varies to a marked degree in differ- ent animals, and that in those hearts in which fibrillation can be elic- ited only with difficulty, recovery can usually be effected either by stop- ping the heart by means of cold and then allowing it to beat again, or by the administration of epinephrine. Of the hearts investigated in this way, that of the rat has been found to be most resistant to stimula- tion; then in order come those of the rabbit, the cat, the dog, and the horse. There is good reason to believe that the heart of man is readily affected. Fibrillation of the ventricle is undoubtedly the main cause of death in most cases of electrocution. Curiously enough, however, it has been stated that, whereas, a current of ordinary intensity (2300 volts alternating current) produces ventricular fibrillation in the heart of cer- tain of the lower animals, at least in that of the horse, a very much stronger current does not do so, and may indeed cause ventricular fibril- lation produced by a more moderate voltage to disappear. Unfortu- nately, however, these stronger currents produce irreparable damage in 196 THE CIRCULATION OF THE BLOOD the central nervous system, so that the method of applying stronger cur- rents, even were it feasible to do so quickly enough, would be of no therapeutic value in removing fibrillation. The disappointing results that have followed the repeated attempts to resuscitate persons killed accidentally by electric shocks is undoubt- edly dependent upon the fact that in the heart of man it is impossible to bring back the normal beat after the ventricles have been thrown into fibrillation. Fibrillation of the ventricle is also the cause of the sudden cardiac failure occurring when blood clots or emboli cause a blockage of the coronary circulation (it is sometimes the cause of angina pec- toris, for example). It must also be remembered in clinical practice that mechanical stimulation of the ventricles may produce fibrillation, so that in attempted resuscitation by cardiac massage care should be taken not to apply this too vigorously. Auricles Although ventricular fibrillation is seldom recovered from, it has been clearly shown in recent years that fibrillation of the auricles is relatively common and that it is by no means immediately fatal. Indeed it is one of the most common of the chronic cardiac disorders in man. Auricular fibrillation can be produced experimentally by the application of a strong electric stimulus to the auricles. If, however, a weaker stimulus is applied, the auricles do not go into typical fibrillation, but come to beat at a very rapid and regular rate, perhaps three or four hundred a minute. This condition is called "auricular flutter," and is quite fre- quently observed in the clinic. The influence of auricular fibrillation and flutter on tie beat of the ven- tricle is an extremely important one in connection with the irregular- ities of the heart observed in man, and this influence in most cases is explained by considering (1) the narrowness of the path (in the auric- uloventricular bundle) along which the impulses have to travel, and (2) the varying conditions of excitability of the ventricular muscle, depend- ing upon the existence of the refractory phase (page 180). In auricular flutter, when three or four hundred impulses per minute are passing along the bundle to the ventricle, the contraction produced by the first one will scarcely have started before the second and imme- diately succeeding ones arrive, so that the ventricle will beat at a rate that is much less than that of the auricle, and a condition simulating heart-block will become established. The characteristic feature which distinguishes this from true heart-block, however, is the fact that the ventricular rate is above normal, whereas in true heart-block the rate is much below normal. By means of the electrocardiogram or by THE PHYSIOLOGY OF THE HEARTBEAT 197 polysphygmographie tracings, it can also be shown that the auricle is beating with perfect regularity although very rapidly. In auricular fibrillation the ventricles obviously will respond at a very irregular rate to the impulses transmitted to them, and the auricular contractions, if examined by the methods above described, will show no regular sequence. Further details of the .method of eliciting these signs will be described later (page 266). CHAPTER XXIII THE BLOODFLOW IN THE ARTERIES THE PULSES Returning to the physical laws that govern the circulation of the blood, we may now consider the temporary changes produced in the bloodflow in the arteries by each systolic discharge. These changes go under the general term of the pulses, of which three may be distinguished: (1) the pressure pulse, or the pulsatile increase of pressure produced by each heartbeat (see page 127) ; (2) the velocity pulse, or pulsatile accel- eration of velocity; and (3) the palpable pulse, or the pulsatile expansion of the walls of the blood vessels produced by the sudden change of blood pressure in their interior. The general characteristics of the three pulses are the same, certain features being however more pronounced in one than in another. General Characteristics Rate of Transmission of Pulse Wave. — The rate of transmission of the pulse wave can be determined by taking simultaneous tracings of the pulses from two far distant parts of the arterial system along with accurate time-tracings. From records (cf. Fig. 98) taken from the apex or the carotid and radial arteries we can determine how long it takes for the beginning of the pulse wave to travel to the radial artery from the point in the aorta from which the carotid artery springs. We shall find that it takes about one-tenth of a second, which, considering the length of the artery involved, would work out as a transmission velocity of about seven meters per second or about seventeen miles an hour. The pulse therefore travels along the blood vessels at a much greater speed than the blood itself is moving, this being, as we shall see immediately, about 0.5 meters per second in the larger blood vessels. The pulse is a wave of sudden increase in pressure and velocity pass- ing along a stream which is flowing in the same direction with a cer- tain more permanent pressure and velocity. A simple physical experi- ment may serve to make this clear: If the first of a row of billiard balls be tapped with the cue, the end balls will fly off while the others are moving slowly along in the direction of the stroke. Each ball becomes accelerated by the ball behind it, and imparts its influence to the ball 198 THE BLOODFLOW IN THE ARTERIES 199 in front. In other words, a pulsatile acceleration of velocity is produced by a pulsatile change in pressure between each two balls. The existence of a pulse wave going in the same direction but quicker than a moving column of fluid can also be illustrated by observing the waves traveling down a stream when a stone is thrown into it. The length of the pulse wave is such that the beginning of it has ar- rived at the periphery of the arterial system before the end has disap- peared from the beginning of the aorta. This is important to remem- ber, for it is a common mistake to think of the wave as being a local one. The determination of the length of the pulse wave depends upon the following equation: L = VT, where L equals the length of the pulse wave, V its velocity of transmission, and T its duration at a given point in the artery. Under ordinary circumstances L would usually work out from 3.25 to 4.5 meters. The rate of transmission of the pulse wave varies according to the rigidity of the walls of the arteries. To understand why this should be so, it will be well for a moment to consider the physical conditions upon which the pulse wave depends. If we connect a piece of rigid tube with the nozzle of a large syringe, with each movement of the pis- ton a wave of pressure will be transmitted to the fluid in the tube, along which it will travel at such a high velocity that it will arrive at the free end of the tube almost instantaneously, and incidentally the out- flow of fluid from the end of the tube with each compression of the pump will be exactly equal to that represented by the movement of the piston. If, on the other hand, an elastic tube is employed, it will be found that the sudden increase of pressure produced by each stroke of the pump causes a distention of the walls, which travels along the tube as a wave at a readily measurable velocity, which is slower the more extensible the tube. Moreover, the outflow of fluid from the free end of the tube will continue for some time after the cessation of the move- ment of the pump. What happens in the tube with each discharge of the fluid is that the portion which is immediately adjacent to the pump undergoes distention and, being elastic, tends immediately afterward to recoil and thus exert a recoil pressure on the fluid contained in the tube. As a result, pressure waves are set up in the fluid in all directions. Those that travel back come to a stop because of the piston, while those that travel distally act on the fluid in front of them so as to accelerate it and by temporarily raising its pressure distend the next segment of the vessel wall, until the end of the tube is reached. From this considera- tion it is clear that the more extensible and elastic the wall of the tube is, the longer will it take for the wave of pressure to travel from one end to the other. 200 THE CIRCULATION OP THE BLOOD Alteration in the rate of transmission of the pulse wave in the arter- ies of man depends entirely upon an application of these principles. When the arteries become hardened in old age, the rate of transmission of the pulse wave is markedly increased. The pulse is also transmitted more rapidly in the vessels of the lower extremities than in those of the upper, since in the former the blood vessels are somewhat more rigid. Delay in the transmission of the pulse wave is further observed as one of the signs of aneurism in a vessel; as is well known, aneurism of the subclavian artery on one side causes a delay of the pulse on that side that is perceptible to the fingers. The Contour of the Pulse Curves For more particular study of the exact contour of the pulse wave, and especially for determining the time relationships of the secondary waves, Fig. 57. — Diagram of Chauveau's dromograph. a, tube for introduction into the lumen of the artery, and containing a needle or vane, which passes through the elastic membrane in its side and moves by the impulse of the blood current; c, graduated scale for measuring the extent of the oscillations of the needle. a large variety of methods of varying degrees of accuracy have been elaborated for each kind of pulse. Those devised for measuring the pressure pulse have already been de- scribed (see page 127), and for the other pulse® they are as follows: Velocity Pulse. — Much ingenuity has been displayed in the elabora- tion of methods for recording the velocity pulse. In one of these the artery is cut across and the ends attached to a tube, into the lumen of which projects a paddle or vane articulated with a light lever, which passes through its wall (see Fig. 57). The vane floats in the blood stream, and the outer end of the lever to which it is attached is con- nected with some device to record its movements, which vary with the velocity of bloodflow (hemodromograph). Another method consists in the application of the instrument known as Pitot's tube used by phys- icists. This consists of a horizontal tube having two side tubes, each of THE BLOODFLOW IN THE ARTERIES 201 which is connected at its outer end with a manometer and prolonged inside the horizontal tube, where they are bent at opposite right angles, so that the inner end of one of them — the proximal tube — points up Fig. 58. Fig. 59. Fig. 58. — Diagram to show principle of Pilot's tubes for measuring velocity pulse. In both tubes the fluid will rise because of lateral pressure, but in the proximal (left-hand) tube it will rise higher than in the distal, because it will also be affected by the velocity of flow. Fig. 59. — Diagram to illustrate the principle of Cybulski's Photo-hematotachometer. The fluid in C stands higher than that in D in proportion to the velocity of flow of the blood along AB. Fig. 60. — Dudgeon's sphygmograph. (From Jackson.) stream, and records not only the lateral pressure but also the pressure produced by the sudden increase in velocity of the flow, while the 202 THE CIRCULATION OF THE BLOOD other — the distal tube — being bent down stream, records merely lateral pressure. A photographic record of the movement of the fluid in the two tubes gives the velocity pulse (see Fig. 58). For physiologic pur- poses the form of apparatus used is constructed as shown in Fig. 59. Palpable Pulse. — To secure a record of the palpable pulse, the so- called sphygmograph is employed, although a tambour having a button in the center which is made to press on the artery may also be em- ployed. The commonest form of sphygmograph is that known as Dudgeon's (Fig. 60). It consists of a small button connected with a spring, the movements of which are transmitted and magnified by means of a system of levers connected with a writing point arranged so as to inscribe its movements on a moving surface. The Analysis of the Curve The general contour of the pulse waves taken by any of the above methods are in general very much the same. The pressure and velocity Fig. 61. — Pulse tracing (sphygmogram) taken by sphygmograph. a d, the period of the pulse curve; b, the primary; c, the dicrotic wave. Time marked in fifths of a second. (From Prac- tical Physiology.) pulse curves are, however, not usually taken for the purpose of observ- ing the contour of the wave but rather for measuring the difference in pressure or velocity actually produced during each pulse. It is a record of the palpable pulse that is usually employed for studying the contour of the wave and the presence of secondary waves. A record of the pal- pable pulse wave (Fig. 61) shows two separate waves on the descending limb of the main wave. If a large number of similar pulse curves are taken from different individuals or from the same individual under different conditions, it will be found that of these two waves the second one is by far the more constant ; and if the waves are timed in relation- ship to the heart sounds, this second wave always occurs immediately after the second sound, allowance, of course, being made for the time required for the pulse to be transmitted from the heart to the artery from which the pulse tracing is being taken. If the observation is made very carefully, it can indeed be shown that the second sound cor- responds exactly to the notch which precedes this wave. The waves that THE BLOODFLOW IN THE ARTERIES 203 precede this notch can not be related to definite changes occurring in the heart. Evidently, then, the secondary pulse waves are not all of equal significance, by far the most important being that which occurs immediately after the second sound, called the dicrotic wave (c), the notch in front of it being called the dicrotic notch. Any secondary waves occurring before the dicrotic are called predicrotic, or if they occur on the ascending limb of the main pulse wave, as they sometimes do, they are called anacrotic. Waves occurring after the dicrotic are called postdicrotic. The relative importance of the dicrotic, in comparison with the pre- dicrotic and postdicrotic waves, is further evidenced by the fact that it alone is seen on a so-called hemataugram, which is the tracing ob- tained by allowing a fine stream of blood, escaping from a pinhole made in the wall of an artery, to impinge upon a moving sheet of white blot- ting paper. That such a tracing shows a dicrotic but no secondary wave, indicates that only the former is present in the blood stream itself, and that the other secondary waves must be produced by some condition arising either in the elastic tissue of the walls of the blood vessels, or in the elastic properties of the instruments used for taking the pulse tracing. The Dicrotic Wave. — Because of its obviously greater significance, we shall first of all consider the exact cause of the dicrotic wave and of the notch preceding it. Theoretically, two possible causes might explain the wave: either it is due to some secondary wave set up at the heart, or it i$ dependent upon waves reflected from the periphery of the cir- culation back along the blood stream, just as secondary waves are re- flected from the walls of a tub of water when a stone is thrown in the center. In considering this second possibility, we are of course making the assumption that at the ends of the arterial system there is a sudden resistance to the onward movement of blood. The frequent branching which occurs when the arterioles open into the capillaries no doubt of- fers many opportunities for the reflection of pulse waves back to the heart, but these waves must be reflected at such varying distances along the arterial system that there can be little opportunity for them to be- come added together so as to form a wave of sufficient magnitude to make itself perceptible in the blood flowing in the larger arteries. These waves are relatively so small and they occur at such different times that the net result of their addition, so far as the production of a larger wave is concerned, must be practically nil. Notwithstanding these con- siderations, it is possible that under some conditions, such as in cases of high arterial tension, certain of the predicrotic or postdicrotic waves may be due to the above causes. 204 THE CIRCULATION OF THE BLOOD That the dicrotic is not a reflected wave is clearly established by the fact that if the distance between the dicrotic wave and the main pulse wave is measured at different points of the arterial stream, it will al- ways be found to be the same, which obviously would not be the case were the dicrotic wave reflected. If, for example, we were to examine the contour of the wave produced by throwing a stone into a tub of water, we should find that near the edge the secondary wave was very close to the main wave, whereas near the center the secondary wave would occur much later. Our problem therefore narrows itself down to an investigation of the cause for the dicrotic wave at the central end of the circulation. It occurs, as we have seen, immediately after the beginning of diastole. That it can not be due to anything taking place in the ventricle itself is evidenced by the fact that such a wave is absent from an intracardiac pressure curve (see page 151), although it is present in the very begin- ning of the aorta. Now, the only structures existing between those two points which could be held responsible for this wave are the semilunar valves — a conclusion which is sustained by the fact that, if the aortic valves are rendered incompetent by hooking them back, or if the pulse beat is examined in patients suffering from an aortic insufficiency, it will be found that the dicrotic wave is not nearly so evident as usual. To understand how the valves are responsible for the production of the wave, the mechanical changes occurring at the root of the aorta must be clearly understood (see page 155). The stretching of the elastic walls of the aorta which occurs with each systolic outrush of blood is fol- lowed by a powerful and sudden contraction of the stretched walls, and the pressure thus brought to bear on the column of blood in the aorta tends to impel it both forward and backAvard. The forward movement adds itself to the wave of increased pressure already produced by the ventricular contraction. The backward component travels as far as the semilunar valve, from which it is reflected, and now proceeds peripher- ally along the blood stream during the time at which the original pres- sure pulse is declining. It therefore imprints itself on the pulse trac- ing as a separate wave, and does so all the more markedly when the decline in the main pulse wave is rapid, as in cases in which the periph- eral resistance is low, but fails to be prominent when, on account of a high peripheral resistance, the decline in the main pulse wave is tardy. This explanation coincides exactly with the well-known clinical fact that the dicrotic wave is conspicuous in pulses of low tension, but ill marked or absent in pulses of high tension. One point remains to be considered, and that is the cause for the sudden decline in the main wave at the cessation of the ventricular out- THE . BLOODFLOW IN THE ARTERIES 205 put, for, it might be said, why should there be such a sudden fall in pressure near the heart, whereas toward the periphery, as we have seen, the pressure between the heartbeats tends to be maintained on account of the elastic recoil of the stretched arterial walls. The explanation usually given is that the sudden cessation of outflow of blood from the ventricle at the end of the sphygmic period causes a negative pressure to be produced in the blood at the beginning of the aorta, thus tending to cause a reflux of blood towards the heart, the effect of which is (1) to bulge the closed valves, and (2) to produce the reflected dicrotic wave. If, while fluid is flowing under pressure along a tube, the flow is sud- denly arrested by turning a stopcock, it is possible by the use of manom- eters to show that a negative wave is set up immediately beyond the stopcock, and that this negative wave travels along the tube at a rate depending on the elasticity of its walls. Causes for Disappearance of the Pulse in the Veins The disappearance of the pulse in the capillaries and its consequent absence in the veins we have already seen to be owing to the combined influence of the elasticity of the vessel walls and the peripheral resist- ance. On account of these two factors the pressure conveyed to the blood during systole is stored up to be released during diastole by the recoil of the stretched vessels. Sometimes, however, the pulse gets through to the veins, either because the elasticity of the vessels is not so marked, or because the peripheral resistance has been lowered (vaso- dilatation). In patients with hardened arteries, or in normal individu- als after taking nitrite, which dilates the peripheral arterioles, a pulse may come through at the periphery and appear in the veins. This may be called the peripheral venous pulse, and it is to be carefully distin- guished from the central venous pulse observed in the large veins, as at the root of the neck, before any valves have intervened to block the transmission of the auricular pressure wave back into the column of blood in the veins. If a pulse is seen in a large vein and there is doubt as to whether it is peripheral or central in origin, this doubt can be immediately removed by locally constricting the vein; if the pulse is peripheral, it will disappear on the heart side of the constriction; if it is central, on the side away from the heart. CHAPTER XXIV THE RATE OF MOVEMENT OF THE BLOOD IN THE BLOOD VESSELS Since the object of the circulation is to maintain an adequate move- ment of blood in the tissues and capillaries, it is evident that besides measuring the pressure of bloodflow, we should also measure the rate of its movement, or, as it is often called, the mean velocity. This measure- ment may be undertaken either for a given vessel or for a complete vascular area, such, for example, as that of one of the viscera or one of the extremities — the mass movement of the blood. Or instead of measuring the mean velocity we may desire to know how long it takes for a particle of blood to traverse a given vascular area. Such a meas- urement is called the circulation time ; it does not at all tell us how long it takes for all the blood to pass through the given area, but only, as stated, the time required for the circulation of a fraction of the blood through a particular field. VELOCITY OP PLOW IN A VESSEL Special methods have been devised for the measurement of each of these three velocities. For the measurement of the velocity of flow through a main artery or vein, methods similar to those employed by hydraulic engineers are employed; that is to say, the volume of blood, in cubic centimeters, which passes a given point is measured for a given time, and the result divided by the cross section of the vessel at the point of observation. The result gives us the mean lineal velocity. To measure the outflow of blood in a given time, the simplest method would be to cut across the vessel and collect the blood in a graduate, but obviously in this method an error wouldx be introduced, because cutting the vessel would lower the peripheral resistance and remove the natural obstruction to the flow present in the intact animal. Moreover, the hemorrhage would in itself introduce a disturbing factor on account of the loss of circulating fluid. To make such measurements of any value, it is obviously necessary to retain the peripheral resistance. For smaller vessels this can be done by introducing in the course of the artery a long glass tube bent in the 206 RATE OF MOVEMENT OF THE BLOOD 207 shape of the letter U (Fig. 62), or by merely allowing the vessel to bleed into a graduated tube and seeing how long the blood column takes to travel from one end to the other. This method is of considerable value in measuring the velocity of flow from small vessels such as the veins coming from glands and muscles. For larger vessels a so-called stromuhr is employed. There are numerous forms of stromuhr; that shown in the diagram (Ludwig's) (Fig. 62) consists of two glass bulbs united above, and connected below with tubes that open flush with the surface of a brass disc. This is pivoted at its center with another similar platform also having flush with the surface the openings of two tubes con- nected with the cut ends of the artery or vein. In a certain position of the platform, the tubes from the artery or vein are exactly opposite those of the bulbs, so that the blood can flow from one end of the vessel Fig. 62. — Forms of apparatus for measurement of blood velocities. i. Volkmann's hemodromometer. The blood vessel is attached to the two short side tubes, and according to the position of the stopcock, the blood flows either directly between them or through the U-shaped glass tube. z. I/udwig's stromuhr. The tubes on the lower end of each of the two glass bulbs pierce a circular brass platform and end flush with its surface. This platform pivots at its center on a similar lower platform with two openings connected with the tubes that lead to the blood vessel. through the bulbs to the other end. To use the instrument the proxi- mal bulb is filled with oil and the peripheral one with physiologic saline. The clip is then removed from the central end of the artery, and the blood flows in and displaces the oil, which in turn displaces the saline in the peripheral end of the artery. When the blood has risen to a mark on the tube joining the two bulbs, the instrument is rapidly rotated so that the oil is brought back again into the proximal position, the rotation being effected so quickly that there is no distinct interruption in blood- flow. The operation is repeated in this way for a given period of time. Counting accurately the number of revolutions, then multiplying the number of revolutions by the capacity of the bulbs, we get in cubic 208 THE CIRCULATION OF THE BLOOD centimeters the amount of blood that has flowed through the instrument in a definite unit of time. This gives us the volume flow and, if the result is divided by the cross section of the vessel in square centimeters, we obtain what is known as the mean lineal velocity. Many modifica- tions have been made of this instrument, but it is unnecessary to go into them here. /" The general result of such measurements has been to show that the I lineal velocity is inversely proportional to the cross section of the vessel I at the point of observation. It is obvious that the volume of blood flowing out of the .heart to the aorta in a given time is exactly equal to that flowing into it by the vena cava, and likewise that the volume floAving into an organ is exactly equal to that which flows out. Conse- quently the lineal velocity will be inversely proportional to the sec- tional area of the vessel. The principle is the same as that which gov- erns the velocity of flow of a stream: when the bed is narrow, the cur- rent is swift, but it becomes sluggish when the bed is wide. If the arteries were of the same caliber as the veins, the mean velocity of the bloodflow through the two would be the same, but actually it is much greater in the arteries because the lumen of these at a given point in the circulation is only from one-third to, one-half , that of-the corresponding vein. It must be understood that we are dealing above with the mean velocity in a unit of time, and that there must be considerable alteration with each systole and diastole, constituting the velocity pulse (page 200). The degree of this alteration with each velocity pulse is much less at the periphery of the circulation than near the heart. As the periphery is reached, the flow becomes more uniform. It must further be re- membered that, although the mean velocity depends essentially upon the area of the vascular bed, yet it is subject to considerable variations as a result of changes either in the force or rate of the heartbeat or in the facility of outflow from the ends of the arterial system — that is, changes in peripheral resistance. It is usually stated that the mean lineal velocity in the carotid artery is about 300 millimeters per second; and in the jugular vein, about 150 mi-llimeters ; whereas in the capillaries, where the total area of the vascular bed has become enormously increased, being perhaps some 800 times that of the aorta, the velocity of -flow is only about half a milli- meter per second. MASS MOVEMENT OF THE BLOOD IN A VASCULAR AREA Methods. — In considering the bloodflow or mass movement of the blood in the different regions of the body, it is usually more practical to RATE OF MOVEMENT OF THE BLOOD 209 measure, not the mean lineal velocity of the inflowing and outflowing blood, but rather how many cubic centimeters of blood are traversing the part per 100 grams of organ or tissue per unit of time. Such meas- urements may be made in a variety of Avays. If there are but one artery and one vein to the part, the stromuhr may of course be employed, and it may be inserted in either the arterial or the venous circuit. For measuring the mass movement of blood through such large viscera as the liver, this is indeed the only method that can be employed, the stromuhr being inserted either in the course of the portal vein and he- patic arteries, or, better still, in the vena cava just below the openings of the hepatic vein, the vena cava being shut off for a moment between the liver and the heart and the blood, as it flows from the hepatic vein, allowed to collect in the stromuhr. For other organs and tissues, how- ever, methods which do not involve any interference with the blood vessels may be employed. One of these is the so-called plethysmograpkic method of Brodie. An organ, such as the kidney, is enclosed in a plethys- mograph (see page 230,) and while a record of its volume is being inscribed on a quickly revolving drum, the vein is suddenly clamped, with the result that the kidney volume expands in proportion to the mass of blood flowing into it. When the expansion has reached a cer- tain degree, the clamp is removed and the bloodflow allowed to pur- sue its course. It is then an easy matter, by graduating the plethys- mograph, to determine how many cubic centimeters of blood must have flowed into the organ in the given time. To avoid serious local asphyxia in the tissue, the clamp must be applied to the vein for only the briefest period of time. This method may also be employed for measuring the bloodflow through the extremities. Thus, if the arm is enclosed in the plethysmograph (Fig. 63) and a band encircling the arm above the plethysmograph is tightened so as to constrict the veins but not the ar- teries, the rate at which the volume of the arm within the plethysmograph expands will correspond to the rate at which blood is flowing into it (Hewlett). For the purpose of measuring blood flow through the upper or lower extremities, a much more serviceable clinical method is that of G. N. Stewart. This depends on the principle that, provided the blood passing from the thorax to the hands or feet is of constant temperature, the rate at which heat is dissipated from the hands or feet will be directly proportional to the rate of movement of the blood through these parts. Fortunately for the method, the hands particularly, but also the feet, are more or less perfect radiators — at least they are to this extent, that if the temperature in their environment is not much lower than the temperature of the blood, then while this is traversing the part, it will 210 THE CIRCULATION OF THE BLOOD lose heat to the environment until the outflowing or venous blood is at exactly the same temperature as the environment; for example, if the hand is placed in water that is a little cooler than that of the blood, and the temperature of the blood in one of the large veins of the hand is measured, it will be found to be the same as that of the water in the water-bath. To measure the rate of flow, therefore, we must ascertain: (1) how much heat has been given out by the part to the water surrounding it in a given time, and (2) the difference in temperature of the inflowing (arterial) and outflowing (venous) blood. We measure the amount of Fig. 63. — Plethysmograph for recording volume changes in the hand and forearm. By observ- ing the rate with which the volume increases when the arm is compressed, the mass movement of the blood can be determined. (From Jackson.) heat given out to the water in calories, a calorie being the amount of heat required to raise the temperature of 1 c.c. of water from 0° C. .to 1° C. Suppose, for example, a hand were placed in 3,000 c.c. of water at 33° C., and that after ten minutes the temperature had risen to 33.5° C., then the amount of calories given out would be 3,000 x 0.5= 1500. Since calories equal cubic centimeters multiplied by change in temperature, it follows that if we divide the figure representing them by the actually observed difference in temperature between inflowing and outflowing blood, the result must equal the number of cubic centimeters of blood that has flowed through the part. The temperature of the in- flowing blood has been found to be practically identical with that of the RATE OF MOVEMENT OP THE BLOOD 211 mouth under the tongue; whereas of course the temperature of the venous blood, as already explained, is equal to the mean temperature of the water during the time that the hand was immersed in it. Further de- tails of the teehnic of this method will be found elsewhere, but it may be said here that it is extremely simple and accurate, and that it requires nothing more than (1) an accurate thermometer ranging between abo^J 40° C. and 50° C., with a scale so drawn out that readings can be made to Moo of a degree, and (2) a well-constructed vessel of about 3,000 c.c. capacity, with double walls, the space between them being packed with some heat-insulating material such as ground cork. Results. — Regarding the results obtained with these methods, it has •been found that the blood supply for each 100 grams of tissue in the viscera, as measured by the stromuhr method, is about as follows: stomach, 21 c.c. ; intestine, 71 c.c.; spleen, 58 c.c.; liver, arterial, 25 c.c.; liver, venous, 59 c.c.; liver, total, 84 c.c.; brain, 136 c.c.; kidney, 150 c.c.; thy- roid gland, 560 c.c. The large blood supplies of the thyroid gland and of the kidney are the most striking results of these observations. By the use of the calorimeter method the bloodflow through the hands and feet of a healthy young man has been found to be about 13 grams per 100 c.c. of hand per minute for the right hand, and about half a gram less for the left. The footflow is only about one-third to one-half that of the hand per 100 c.c. of tissue — a difference which is largely owing to the greater proportion of skin and the smaller proportion of bone m the hand. The average footflow or handflow for a given indi- vidual under ordinary conditions is remarkably constant from time to time, but it is extraordinarily sensitive to changes in the temperature of the environment in which the subject has been living for some time previous to the measurement. In one individual, when the room temperature was 20° C., the flow in the right hand, expressed in grams of blood per 100 c.c. of hand or foot, was 10.1; when it was 22.8° C., the flow was 12.8; when it was 25° C., 12.1 ; when it was 30° C., 18.5. On account of the influence of temperature on the flow, it is extremely important that the measurements should be made in a small room the temperature of which is kept constant, or if it must be made in the wards, the bed should be sur- rounded by curtains. The measurements made on the hands of dispensary patients shortly after coming in from outside air are very likely to be fallacious. The importance of making such bloodflow measurements in the clinic will be alluded to later. Of course the measurements made by the above method in man tell us only the rate of flow in the periphery of the body, and furnish us with no in- formation regarding the flow of blood through the viscera. It is, how- ever, a well-established fact that the bloodflow in the central part of the 212 THE CIRCULATION OF THE BLOOD circulation is more or less reciprocal with that at the periphery, an increase in the one place being accompanied by a corresponding de- crease in the other. The Visceral Bloodflow In Man The visceral bloodflow in man can be measured indirectly in the case of the lungs, either, ( 1 ) by finding the quantity of oxygen absorbed by the blood during an interval of time that is less than that required for the blood to travel once round the circulation (60 seconds) and comparing this with the oxygen content of samples of arterial and venous blood, or (2) by causing a person to breathe a known quantity of nitrous-oxide gas and then finding the concentration of this gas in the blood after leaving the lungs. In the former method the difference in oxygen percentage be- tween arterial and venous blood will be less for a given absorption of oxygen from the alveoli the more rapid the circulation of blood through the lungs, and in the latter method for the absorption of a given amount of nitrous oxide, the less will be the concentration of this gas in the blood the more rapid the circulation. Obviously these estimations must be made only over periods of time, less than that taken for any of the blood to complete one circuit of the circulation. The methods are admittedly only approximate, but the results are of much interest, mainly because of the indication they give us as to the amount of blood pumped out by the ventricle with each heartbeat, or during a given period of time. The results have been found to. vary considerably; thus, one author (Krogh) places the output of blood per minute as between 2.8 and 8.7 liters, which would correspond, at a pulse rate of 70, to an output per heartbreat of from 40 to 120 c.c. An immediate and very marked increase has been found to occur during muscular work. By comparing the bloodflow through the hand with that through the lungs, an estimate can be formed in a given individual as to the relative magnitude of the peripheral and visceral moieties of blood. Interesting results, which will be referred to later, have been obtained from such measurements. The Work of the Heart Meanwhile it is of interest to note that we may calculate from the ventricular output of the blood the amount of work that the heart is doing in maintaining the circulation. Of course the calculation is again only approximate, since we have to assume certain figures. If we assume that in a 70-kilogram man the quantity of blood is 4,200 c.c. (see page 85), and that it takes about one minute for all the blood to complete a cir- culation, then the work performed by the left ventricle .in one minute BATE OF MOVEMENT OF THE BLOOD 213 will be equal to that done in raising the above quantity of blood to a height corresponding to the mean pressure in the aorta. If we take this pressure as 130 millimeters of mercury, which would correspond to -a column of blood 1,755 meters high (13.5 x 130=1755 mm. blood, or 1.755 meter), the work done by the left ventricle would be 1.755x4.2=7.37 kilogram-meters in one minute, or in twenty-four hours roughly about 10600 kilogram-meters. The work done by the right ventricle is probably about one-third that of the. left, this being about the ratio of the pres- sures in the two chambers. The total work of the two ventricles is there- fore about 14000 kilogram-meters. This represents an enormous amount of work; indeed it has been computed that it is sufficient to raise a man of 70 kilograms to about twice the height of the highest skyscraper in New York. The work thus expended in forcing the blood through the capillaries becomes converted by friction in the small blood vessels into heat, the heat equivalent of the above amount of work being roughly about 350 calories (see page 537). THE CIRCULATION TIME The circulation time, or the time taken by a drop of blood to travel between two points in the circulation, can be determined in laboratory animals by a variety of methods, all depending on the principle of seeing how long it takes for a drop of some substance injected into an artery to appear in the corresponding vein. For example, to determine the time taken for a drop of blood to pass from the jugular vein into the carotid artery in a rabbit, a solution of methylene blue in isotonic saline is in- jected into the former vessel and the moment of its appearance through the walls of the artery determined by a stop-watch. If the walls are too thick to admit of the employment of this method, a strong solution of sodium chloride may be substituted for the methylene blue, and the mo- ment of its appearance at another point of the circulation determined by observing the electrical conductivity of the vessel. Since the con- ductivity of a blood vessel depends partly on the concentration of elec- trolytes in the blood flowing through it, the moment at which the salt solution appears will be indicated by a change in electrical resistance (G. N.Stewart). By such methods, it has been found that the time for the pulmonary circulation is very short compared with that of the systemic circulation. In a rabbit it is usually a little less than four seconds; in an average- sized dog of about 12 kilograms, it is about eight seconds; and in man it is computed to be about fifteen seconds. On the other hand, the cir- culation time in such viscera as the spleen and kidney is relatively long, 214 THE CIRCULATION OF THE BLOOD and more susceptible than that of the lungs to different conditions of temperature. In a dog in which the pulmonary circulation time was about 8.5 seconds, that of the spleen was about 11 seconds, and of the kidney about 17.5 seconds. The shortest circulation time of all is of course that in the coronary artery, but that through the retina can not fall far behind it. % To determine the total circulation time, we must know: (1) the average amount of blood passing by each part in a given time, and (2) the average circulation time of each part. From such computations, which however are obviously subject to considerable error, it has been reckoned that the total circulation time in man must lie somewhere between 1 and 1.25 minutes. MOVEMENT OF BLOOD IN VEINS Before leaving this part of our subject, a few words may be said con- cerning the forces concerned in the movement of blood in the veins from the capillaries to the heart. By the time that the venules are reached, owing to friction in the capillaries the blood will have lost most of the force imparted to it by the heart action. Nevertheless, this remaining vis a tergo must be considered as the basic cause for the movement of the venous blood near the periphery. As the venules get larger, two other factors come into play: the massaging influence of the muscles, and the valves of the veins. By the movements of the muscles the veins which lie between will be rhythmically compressed, and this will tend to cause the blood to be moved forward and backward in them, the back- ward movement being however prevented by the operation of the valve's. When the tonicity of the muscles is subnormal, as in conditions of ill health, the absence of this massaging action permits the blood to stag- nate in the veins, especially in those of the lower extremities in upright animals, with the consequence that the veins become dilated, particularly just above the valves, thus causing the condition known as varicose veins. As the thorax is approached, two other factors become operative: the aspirating influence of the thorax during inspiration, and the negative intraventricular pressure (see page 152). There is no doubt that the former of these is of considerable importance in maintaining the venous return near the heart, for although the change of pressure induced by in- spiration amounts to only 5 millimeters of mercury, yet it acts so slowly that it produces a considerable influence. The aspirating effect of the ventricle at the beginning of diastole is, however, of no sig- nificance in attracting blood to the heart, for although, as we have seen, it may be considerable, yet it lasts for so short a time that it could not RATE OF MOVEMENT OF THE BLOOD 215 overcome the inertia of the column of blood in the vena cava. Even if the negative pressure did last for a longer period, it could not attract more than a small amount of blood, because it would cause the thin collapsible walls of the veins to come together and thus block the pas- sage towards the heart. CHAPTER XXV THE CONTEOL OF THE CIRCULATION The available blood in the body is parceled out to the various organs and tissues according to their relative activities, and, since these vary from time to time, the question arises as to the nature of the mechanism or mechanisms involved in bringing about this adjustment. Two possible methods of increasing the supply are: an increase in the mass movement of all the blood in circulation, and a reciprocal adjustment of the resistance to the flow in different vascular areas brought about by vasodilatation in one and vasoconstriction in others. Both of these methods might operate together. Two agencies can be thought of as responsible for bringing about the above changes: (1) chemical substances or hormones, present in the blood, and (2) the nervous system. The influence of chemical substances, or hormones, (page 729) in the control of the circulation is undoubtedly an important one, and of those known at the present time two groups may be mentioned: (1) sub- stances which alter the hydrogen-ion concentration of the blood, and (2) so-called pressor and depressor substances, produced either by duct- less glands, such as the adrenal, or by the activity of tissues. An in- crease in hydrogen-ion concentration of the blood not only affects the heartbeat (see page 168), but causes a marked dilatation of the Mood vessels, so that both the central and the peripheral changes will be such as to encourage an increased flow of blood through the active organs or viscus. Thus, during muscular "activity of the leg muscles there will be a tendency to an increase in the hydrogen-ion concentration of the blood as a whole, resulting in a greater cardiac activity and a greater outrush of blood through the aorta, and at the same time the vessels of the acting muscle will have become especially dilated because of the production by the active muscles either of lactic acid or of carbonic acid. The active muscle also produces such substances as imidazole, which have a powerful vasodilating action. Such substances are sometimes called depressor. Though the hormone control of the circulation is undoubtedly of great importance, it is probably much less so than that exercised through the nervous system, and here again the control is centered partly in the 216 THE CONTROL OF THE CIRCULATION 217 heart and partly in the peripheral resistance. The nerve control of the heart is effected through the vagus and sympathetic nerves, and that exercised on the blood vessels, /through the so-called vasoconstrictor and vasodilator nerves. The activity of the nerve centers from which the cardiac and vaso- motor impulses are discharged is controlled by afferent impulses com- ing from the various regions of the body. When a gland becomes more active, we must suppose that stimulation of the sensory fibers has caused afferent impulses to be transmitted to the cardiac and vasomotor centers, upon which they act in such a way as to produce increased heart ac- tion and a local dilatation of the blood vessels of the active gland, with perhaps a constriction of the blood vessels of the rest of the body. THE NERVE CONTROL OF THE HEARTBEAT The Vagus Control With regard to the control exercised through the vagus nerve, we have already seen that the cutting of the two nerves in the neck causes the heart to quicken and the arterial blood pressure to rise, whereas a stimulation of the peripheral end of the nerve causes the heart to be- come slowed, if not stopped altogether, and the blood pressure to fall. For the more detailed investigation of the nature of the vagus control of the heart, it is necessary to observe the exposed heart itself — an ex- periment which, for obvious reasons, can be most simply performed in a cold-blooded animal, such as the frog or turtle, but which can also be performed in mammals provided artificial respiration is maintained. The general effect of the vagus in both groups of animals is the same, although apparent differences may exist on account of the relative im- portance of the different parts of the heart in the origination and propa- gation of the heartbeat. The Cold-Blooded Heart. — If the vagus nerve on the right side in the turtle (the left nerve is inactive in this animal) is stimulated with a very feeble electric current, while simultaneous records are being taken of the contractions of the auricles and ventricles in the manner shown in the accompanying tracing (Fig. 64), it will often be found that there is a weakening of the auricular beats without any change in those of the ventricle. If the strength of stimulus is somewhat increased, the auricular beat, besides becoming weaker, will also become slower, but meanwhile the ventricular, although also slower, may become distinctly stronger. At first sight this result may be a little confusing, because it would seem to indicate that the vagus nerve weakens the auricular, 218 THE CIRCULATION OP THE BLOOD but strengthens the ventricular beat. It is clear, however, that the strengthening of the ventricular beat is merely due to the fact that the cavity has become better filled with blood during diastole as a result of the slowing of the auricle. These results indicate, then, that with weak stimulation the vagus exerts its direct influence only on the auricle. If Fig. 64. — Simultaneous tracings from auricle and ventricle of turtle's heart. Between the crosses the vagus was stimulated, with the effect that the auricular beat diminished in force but not in frequency, while the ventricular beats were practically unaffected. (From Howell's Physiology.) the stimulation is strong enough both auricles and ventricles cease to beat altogether, and if the stimulus is maintained, the inhibition may go on for a very long time (Fig. 65). Usually, even though the stimulus is maintained the heart begins to Fig. 65. — Effect of vagus stimulation on heart of turtle. Note the after effect of augmentation. beat again "after a time, at first only occasionally but gradually more rapidly. This is known as escapement, and it indicates that the energy pent up in the heart during the vagus inhibition has at last overcome the inhibiting influence of the nerve, which is meanwhile becoming fatigued. All of these results could be quite satisfactorily explained on the assumption that the action of the vagus is confined to the sinus, THE CONTROL OF THE CIRCULATION 219 which, it will be remembered, dominates the beat in the rest of the heart. There is evidence, however, that the vagus also directly affects the rhythm of the ventricle. It may be stated as a general conclusion from these results that the influence of the vagus upon the heartbeat is chiefly centered upon those parts of the organ in which the rhythmic power is most highly developed. Besides affecting the rate and strength of the heartbeat, the vagus also exercises a control on the conductivity of the cardiac muscle. Thus, if a partial block is instituted in the turtle heart by applying a clamp be- tween the auricles and ventricles, stimulation of the vagus enfeebles the auricular beat and may also cause a complete heart-block as shown in- the tracing reproduced in Fig. 66. It is important to point out here, however, that under certain conditions the vagus may appear to increase rather than decrease the conductivity of the tissue in the auriculoven- Fig. 66. — Tracing to show that vagus stimulation may diminish transmission from auricles to ventricles. It shows the effect of stimulating the left vagus on partial (2/1) block produced on heart of turtle by application of clamp at auriculoventricular junction. Stimulation at •!• depressed the conductivity and weakened the auricular contractions (lower tracing) without slowing their rate. The result was an increase in the degree of block with cessation of ventricular contractions (upper tracing). Initial auricular rate =: 35 per minute. (From Carrey.) tricular junction; for example, it has been observed in the turtle heart that when a clamp is so tight as to produce complete block — that is to say, to render the ventricle inactive while the auricle is still beating at the usual rate — stimulation of the vagus, besides causing the auricles to become distinctly slowed, may at the same time cause the ventricles to respond to the auricular beats. This result is probably due to the better chances of slow beats getting through the junction than those which are so frequent as to crowd one another on the narrow bridge which the con- stricted tissue offers to their passage (Fig. 67). Very important work was contributed in this field by G. B. Mines13 shortly before his lamentable death. He found that the local applica- tion of atropine to the sinus eliminates the effect of stimulation of the (intracranial) vagus on the rate of the heartbeat, while the effect on the 220 auriculoventricular junction and on the ventricle remains. After the atropinization, vagus stimulation delays the transmission of beat from auricle to ventricle and shortens the time of each beat in the ventricle. It was further found that by the local application of atropine various parts of the ventricle can be rendered irresponsive to the influence of the vagus and the effects of this nerve on the form of the cardiogram modified at will. These results have an important bearing in the in- terpretation of the cause of the T-wave of the electro-cardiogram which will be referred to later. Mines' results show that the proba- ble explanation is that the T-wave is due to the greater duration of the excitatory state at the base than at the apex, for by altering the relative duration of this state at base and apex by the above methods, he could cause the T-wave to appear or disappear. The direct excitability of the heart muscle to external stimuli is also depressed during vagus stimulation. This effect is, however, not evi- Fig. 67. — Tracing to show that vagus stimulation may facilitate transmission from auricles to ventricles. It shows the effect of right vagus stimulation on heart-block produced in the turtle by a clamp. Upper tracing records ventricle; lower tracing, auricles. Weak faradization of the right vagus nerve beginning at A affected the degree of block only at f1 , when a lengthened period between auricular contractions caused a single ventricular contraction. At B stronger faradiza- tion of the same nerve produced marked slowing of the auricles, in consequence of which the block disappeared and the ventricles contracted after each auricular contraction. Block reappeared when the rate again became rapid. Initial auricular rate = 36 per minute. (From Carrey.) dent in the case of all hearts, but is seen in those of certain fishes (e. g., the eel). The Mammalian Heart. — The action of the vagus on the mammalian heart may be investigated either by exposing the heart and connecting the auricles and ventricles with specially designed recording levers (myocardiograph), or if we desire to study the influence on the heart as a whole, by taking a blood-pressure tracing from one of the large arteries by means of a spring manometer. The results are in general similar to those observed on the frog or turtle heart, the main effects being de- veloped on the auricle. Considerable differences are found in the effect on the heart as a whole in different animals, particularly with regard to the facility with which escapement occurs. In the dog when the vagus THE CONTROL OF THE CIRCULATION 221 is continuously stimulated, the heart is likely to remain inhibited for a long time, whereas in the cat the inhibition is very quickly broken into by escapement. If the tracing is taken directly from the heart, it will frequently be observed in the dog that, when the escapement occurs dur- ing vagus stimulation it is only the ventricle that is beating, the auricles still remaining inhibited. If the stimulation of the vagus is discontinued after some time in an animal whose blood pressure is being recorded, the pressure will not only quickly recover, but will usually overshoot the normal level, mainly because of the asphyxia which has been produced during the period of inhibition. The asphyxia raises the hydrogen-ion concentration of the blood and this stimulates both the vasoconstrictor center and the heart action (page 168). The increased heart action is, also partly owing to the fact that during vagus inhibition the beating power of the heart becomes improved (page 225). As an outcome of recent work,14 it has been shown that the right vagus nerve acts mainly on the sinoauricular node, and the left vagus on the auriculoventricular bundle. This is in agreement with the observations described above on the cold-blooded heart (page 217). Stimulation of the right vagus ahvays causes slowing and weakening of both the auricular and the ventricular beats, but stimulation of the left vagus is sometimes observed to have but little influence on the auricular beat, although it may produce a condition of partial heart-block; or, if a clamp is ap- plied to the auriculoventricular bundle so as to produce a partial heart - block, then during stimulation of the left vagus, the block may become complete. There is, however, a considerable overlapping of these in- fluences, at least in the case of the left vagus, for this nerve also acts considerably on the ventricle, influencing perhaps not so much the rate as the force of the contraction. It has been found experimentally that, in order to demonstrate the specific action of the left vagus on the bun- dle, it is most suitable to study the relationship between auricular and ventricular beats when the auricle is beating rapidly as during the application of artificial (electrical) stimuli to it. Ordinarily the con- traction produced by each stimulus passes into the ventricle, but during stimulation of the left vagus it is found that every contraction does not pass. These experiments raise the question as to what the influence of either nerve may be in blocking impulses from the auricles to the ven- tricles when auricular fibrillation is present. It might be expected that the left vagus would prove more effectual in this regard, but actually it has been found that both vagi have the same effect. Tonic Vagus Action. — Impulses are constantly passing along the vagi to the heart. On account of this so-called tonic action, the heart rate 222 THE CIRCULATION OF THE BLOOD increases when the continuity of the vagus nerve is broken either by cutting or by freezing a portion of nerve (Fig. 26). The effect is usually inconspicuous when one nerve only is cut, but in most mammals it be- comes quite marked when both are cut. Change in the heart rate pro- duced by muscular effort is much more gradual in animals with marked vagus tone, such as hunting dogs, than in those with little vagus tone, as in domestic rabbits. The degree of vagus tone therefore bears a relation- ship to the staying power of the animal for prolonged muscular effort. It is usually ill developed in cold-blooded animals. It is quite marked in the case of man, as is evident on observing the heartbeat before and after giving a sufficient dose of atropine to paralyze the termination of the vagus in the heart. The exaet location of the nerve cells that form the center of discharg- ing impulses along the vagus fibers to the heart can not be made out with certainty, but they are no doubt part of the great motor nucleus (ambiguus), from which arise the fibers not only of the vagus but of the glossopharyngeal nerve. The tone of this vagus center is almost without doubt dependent upon the constant transmission to it along the sensory or afferent fibers of impulses coming from various portions of the body. According to the strength or number of these impulses, the tone may be increased or diminished, thus altering the rate of the heart. It is possible of course that the tone can be maintained, independently of the afferent impulses, by the action on the center of chemical meta- bolic products or hormones produced in the cells or carried to them in the blood. We know at least that, like the respiratory center, that of the vagus is excitable by such hormones as the hydrogen-ion concen- tration of the blood. The tonicity of the vagus center is, however, mainly dependent upon the passage to it of afferent impulses, and as evidence for this conclusion may be cited the observation that, after section of most of the afferent nerves to the medulla (as by cutting the spinal cord high up in the cervical region), subsequent section of the two vagi does not produce anything like the usual degree of change in the heart rate. The Afferent Vagus Impulses. — The afferent vagus impulses may come from practically any part of the body, having been first discovered by the simple experiment of tapping the abdomen of the frog with the han- dle of a scalpel, when slowing of the heart rate is observed. Cutting the vagi abolishes the reflex. Similar cardiac inhibition is produced by me- chanical stimulation of the tail or gills of an eel. In mammals stimula- tion of the central end of any sensory nerve usually slows the heart, though sometimes the opposite effect occurs. The pulmonary branches of the vagus are particularly sensitive in producing reflex inhibition, and distinct results are usually obtained: by stimulation of the termina- THE CONTROL OF THE CIRCULATION 223 tions of the fifth nerve in the mucosa of the upper respiratory passages, as by inhaling ammonia vapor; by stimulation of the sensory nerve end- ings in the pharynx, as by swallowing ; and of the mucosa of the larynx, as when a substance is "swallowed the wrong way." The sensory nerves of the abdominal viscera seem to be particularly active on the vagus center, as is seen in irritation of the sensory nerves of the stom- ach such as occurs in gastritis. Profound, inhibition may also be caused by violent stimulation of the mesentery, as from a blow on the abdo- men, or by irritation of the sensory nerves of the intestine, either me- chanical or because of disease. Another interesting illustration of affer- ent vagus stimulation is obtained by pressure on the outer canthus of the eye. This oculomotor vagus reflex, as it is called, is very marked in some individuals. Through which of these afferent paths it may be that the constant stimuli are transmitted to the vagus center to enable it to maintain its tone, can not be said, although it is very likely to be through the vis- ceral nerves. In considering the cause for an observed change in heart rate, we must of course bear in mind the possibility that the action may have occurred, not through the vagus center, but through the sympathetic center. Thus, when the heart becomes quicker, it may be owing either to diminution in the vagus tone or to an increase in the discharges along the sympathetic nerve from the augmentor center. That such reflex action through the augmentor center does occur under experi- mental conditions has been clearly shown; for example, if both vagus nerves are cut and the peripheral end of one of them stimulated mod- erately, so as to hold the heart at about its normal rate, the stimulation of certain sensory nerves may cause increase in the heart rate. Reflex sympathetic control of the heartbeat is however no doubt much less important than control through the vagus center. When it does exist it means that the actual rate of the heartbeat at any given moment must represent the algebraic sum of two opposing influences, with that of the vagus preponderating. The advantage of such a double inner- vation is that it insures prompter adjustment of the beat. If, for ex- ample, for any reason quickening of the heart rate is necessary, it is brought about most promptly if the vagus tone is diminished at the same moment that the sympathetic tone is increased. Such reciprocal action of antagonistic influences is the usual rule in the animal economy. Thus, when the knee joint flexes, it does so not merely because stimulating impulses are transmitted to the hamstring muscles, but also because at the same moment inhibiting impulses are transmitted to the extensor muscles (see page 814). 224 THE CIRCULATION OF THE BLOOD Several possibilities have to be kept in mind when we attempt to determine the exciting cause for an observed change in the heart rate in man. Thus, a slowing of the rate may be due to mechanical stimulation of the vagus trunk, as in pressure on the nerves by a tumor or aneurism in the neck. That such mechanical irritation may stimulate the vagus is easily demonstrated in many individuals by applying pressure to the vagus where it lies in the neck in front of the sixth cervical vertebra. Such pressure sometimes produces so profound an inhibition of the heart that temporary loss of consciousness occurs. It is often an unsafe ex- periment to perform. Change in the heart rate in man may be caused by direct stimulation of the vagus center, as by the presence of a tumor or a blood clot in the medulla, or by the action on the center of some unusual hormone in the blood. A general increase in intracranial pressure also stimulates the vagus center. The slowing of the heart which occurs in asphyxia might be due either to the action of hormones (hydrogen-ion concentration) in the blood as the result of the asphyxia, or to the increased intra- cranial pressure. That the latter is the more important cause is shown by the fact that, if the rise in blood pressure is prevented by connecting an artery with a mercury valve, — that is, with a tube dipping into a cylinder of mercury to a depth corresponding to the normal blood pressure, so that when the pressure tends to rise the blood escapes, — the sloAving of the heart is not observed. The excitability of the afferent vagus fibers in the lungs is greatly increased during the earlier stage of chloroform administration. Finally it should be pointed out that, although we have no voluntary control of the activity of the vagus center, its activities are subject to great variation as a result of impulses transmitted from centers higher up in the cerebrospinal axis. It is by the operation on the vagus center of such impulses that changes in heart rate occur during emotional ex- citement, fright, etc. The increased heart rate in muscular exercise is probably dependent upon a number of causes, such as the irradiation of the motor impulses on to the cardiac centers (see page 412), the rise in temperature and changes in the hydrogen-ion concentration of the blood, etc. Mechanism of Action of Vagus on the Heart. — Physiologists have nat- urally been curious as to the exact manner in which the vagus nerve brings about inhibition of heart action. Similar inhibition as a result of stimulation of efferent nerves exists in the case of the dilator fibers to the blood vessels (page 234) and the sympathetic nerve to the intes- tine (page 4.67). Inhibition of voluntary muscles can be produced only through the central nervous system by stimulation of afferent nerves Left Ant Caval vein .Ttiyht Ant Cavsl vein ^m^^^^f neuron "euron Auricular ,- ning from n Sinus to auricle [.Position of Hook from 'Heart lever vonBezold's Oanqlion in Auricular seprum /„,, , _ I 'Bidders Ganglion in aurkulo-ventricular junction JtimulaHng electrodes in sino-auricular junction [Crescent] Sympathetic fibres = dotted lines Fig. 68. — Diagram to show the innervation cf the heart in the frog or turtle. The electrodes are represented as applied to the white crescentic line where they will stimulate some postganglionic fibers. (From Jackson.) THE CONTROL OF THE CIRCULATION 225 (page 814). It is not the nerve fibers themselves that are responsible for the inhibitory effect, for it has been found that if the peripheral end of a cut vagus nerve is connected with the central end of one of the anterior roots of the cervical portion of the spinal cord, the axons of the latter when they grow down into the vagus trunk during the regeneration which follows, stimulation of the regenerated fibers will still produce inhibition of the heart. The nature of the fibers can not therefore be the factor upon which the inhibiting influence of the vagus is dependent. This leaves the terminal apparatus of the vagus fibers in the heart as the structures in which the stimulus conveyed to them is rendered inhibitory in nature. There has been considerable speculation as to what kind of change must be occurring in the heart in order to cause the inhibition, but practically nothing that is definite is known. One significant fact, how- ever, is that the electrical current led off through nonpolarizable elec- trodes from two portions of the auricle one of which is injured, does not take the same direction when the vagus nerve is stimulated as that which it takes when the motor nerve of a similarly observed muscle is stimu- lated. A positive instead of a negative variation is observed. Now, since a negative variation is always accompanied by active chemical breakdown changes occurring in the muscle to supply its energy of contraction, it is assumed that the positive variation accompanying stim- ulation of the vagus must indicate that, instead of a katabolic process, a building up, or anabolic process, is being excited. This conclusion would fit in perfectly with the well-known fact that, after the heart has been held in standstill for some time by vagus stimulation, the beats are stronger after the inhibition has passed off than they were before. The vagus seems to have a conserving influence on the heart. During the inhibition produced by it energy material is apparently stored up in the heart, so that when the beat is reestablished it is stronger than before. The Manner of Termination of the Vagus Fibers in the Heart. — This subject is of considerable pharmacologic and therefore therapeutic in- terest. In approaching the problem it must be remembered that the vagus fibers belong to the so-called cerebral autonomic system of nerves (see page 882). They are therefore fibers which have cell stations situ- ated near their peripheral termination — cell stations, that is to say, in which ganglionic medullated fibers, by forming synapses around nerve cells, become connected with postganglionic nonmedullated fibers. The existence of ganglia in the heart, particularly of the frog, has been known for a long time. These ganglia are located at the sinoauricular junction, at the interauricular septum, and in the ventricle near the 226 THE CIRCULATION OF THE BLOOD auriculoventricular junction. The function of the ganglia is to serve as cell stations on the course of the vagus nerves. (Fig. 68.) Nicotine is a drug which in certain concentrations, if applied locally to sympathetic ganglia, specifically paralyzes the synapses between the ends of the preganglionic fibers and the cells from which the post- ganglionic fibers arise. If this drug is applied in a 1 per, cent solution to the heart, stimulation of the vagus trunk no longer produces inhibi- tion, but if the stimulus is applied to a portion of the heart known as Fig. 69. — Frog heart tracing showing the action of nicotine. The vagus trunk was stimulated as indicated. In the normal (lower) tracing inhibition occurs but after nicotine (second tracing) no inhibition follows. Stimulation of the crescent in the next two lines still is followed by inhibi- tion. The final effects of the drug are' shown in the last two (upper) tracings. (From Jackson.) the white crescentic line, inhibition still occurs, because at this point the postganglionic nerve fibers come near to the surface and therefore are stimulated (Fig. 69). On the other hand, atropine is a drug which paralyzes the postganglionic fibers, so that after its application to the heart inhibition can not be produced by stimulating either the vagus trunk or the white crescentic line. Pilocarpine and muscarine are drugs which have an action exactly opposite or antagonistic to that of atro- Medulla oblong a ta N.XI Cervical l Accessory n. to trapeziu Spinal medulla — (cord) Kami communican- tes going to Symp. gang, (preganglionic) Ansa subclavia- (Annulus of Vieussens)r Thoracic— 3 nerves ^ N.I Postganglionic fibers are dotted thus -Jugular ganglion (Gang, of the root) -Depressor (Fall in pressure or slowing of heart.) (Sensory) separate nerve in rabbit and opossum. Vodosum ganglion (Gang, of the trunk) lHHaurnr^ton] -Inhibitory cranial autonomic fibers -Superior cervical ganglion -Descending sympathetic fibers in cord ^Cervical vago- sympathetic trunk ^Electrodes (slowing or stoppage of Subclavian heart. Augmentation in some animals.) -Aortic arch First thoracic qanqlion (Stellate) 5 Electrodes (Acceleration, or augmentation of heart.) Fig. 70. — Schematic representation of the innervation of the heart of the mammal. The red continuous lines represent the sympathetic (accelerator) preganglionic fibers, and the broken red lines, their postganglionic fibers. The cell stations are in the inferior cervical and stellate ganglia, some extending up to the superior cervical ganglion. The green continuous lines represent the vagus preganglionic fibers, and the broken green lines, their postganglionic fibers. The cell stations in this case are located in the heart itself. It will be observed that electrodes applied to the so- called vagus low down in the neck may stimulate some sympathetic fibers. (From Jackson.) THE CONTROL OF THE CIRCULATION 227 pine; that is, they stimulate the postganglionic fibers and produce a slowing and possibly an enfeebling of the beat. In the mammalian heart a large number of the fibers in the right vagus nerve proceed directly to the sinoauricular node, where it can be shown histologically that considerable masses of nervous tissue exist. On the other hand, the great majority of the fibers in the left vagus proceed to the auriculoventricular bundle, in which also nervous struc- tures are abundant (page 184). As already indicated, the experimental results which follow stimulation of either nerve can be explained by the influence which 'the nerve exerts on the particular structure to which the majority of its fibers proceed. In brief, stimulation of the right vagus is likely to produce slowing and weakening of the beat, whereas stimulation of the left vagus is more likely to institute a condition of partial heart-block. On account of the different results which may be obtained by stimu- lating the vagus, some authorities have assumed that the heart must contain four kinds of fiber, more strictly, of vagus nerve endings, one for each kind of influence which the vagus can develop. These four influ- ences are, it will be remembered, on the strength, the rate and the propagation of the heartbeat, and the excitability of the cardiac muscle. It is, however, almost certainly unnecessary to make such an assump- tion, for the results can be explained as merely dependent upon dif- ferent degrees of stimulation of the same kind of fiber and upon the exact part of the heart to which the fiber runs. Sometimes, for ex- ample, when the right vagus nerve is stimulated very feebly, there may be only a diminution in the force of the beats -without any change in their rate, indicating that the effect has been upon the musculature of the auricular walls and not on the sinoauricular node. If the stimulus is increased a little, then both an enfeebling and a slowing of beat occur, indicating that the stimulus has now passed both to the auricular mus- culature directly and to tlie sinoauricular node. The Sympathetic Control The effect of the sympathetic nerve on the heart may be described as being exactly opposite to that of the vagus. The pathway along which the fibers of this nerve travel to the heart is more or less a devious one. They arise in the mammal from nerve cells in the gray matter in the upper thoracic portion of the spinal cord. The fibers leave by the cor- responding spinal roots and pass by the white rami communicantes into the sympathetic chain, up wrhich they travel to the stellate and inferior cervical ganglia. Around the nerve cells of the stellate ganglion the fibers end by synapsis, and the axons of the cells are then continued on 228 THE CIRCULATION OF THE BLOOD as postganglionic fibers, proceeding to the heart through branches com- ing off from the stellate ganglion itself, or from the ansa subclavii or. inferior cervical ganglion. (Fig. 70). In cold-blooded animals, such as the frog, the sympathetic fibers run up to the upper end of the cervical sympathetic and join the vagus immediately after it leaves the cranial cavity. They then proceed along with this nerve — forming the vago- sympathetic — to the heart. The effect of stimulation is shown in Fig. 71. The sympathetic nerve differs from the vagus in that a much longer la- tent period elapses before its influence becomes effective, and this persists for a much longer period after the stimulus is withdrawn. If the vagus B. Fig. 71. — Tracings showing the effects on the heartbeat of the frog resulting from stimulation of the sympathetic nerves prior to their union with the vagus nerve. (From Brodie.) and sympathetic are stimulated at the same time, as by exciting the vago- sympathetic in the frog, the first effect observed is that of the vagus usually followed, after removal of the stimulus, by the sympathetic ef- fect. If the stimulus is maintained for a long time, so that the vagus becomes fatigued, escapement will occur earlier than with pure vagus stimulation, and augmentation may become apparent. The sympathetic influence is, however, never so strong as that of the vagus. The two nerves are therefore not antagonistic in the sense that the one neutralizes the effect of the other; but when both are stimulated, the heart responds first to the vagus and later to the sympathetic. CHAPTER XXVI THE CONTROL OF THE CIRCULATION (Cont'd) THE NERVE CONTROL OF THE PERIPHERAL RESISTANCE As already explained, the nerve control of the peripheral resistance takes place through the action of vasoconstrictor and vasodilator nerve fibers on the musculature of the arteriole walls. The vasoconstrictor impulses like those in the vagus of the heart are tonic, so that when a nerve containing such fibers is cut, the corresponding blood vessels un- dergo dilatation (see page 135), and when their peripheral ends are stim- ulated artificially, constriction occurs. On the other hand, the vasodi- lator impulses do. not appear, at least under ordinary circumstances, to be tonic, so that the cutting of such fibers does not cause vasoconstriction ; their stimulation, however, causes marked dilatation. Vasomotor fibers are contained in most of the efferent (motor) nerve trunks, and to detect their presence the nerve must be either cut or stimulated and the condition of the blood vessels of the innervated area observed. Methods for the Detection of Constriction or Dilatation Several methods, varying with the exact area under observation, can be used for the detection of vasoconstriction or dilatation. In many cases visual inspection is sufficient, as in the well-known experiment of Claude Bernard on the blood vessels.in the ear of the rabbit (see Fig. 106). When this is held with a light behind it, and the cervical sympathetic of the corresponding side is cut, marked dilatation will become evident and vessels will spring into view where previously there were none visible. Visual inspection is usually also a satisfactory method of demonstrat- ing vasodilatation or constriction in exposed glands, in mucous pas- sages and in the vessels of the skin. Another comparatively simple method is the observation of the tem- perature of the part, this being particularly useful when the vascular area is one situated in the peripheral part of the body, such as the hand or foot (see page 209). When dilatation occurs the temperature of the part rises, because the warmer blood from the viscera flows with greater freedom through the peripheral regions, where it is cooled off by radia- tion. When a thermometer is placed between the toes of a dog or cat, a 229 230 THE CIRCULATION OF THE BLOOD distinct rise in temperature will be observed when the sciatic nerve of the corresponding limb is cut. The application of this principle in deter- mining the mass movement of blood by the amount of heat given off from the hands or feet has already been explained. Other methods depend up.on observation of the outflow of blood from the veins of the part. A simple application of this method can be used in the case of the ear of the rabbit. If the tip of the ear is cut off, bleeding under ordinary circumstances is only very slight, but if the cervical sympathetic is cut, it becomes quite marked, slowing down again or even stopping entirely when the peripheral end of the nerve is stimu- lated. By making measurements of the volume of the outflow of blood from a vein by this method, the extent of constriction or dilatation can tube to recorder oil enclosed by membrane Fig. 72. — Roy's kidney oncometer. (From Jackson.) be followed quantitatively. Vasodilatation also causes changes in the character of the venous flow, the usually continuous flow becoming pul- satile and the color of the blood brightening. Comparison of the pressures in the arteries and the veins of a part is also often of value in the detec- tion of changes in the caliber of the blood vessels, for, of course, the greater the . difference in pressure between the tAvo manometers, the greater must be the resistance offered to the flow. For experimental purposes, however, the standard method is that known as the plethysmofjraphic. For this purpose the organ or tissue is enclosed in a so-called plethysmograph or volume recorder, the prin- ciple of which will be clearly seen by consultation of the accompanying diagram of one adapted for the kidney (Fig. 72). Any increase de- tected by this means in the volume of the part must be due either to THE CONTROL OF THE CIRCULATION 231 an increase in blood flowing into the vessels because of increased heart action or to a local vasodilatation; and vice versa, when shrinkage oc- curs. We can not tell from the volume tracing itself which of these changes is really responsible for the observed alteration, but we can do so by simultaneously observing the mean arterial blood pressure. If this falls when the volume decreases, it means that the volume of blood flow- ing to the part must have become diminished. If, on the other hand, the blood pressure remains constant or rises while the volume decreases, it means that the blood vessels have locally constricted. Methods for the Detection of Vasomotor Fibers in Nerve Trunks If we wish to find out through which nerve trunks a given vascular area is supplied with vasoconstrictor or vasodilator impulses, we should proceed by the use of one of the above described methods to observe the effect produced on the vessels by cutting the nerve and then by stimu- lating the peripheral end of the cut nerve. As a result of such observa- tions it has been found that the vasomotor fibers are frequently dis- tributed so that those having a vasoconstricting action are collected mainly in one nerve trunk and those having a dilating action in another; in some nerve trunks, however, the relative numbers of the opposing fibers are about equal. Nerves containing a great preponderance of vaso- constrictor fibers are the great splanchnic and the cervical sympathetic ; and those containing a great preponderance of vasodilator are the chorda tympani nerve to the submaxillary gland and the nervi erigentes to the external genitalia. It must be clearly understood that, although one kind of vasomotor fiber may preponderate in one of these nerves, yet the opposite kind is also present. In the cervical sympathetic, for example, some vasodila- tor fibers extending to the blood vessels of the mucous membrane of the nose and cheeks can readily be demonstrated, as shown by the flushing of these parts when the peripheral end of the nerve is stimulated; and similarly, even in the great splanchnic nerve itself, vasodilator fibers supplying the suprarenal capsule can quite readily be made out. When' the vasoconstrictor fibers greatly preponderate over the vasodilator, the effect of the latter may be demonstrated by taking advantage of the fact that ergotoxine paralyzes the vasoconstrictor but not the vasodilator fibers, so that after its administration stimulation of the great splanch- nic nerve gives rise to a vasodilatation instead of a vasoconstriction. The presence of vasoconstrictor fibers in the so-called vasodilator nerves (chorda tympani and nervi erigentes) has not however, been demon- strated. A good example of a nerve trunk containing about an equal admix- 232 THE CIRCULATION OF THE BLOOD ture of both kinds of vasomotor fibers is the sciatic. If the hind limb of a dog is placed in a plethysmograph and simultaneously a record of the mean arterial blood pressure taken, it will be found on cutting the sciatic nerve that the volume of the limb increases, whereas the blood pressure remains practically constant. Before placing the limb in the plethysmo- graph, the muscles must of course be paralyzed by means of curare; otherwise muscular contractions would confuse the result. If the peripheral end of the cut nerve is now stimulated, vasoconstriction will readily be observed. So far, then, the results demonstrate the presence of vasoconstrictor nerve fibers alone. To demonstrate the presence of vasodilators a different procedure is necessary. This is based on the following facts: (1) The vasodilator nerve fibers degenerate more slowly than the vasoconstrictor; (2) they are less depressed in their excitability by cooling the nerve; and (3) they are more sensitive to weak slow faradic stimulation than the vasocon- strictor fibers. Accordingly, if we cut the sciatic nerve two or three days before the actual experiment, and then, while observing the volume of the limb, proceed to stimulate the half-degenerated nerve with feeble electric stimuli of slow frequency we shall usually observe a dilatation of the limb instead of constriction; and even if we cool a stretch of a freshly cut nerve before applying the stimulus, the same result will often be obtained. The Origin of Vasomotor Nerve Fibers Having seen how the presence of vasomotor fibers may be detected in peripheral nerves, we must now proceed to trace them back to their origin from the central nervous system. The method for doing this con- sists, in general, in observing the effect on the blood vessels produced by cutting or stimulating the various nerve roots through which the fibers might pass on their way to the nerve trunks. As a result of such obser- vations it has been found that all of the vasoconstrictor fibers emanate from the spinal cord in the region between the level of the second thoracic and that of the second or third lumbar spinal roots, but from nowhere else in the cerebrospinal axis. Section of the spinal cord below the level of the second lumbar spinal roots produces no change in the volume of the hind limb, provided the muscles be thoroughly curarized, nor does stimulation of the lower end of the cut spinal cord have any effect. If the last two thoracic or the first two lumbar spinal roots are stimulated, however, evidence of vasoconstriction will be obtained. The restriction of the origin of vasoconstrictor fibers to the above- mentioned regions of the spinal cord indicates that in proceeding to the mixed nerve trunks they must travel along special nerve paths. THE CONTROL OF THE CIRCULATION 233 These are provided by the sympathetic chain and its branches (Fig. 228). The vasoconstrictor fibers in the anterior spinal roots leave the latter by way of the corresponding white rami comnmnicantes, and pass into the neighboring sympathetic chain, along which they either ascend or descend, according to their ultimate destination. In their course they come into contact with the sympathetic ganglia, through one or two of which they may pass without any change, but ultimately each fiber ar- rives at some ganglion, in which it terminates by forming a synapsis around one of the ganglionic nerve cells. The axon of this nerve cell then continues the course by the nearest gray ramus communicans back to the spinal nerve beyond the union of its anterior and posterior roots. Up to the point where the fiber forms a synapsis with a ganglionic nerve cell, it is medullated and is known as the preganglionic fiber. Beyond the nerve cell, it is nonmedullated and is known as postganglionic (page 877). The exact ganglion in which a given vasoconstrictor fiber becomes connected with a nerve cell can be determined by the nicotine method of Langley. Local application to the ganglion of a weak solution of this drug (1 per cent) paralyzes the synaptic con- nection, so that a stimulus applied to the preganglionic fiber no longer produces its effect. Suppose, for example, that a vasoconstrictor fiber has been found by the stimula- tion method to travel through several ganglia, and we wish to determine in which of these the synapsis occurs: we can do so by applying the stimulus at a point central to the ganglia after painting each of them in turn with the nicotine solution. If the application of the drug to a given ganglion is found to cause no alteration in the effect produced by stimulation, then we know that there can not be any synaptic connection in that ganglion, and we proceed in the same way till we have located , the ganglion in which synapsis occurs. It is important to remember that the post- ganglionic vasoconstrictor fibers in a gray ramus communicans do not come from the preganglionic fibers of the corresponding spinal rcot, but from fibers coming through white rami at a higher or a lower level. The above description applies to the vasoconstrictor fibers proceeding to the vessels of the anterior and posterior" extremities, those for the former arising (in the dog) from about the fourth thoracic to the tenth ; and those for the latter, from the lowest thoracic and the first three lumbar nerve roots. The cell station for the fibers to the fore limbs is in the stellate ganglion, and for the hind limbs in the last two lumbar and first two sacral ganglia of the abdominal sympathetic chain. The vasoconstrictor fibers to the vessels of the head and neck run a somewhat dif- ferent course, there being no convenient cerebro spinal nerve along which the post- ganglionic fibers may run. The fibers to the blood vessels of the head leave the cord by the second to the fourth or fifth thoracic roots and pass by the corresponding white rami communicantes into the sympathetic chain, up which they run, passing through the stellate ganglion, the ansa subclavii, and the inferior cervical ganglion, then ascending in the cervical sympathetic to the superior cervical ganglion, where their cell station exists. The postganglionic fibers on leaving this ganglion travel to their destination mainly along the outer walls of the blood vessels. The vasoconstrictors to the abdominal viscera are carried by the splanchnic nerves, the fibers of which come off from the lower seven thoracic and the uppermost lumbar 234 THE CIRCULATION OP THE BLOOD roots. The thoracic fibers pass down the sympathetic chain, which they leave by the great splanchnic nerves. The lumbar fibers form the lesser or abdominal splanchnic nerves. As preganglionic fibers, therefore, these fibers are carried by the greater and lesser splanchnic nerves into the abdomen, where the former comes into close relation- ship with the suprarenal glands, giving off a branch fco the suprarenal ganglion. The main course of the nerve is continued on to the solar plexus, in the various .ganglia of which most of the preganglionic fibers end by synapsis, the postganglionic fibers then proceeding along the blood vessels to the vessels of the abdominal viscera. (See also page 879). Vasodilator fibers have a more varied origin than vasoconstrictor, and they run an entirely different course. Vasodilator impulses may be transmitted by fibers arising from practically any level of the cerebro- spinal axis, not only by the motor roots, but by the sensory as well. Thus, they pass out of the spinal cord in the posterior sacral roots to enter the nerves of the hind limbs, as has been demonstrated by observ- ing an increase in the volume of the curarized limb during electrical stimulation of the exposed rootlets. The apparent inconsistency of these observations with the well-known law concerning the direction of the impulses contained in the posterior spinal roots is explained by assum- ing that the dilator impulses are transmitted along the ordinary sensory fibers, since th'ere are no efferent fibers in these roots. They are impul- ses which go against the ordinary stream (antidromic). In support of this explanation it is of importance to note that at their termination near the skin many sensory fibers split into several branches, some of which run to blood vessels, and others to receptor organs' (page 797). Stimulation of the latter branches may cause dilatation of the local blood vessels nearby, indicating that impulses must be transmitted up to the point at which the branching occurs and then down the vascular branch, this result being obtained even after the main trunk of the nerve has been cut above the division. For the blood vessels of the anterior extremity, the vasodilator impulses are similarly transmitted through the posterior spinal roots of the lower cervical region of the spinal cord. The vasodilator fibers to the abdominal viscera are transmitted with the splanchnic nerves, but they may also be derived from the posterior spinal roots, for it has been found that stimulation of posterior rcots in the splanchnic area causes dilatation in the intestine (Bayliss). Vasodilator fibers are also contained in the cranial nerves, par- ticularly the seventh and the ninth, being distributed in .the former nerve to the an- terior portion of the tongue and the salivary glands, and in the latter to the posterior portion of the tongue and the mucous membrane of the floor of the mouth. The vaso- dilator fibers for the mucous membrane of the inside of the cheeks and nares have their course in the cervical sympathetic, being distributed to the buccofacial region in the branches of the fifth cranial nerve. There is evidence to show that the vasodilator fibers, like the vasoconstrictor, become connected by synapsis with nerve cells somewhere in their course. In the case of the vasodilator fibers in the chorda tympani and nervi erigentes, such cell stations have been clearly demonstrated in the hilus of the submaxillary gland in the former nerve THE CONTROL OF THE CIRCULATION 235 and in the hypogastric plexus situated on the neck of the bladder in the latter. It is therefore commonly assumed that, although not recognizable by histologic methods, such terminal cell stations must also exist in close association with all blood vessels to which the vasodilator fibers run. Whether or not such peripheral cell stations exist, there is a marked difference between the course of vasodilator and of vasoconstrictor fibers. The Vasomotor Nerve Centers Our next problem is to trace these fibers farther into the central nervous system, and find the location and study the characteristics of the nerve centers from which they are derived. We must postulate the existence of both vasoconstrictor and vasodilator centers, but since there is no adequate evidence at the present time which enables us to locate the latter, we must confine our attention to the vasoconstrictor centers. These exist at two levels in the cerebrospinal axis: (1) in the gray mat- ter of the spinal cord, and (2) in the gray matter of the medulla oblongata. The spinal, or as they are often called, the subsidiary vasoconstrictor centers, are represented by certain cells of the lateral horn of gray mat- ter in the thoracic portion of the spinal cord, from which the pregan- glionic vasoconstrictor fibers above described are derived. The exact location of the nerve cells composing the chief centers in the medulla has not as yet been definitely made out; they undoubtedly lie near those of the vagus center (see Hanson). The axons of the medullary cells de- scend in the lateral columns of the spinal cord to end by synapses around the cells of the subsidiary vasoconstrictor center in the lateral horns. The experimental evidence which indicates the existence of chief and subsidiary centers is quite definite. Thus, if the spinal cord is cut at the lower cervical region (below the phrenic nuclei, so as not to interfere with the movements of the diaphragm), the arterial blood pressure falls profoundly, because the pathway connecting the two centers is broken. After several days, however, the blood pressure will gradually rise again. If after this has occurred, the spinal cord is destroyed by pushing a wire down the vertebral canal, the arterial blood pressure will again fall, indicating that the vascular tone which had been reacquired after sec- tion of the pathway between the main and the subsidiary centers must have been brought about by the development in the subsidiary centers of an independent power of reflex tonic action. This experiment there- fore demonstrates that in the intact animal the subsidiary centers do not by themselves discharge tonic impulses. In other words, the subsidiary centers ordinarily serve • merely as transfer stations for the tonic im- pulses coming from the chief center, but when these impulses no longer 236 THE CIRCULATION OP THE BLOOD arrive, then a hitherto dormant power of tonic activity becomes devel- oped in the subsidiary centers. Independent Tonicity of Blood Vessels Even after complete disconnection of the spinal cord from the blood vessels, as by cutting of the splanchnic nerve to the abdomen or abla- tion of that portion of the lower spinal cord from which the fibers to the hind limb arise, the disconnected blood vessels, although at first completely dilated, may later reacquire an independent tone of their own, indicating therefore, that they must possess some neuromuscular mechanism which can act independently of the nerve centers, and which may be stimulated to activity by the presence of hormones in the blood. The hormone was at one time thought to be cpinephrine (see page 745). Epinephrine control is indicated in the effect produced upon arterial blood pressure by stimulation of the great splanchnic nerve. Careful analysis of the curve, shown in Fig. 29, shows that the rise is both im- mediate and delayed; that is, the curve mounts immediately, then flat- tens out a little, and then assumes a further rise. This delayed response seems to depend upon the excretion of epinephrine into the blood, for it does not occur when the suprarenal veins are occluded, and is much de- layed by temporarily clamping the suprarenal veins on the same side as that on which the splanchnic nerve is stimulated. It has been stated by certain observers that, after occlusion of the adrenal veins, there is a downward tendency of the blood pressure, which however develops with extreme slowness; and that a distinct elevation of blood pressure follows the removal of a clamp temporarily placed on the adrenal veins. This rise is {^renounced if the splanchnic nerve is stimulated during the occlusion of the veins. It must of course be understood that the imme- diate rise in blood pressure following splanchnic stimulation is caused by vasoconstriction in the splanchnic area itself, as is evidenced by the fact that it does not occur, or is only very faint, when the abdominal blood vessels are ligated prior to the stimulation of the splanchnic nerve. Even after ligation of the adrenal veins and of the blood vessels of the splanchnic area, stimulation of the splanchnic nerve may still cause a slight rise in arterial blood pressure, possibly because some fibers may run from the splanchnic to vascular areas not situated within the realm of the splanchnic nerve — for example, the blood vessels of the lumbar muscles. CHAPTER XXVII The activities of the vasomotor center are controlled partly by hor- mones and partly by afferent impulses. The Hormone Control As with the respiratory center, the chief hormone, is the hydrogen-ion concentration of the blood. When this is increased, as in asphyxia, the vasoconstrictor part of the vasomotor center becomes stimulated, so that the blood vessels are constricted and the blood pressure rises. Tak- ing, as our criterion of hydrogen-ion concentration, the tension of the carbon dioxide in the blood (see page 354), we may proceed to investi- gate the relationship by observing the blood pressure during changes in the carbon-dioxide tension brought about by causing the animal to breathe atmospheres containing known percentages of the gas (Mathi- son15). Thus, if a decerebrate cat is made to respire an atmosphere containing 5 per cent or more of carbon dioxide, an immediate rise occurs in the arterial blood pressure. That the inhaled carbon dioxide acts by raising the hydrogen-ion concentration of the blood is indicated by the fact that a similar rise in blood pressure can be obtained by intra- venous injection of a weak solution of lactic acid (2 c.c. N/15) in a de- cerebrate cat. Instead of injecting the lactic acid, we may cause it to be produced in the muscles of the animal itself by greatly diminishing their oxygen supply. When a decerebrate cat, for example, is made to breathe an atmosphere of almost pure nitrogen, there is, after a latent period of about 30 seconds, a sudden rise in. arterial pressure. The existence of this latent period in the latter case, as compared with its absence when carbon dioxide is inspired, is owing to the time taken for lactic acid to be produced in the muscles on account of the oxygen deprivation. It is important to note in the above experiment that decerebrate animals are employed so as to avoid the necessity of using anesthesia, under which the results are much less definite. The fact that oxygen depriva- tion causes excitation of the vasoconstrictor center has been known for 237 238 THE CIRCULATION OP THE BLOOD some time, but the explanation that has usually been given has been that it is due to a direct effect of oxygen want on the center. The sensitivity of the medullary center towards the hydrogen-ion is many times greater than that of the subsidiary centers in the spinal cord. If an animal is kept alive by artificial respiration for some time after cutting the cervical spinal cord, the subsidiary vasomotor centers will, as we have seen, gradually acquire a tonic action, and the lowered blood pressure will gradually rise again. If, when this has been attained, the animal is made to breathe an atmosphere rich in carbon dioxide, a sud- den rise in blood pressure will occur, but to produce it a very * much greater percentage of this gas must be inspired than when the pathway ••; •* * * *;\. »• \: -•* 5 *Ill between the chief and subsidiary centers is intact.' -Whereas 5 per'ceiit carbon dioxide is sufficient to cause a rise of pressure in an animal hav- ing its chief vasomotor center, it takes 25 per cent and upward to pro- duce a like effect on a spinal animal; and similarly, although 2 c.c. of N/15 lactic acid will stimulate the chief vasomotor center, it takes 5 c.c. of N/2 to excite the spinal-cord centers. The Nerve Control However important hormones may be in maintaining a tonic stimula- tion of the center, the more sudden changes in activity are mainly brought about by afferent nerve impulses. The afferent impulses are of two classes: (1) those causing a rise in blood pressure, called pressor, and (2) those causing a fall in blood pressure, called depressor. The effect produced on the arterial blood pressure by stimulation of either pressor or depressor fibers is usually more or less evanescent, especially in the case of the depressor fibers; and when the change fol- lowing stimulation of the nerve passes off, the blood pressure always returns to its former level. This indicates that the afferent impulses do not affect the tonic control which the vasomotor center exercises on the blood vessels. It has, therefore, been assumed by Porter16 that there are really tAvo kinds of vasomotor centers: one concerned merely in the bringing about of temporary reflex changes, the other concerned in the maintenance of the vascular tone. It may be that the activities of the former are primarily dependent upon afferent impulses, and the latter, upon hormones. Justification for this view has been found in observa- tions made on the effects of stimulation of pressor and depressor fibers in animals under the influence of curare or alcohol. With the former drug, stimulation of a nerve containing a preponderance of pressor or depressor fibers produces double its usual effect, but the mean level of the blood pressure apart from this effect remains unchanged. With the latter drug (alcohol), on the other hand, the reflex response entirely THE CONTROL OP THE CIRCULATION 239 disappears, although it immediately reappears when the alcohol effect has passed off, and there is no evidence of a change in tone. The tonic and the reflex mechanisms of the vasomotor center can not therefore be identical. At the present stage of our knowledge, it is only possible for us to study the effect of stimulation of pressor and depressor fibers on the vasoreflex center. Such fibers are contained in practically every sen- sory nerve of the body, and it would appear that a fairly equal mixture of both kinds of fiber exists in most of these nerves. Pressor and Depressor Impulses. — Depressor impulses are alone present in the cardiac depressor nerve. Sometimes as in the rabbit, this exists as an independent nerve trunk, originating by two branches, one from the superior laryngeal, the other from the vagus, and descending close to Fig. 73. — Fall of blood pressure from excitation of the depressor nerve. The drum was stopped in the middle of the curve and the excitation maintained for seventeen minutes. The line of zero pressure should be 30 mm. lower than here shown. (From Bayliss.) the vagus trunk, to end around the arch of the aorta. In other animals the depressor is bound up with the vagus trunk from which it can some- times be separated by careful dissection. The first prerequisite in inves- tigating the cause of the changes produced by stimulation of these nerves is the elimination of any chance of an alteration in heartbeat as a result of simultaneous stimulation of afferent vagus fibers. This may be done either by cutting both vagi or by administering atropine. Stimulation of the central end of the cardiac depressor nerve in such an animal causes an immediate fall in blood pressure, accompanied by an increase in volume which can be demonstrated either in the hind limb or in one of the abdominal viscera — evidence of general vasodilatation (Fig. 73). When the central end of a sensory nerve, such as the sciatic, is acted on by a stimulus of moderate strength, it will usually be found that the arterial blood pressure rises and that the volume of the limb or of some 240 THE CIRCULATION OF THE BLOOD - abdominal viscus becomes diminished — evidence of general vasoconstric- tion. But when the sensory nerve is stimulated with extremely weak faradic shocks, an entirely different result is likely to be obtained; namely, a fall of blood pressure and an increase in volume of the limb or viscus, indicating that in this manner we have stimulated depressor fibers. By careful experimentation with quantitatively graduated elec- trical stimuli, it has been found by Martin and others17 that on stimu- lating an afferent nerve with weak shocks, a fall in blood pressure is the first effect to be observed, and that this becomes more and more marked as the strength of the stimuli is increased, until a certain opti- mum is reached, after which the fall in blood pressure becomes less evi- dent. When a certain strength of stimulation is exceeded, a rise instead of a fall occurs. After this point additional increase in stimulation causes more and more marked elevation of blood pressure through a very long range of stimuli. Stimulation of two afferent nerves at the same time usually produces a greater reflex vasomotor change than the stimulation with an equiva- lent strength of current of either nerve alone. That is to say, the effect produced by stimulating the central end of both sciatics simultaneously will be greater than that produced by stimulating either alone with double the strength of stimulus. As has been stated above, the reflex change in blood pressure is often quite transitory in nature, although the stimulation of the pressor nerve is maintained. When this decline has occurred, the pressor reaction can often be renewed by shifting the stimulation to a second nerve. These facts concerning the greater efficacy of combined stimulation of several nerves are of considerable importance in connection with the general question of reflex changes in blood pressure. For instance, many of the pressor fibers found in the sciatic nerve are connected with the receptors that mediate the sensations of the skin. When these receptors are stimulated, as by heat or cold, reflex changes in blood pressure occur (pressor reaction), (Fig. 74), and it is important to remember that localized stimulation of the skin is less efficient in bringing about such vascular changes than stimulation applied over large areas, even when the local stimulus is intense and the general stimulus mild in character. Jumping into a moderately cold bath will cause a much greater rise in arterial blood pressure than plunging the hand into ice cold water. Mechanism of Action of Pressor and Depressor Impulses. — When we consider the exact mechanism by which these afferent impulses operate, we have to bear in mind four possibilities: the reflex fall produced by stimulation of a depressor afferent fiber may be due either to a stimula- tion of the vasodilator part of the center or to an inhibition of the tone THE CONTROL OF THE CIRCULATION 241 cf the vasoconstrictor part; and, conversely, a rise in arterial pressure caused by vasoconstriction may be dependent either on a stimulation of the vasoconstrictor part of the center or on an inhibition of the tone of the vasodilator part. All of these changes have, as a matter of fact, been shown to occur, at least under certain conditions, although the evidence Fig. 74. — The effect of strong stimulation (heat) of the skin of the foot on the arterial blood pressure and respiratory movements. Upper tracing, thoracic movement; lower tracing, arterial blood pressure. for the inhibition of dilator tone is as yet a little uncertain (see Pig. 75). Without going into the subject in detail, we may nevertheless take as an example of the methods by which the information has been ob- tained, the experiment performed by Bayliss,18 showing that the vasodi- lation which results from stimulation of the depressor nerve is owing partly to removal of vasoconstrictor tone and partly to vasodilator 242 THE CIRCULATION OF THE BLOOD stimulation. The volume of the hind limb of a curarized and vagotomized rabbit increases when the central end of the cardiac depressor nerve is stimulated. In order to determine whether this dilatation is due solely to the removal of vasoconstrictor tone, the above experiment was repeated on a rabbit in which the sympathetic chain had been cut below the level of the second lumbar spinal roots. By such an operation all the vaso- constrictor fibers to the vessels of the hind limb are severed, but the vasodilator fibers, since they emanate through the sacral sensory roots, are left intact. It was nevertheless found on stimulating the depressor nerve that dilatation of the hind limb still occurred, thus indicating Fig. 75. — Diagram showing the probable arrangements of the vasomotor reflexes. A. Muscle of arteriole. D. Vasodilator nerve fiber terminating on A and inhibiting its natural tonus, as indicated by - sign. D, Vasoconstrictor fiber also ending in A, but exciting it (+). These two kinds of fiber arise from the dilator center (DC) and the constrictor center (CC) respectively. F. Afferent depressor fiber, dividing into two branches, one of which (-) inhibits the con- strictor center, while the other (+) excites the dilator center causing dilatation of the arteriole and fall of blood pressure. R. Pressor fiber exciting CC and inhibiting DC, and therefore causing vasoconstriction and rise of blood pressure. a, b, c, and d represent the synapses of the pressor and depressor branches with the efferent neurons. (From Bayliss.) that stimulation through vasodilator fibers must have taken place. Con- versely, in another experiment, instead of the sympathetic chain, the spinal cord was cut below the level of the second lumbar segment, thus THE CONTROL OF THE CIRCULATION 243 severing the dilator but not the constrictor path, and again depressor stimulation caused the volume of the limb to increase, indicating that an inhibition of constrictor tone must have occurred. Reciprocal Innervation of Vascular Areas It must not be imagined that changes in the caliber of the blood ves- sels occurring in one vascular area are necessarily occurring all over the body. On the contrary, a most important reciprocal relationship exists in the blood supply to different parts. After food is taken, for example, more blood is required by the digestive organs than when they are at rest, and this is insured by dilatation of their own vessels along with reciprocal constriction of those of other parts of the body. On account of the relatively great capacity of the abdominal vessels, their dilatation during digestive activity is usually greater than the reciprocal constriction of the other vessels, so that the diastolic blood pressure falls, necessitating a more powerful cardiac discharge in order to maintain the mean pressure. After taking food, the systolic pressure does not as a rule fall so much as the diastolic, if it falls at all; and the pres- sure pulse therefore becomes greater and causes a greater live load to be applied to the vessels with each heartbeat. During the sudden strain that is thrown on them, weakened arteries may give way, especially in the brain. Another example of reciprocal action of the vascular system is seen in muscular exercise. The vessels of the active muscles dilate, while those elsewhere constrict. The local dilatation in this case is, however, not entirely at least a nervous phenomenon, being caused in fact, as we shall see, by hormone action on account of the local increase in hydro- gen-ion concentration (see page 414). There can be little doubt that local irritants to the surface of the body, such as hot applications, lini- ments, etc., act in the same way; they cause local dilatation of the super- ficial and perhaps of the immediately underlying vessels and constric- tion of those elsewhere in the body. Application of cold to local areas of skin similarly causes local constriction accompanied by reciprocal dilatation elsewhere. This action of cold is very marked in some parts of the body, such as the hands, where by Stewart's method (page 283) it can be shown, not only that the bloodflow of the hand to which the cold is applied is greatly curtailed, but also that of the opposite side. Experimental demonstration of reciprocal vascular innervation is fur- nished by numerous experiments. If the central end of the great auric- ular nerve of the ear is stimulated in a rabbit, a blanching of the ves sels of the ear occurs at the same time as a rise in arterial blood pres- 244 THE CIRCULATION OF THE BLOOD sure (Loven reflex). Similarly when the central end of one of the sen- sory roots of the leg of a dog is stimulated, there is a rise in arterial blood pressure and an increase in the volume of the limb. THE INFLUENCE OF GRAVITY ON THE CIRCULATION If the arterial blood pressure is measured in the arm and leg in a man standing erect, a difference corresponding to the hydrostatic effect of gravity will be found between the two readings. In comparison with the high pressure normally existing in the arteries, this difference is, however, of little significance. On the other hand, in the veins, where the average pressure is low, gravity would cause serious embarrassment to the circulation of blood were it not for the valves and the forces which move the blood beyond them (page 214). In erect animals the part of the circulation in which blood might stag- nate as a result of gravity is the splanchnic area. Were such stagna- tion to occur, the blood would not be returned to the right heart, so that the arteries would not receive sufficient blood to maintain an ade- quate circulation, particularly in the vessels of the brain. Simple experiments devised by Leonard Hill19' 2S illustrate these prin- ciples. When a snake, for example, is pinned out on a long piece of wood and an opening made opposite the heart, this organ can be seen to fill adequately with blood as long as the animal is maintained in the horizontal position. When placed vertically, however, the heart be- comes bloodless. If now the tail end of the animal is placed in a cylinder of water so as to overcome the effect of gravity, the heart will be seen to fill again with blood. Evidently in such an animal there is no mechan- ism to compensate for gravity. If a domestic rabbit with a large pendulous abdomen is held in the vertical tail-down position, stagnation of blood in the splanchnic ves- sels occurs to such an extent that in from fifteen to twenty minutes the animal dies from1 cerebral anemia. If an abdominal binder is first of all applied, the vertical position will not have the same consequences. This experiment illustrates clearly the possible evil effects that gravity may produce in animals in which no mechanism exists to compensate for it. Placing an animal such as a dog under light ether anesthesia in the vertical tail-down position produces an immediate fall in arterial blood pressure, as shown in the tracing (Fig. 76), followed by a certain de- gree of compensation even while the animal is still in the erect position. The extent to which this compensation occurs varies with the depth of the anesthesia. If the experiment is repeated after administering a large dose of chloroform, not only will the initial fall be much greater, but THE CONTROL OF THE CIRCULATION 245 subsequent compensation will be practically absent. The application of these facts in the operating room will be self-evident. Leonard Hill has shown that three factors are involved in the com- pensating mechanism: (1) the tonicity of the abdominal musculature; Fig. 76. — Aortic blood pressure, showing the effect of posture: A, vertical, head-up; B, hori zontal; C, vertical, head-down; D, horizontal. (L.H.) (2) the tone of the splanchnic blood vessels; (3) the pumping action of the respiratory movements. The importance of the first-mentioned fac- tor can be readily shown by making a crucial incision of the abdom- inal walls in an animal in the erect position (Fig. 77), and that of Fig. 77.— Tracing to show the effect of gravity on the arterial blood pressure. At A, the animal was placed in the vertical position; at B, the abdomen was compressed; at C, a crucial incision was made in the abdomen; at D, the pleural cavity was opened; at F, the animal was returned to the horizontal position. (From Leonard Hill.) the second factor by cutting the great splanchnic nerves, or the spinal cord. After such an operation, even while in the horizontal position, as we have seen, the blood pressure falls to a considerable extent. If the animal is now placed in the vertical tail-down position, however, it falls 246 THE CIRCULATION OF THE BLOOD to the zero line and the animal soon dies (Fig. 78). The influence of the third factor is not so great as of the other two, but can be shown by the increased respiratory activity which is likely to develop in the vertical Fig. 78. — The effect of gravity on the aortic pressure after division of the spinal cord in the upper dorsal region. By placing the animal in the vertical feet-down posture, the pressure fell almost to zero, but on returning it to the horizontal posture, the circulation was restored. (From Leonard Hill.) tail-down position, the anemic condition of the respiratory center being no doubt the cause of the increased respiration. CHAPTER XXVIII PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA Up to the present we have been considering the circulation of the blood from a general point of view. There are certain organs and tissues, how- ever, in which the general mechanism is altered in order to meet pecu- liar requirements of blood supply. Thus, it is evident that the brain, incased as it is in the rigid cranium, will be unable to contract and expand as a result of vasoconstriction or vasodilation. On the other hand, we know that the blood supply to this organ does vary con- siderably from time to time. What is the nature of the mechanism by which such changes are brought about? In the case of the liver the cir- culation is peculiar on account of the fact that blood is carried to the organ by two vessels, in one of which it is supplied under high pressure and in the other, under low pressure. We must investigate the rela- tionship of these two sources of blood supply. The circulation through the coronary and pulmonary vessels must likewise receive special atten- tion on account of the highly specialized functions of these organs. THE CIRCULATION IN THE BRAIN Anatomic Peculiarities Serious curtailment of the blood supply to the brain is guarded against by the existence of the circle of Willis. Besides the four main arteries — the vertebrals and the two carotids — the spinal arteries contribute to the blood supply of the circle, and consequently in certain animals, such as the dog, the four main arteries may be ligated without causing death. In man, however, ligation of both carotids is usually fatal. The free anastomosis displayed in the circle of Willis is not maintained in the case of the arteries which run from it to supply the bratti structure. On the contrary, these vessels are more or less terminal in character; that is to say, the capillary system produced by the different vessels does not freely anastomose, so that the obstruction of one vessel, or an important branch, is followed by death of the supplied area. The vessels which go to the pia mater, however, break up into numerous smaller branches, which freely anastomose before entering the brain tissue. 247 248 The venous blood is collected by the small, very thin-walled and valve- less cerebral veins. These run together to form larger veins dis- charging into the sinuses, the openings into which are kept patent by the arrangement of dura mater around the orifices. The sinuses exist between the dura and skull and are so constructed that they can not be compressed, particularly those at the base of the brain. From them the blood is conveyed mainly to the internal jugular vein, some of it however escaping by the anastomoses existing between the cavernous sinus and the opththalmic veins, and by the venous plexus of the spinal cord. The most striking peculiarities of the veins are their patulous con- dition and the absence of valves, so that any change in the blood pres- sure in the internal jugular vein must be immediately reflected in that of the venous sinuses. This explains why compression of the abdomen Fig. 79. — Schema to show the relations of the Pacchionian bodies to the sinuses: d, d, Folds of the dura mater, inclosing a sinus between them; v.b., the blood in the sinus; a, the arachnoidal membrane; p, the pia mater; Pa., the Pacchionian body as a projection of the arachnoid into the blood sinus. (From Howell's Physiology.) causes venous blood to flow from an opening made in the longitudinal sinus. In considering the cerebral circulation, another factor that must be borne in mind is the presence of cerebrospinal fluid. This is contained in the subarachnoid spaces of the brain and spinal cord, these spaces, in the case of the brain, being often considerably enlarged to form the cisternse. The cerebrospinal fluid is also present in the ventricles of the brain, which it will be remembered communicate with the subarachnoid spaces through the foramen of Magendie, etc. It is unlikely that the cerebrospinal fluid is of much importance in connection with the control •of the blood supply to the brain tissue. It may be merely a lubricating fluid; at least it is so small in amount (60 to 80 e.c. in man) as to be apparently of little value in bringing about an alteration in brain volume. PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA 249 Although normally so scanty, its secretion can become remarkably stim- ulated under certain conditions as in fractures of the base of the skull. Under these conditions in man, it may drain away at the rate of about 200 c.c. a day or more. The fluid is apparently secreted from the choroid plexus, for when the pathways by which the ventricles communicate with the subarach- noid space are obstructed, it collects in the ventricles, producing internal hydrocephalus. Under certain conditions its absorption is also very rapid, as shown experimentally by the rapidity with which physiologic saline is absorbed when it is injected into the subarachnoid space. This absorption is believed to occur through the Pacchionian bodies, which are minute sac-like protrusions of the arachnoid into the interior of a venous sinus. The membrane that separates blood and cerebrospinal fluid is extremely thin at these places (Fig. 79). Physical Conditions of Circulation On account of these anatomic peculiarities, the physical factors con- trolling the circulation of blood to the brain are considerably different from those obtaining in any other part of the body, with the possible exception of the bones. In other vascular areas, we have seen that, when dilatation or constriction of the vessels occurs, a marked increase or diminution of the volume of the part becomes evident. Such a change in volume is evidently impossible in the case of the brain because of the rigid cranium in which it is contained. In fact, from a physical point of view we must consider the blood vessels of the brain as pro- jecting into a rigid case filled with incompressible material. Under these conditions it is obvious that the vessels as a whole could neither contract nor dilate without some increase or decrease in the volume of the contents of the cranial cavity (Leonard Hill19). Some have thought that the cerebrospinal fluid as it flows into or out of the spinal cord might accomplish this alteration in the cranial con- tents, but the relatively small amount of available cerebrospinal fluid, the smallness of the openings between the brain and the spinal cord, and the lack of experimental evidence that such changes in Volume of cere- brospinal fluid in the spinal cord do actually occur, all stand in contra- diction to such a view. However, although the vessels as a whole might not contract or expand, yet some vessels, like the arteries, might con- tract simultaneously with a corresponding dilatation of other vessels, such as the smaller cerebral veins. In admitting the possibility of some reciprocal relationship between arteries and veins, we must remember that it is only before the well-protected sinuses are reached that a change in the caliber of the veins would be possible. But it is difficult 250 THE CIRCULATION OF THE BLOOD to see how such reciprocal dilatation and constriction could be of any advantage except perhaps in causing certain areas to receive more blood than others. A reciprocal relationship might also exist between adjacent arterioles as well as between arterioles and veins; when, for example, the arm center becomes active, it is conceivable that its arterioles might dilate at the same moment that those of a neighboring, less active center become constricted. Alterations obviously might oc- cur without causing any perceptible change either in the volume of the brain as a whole or in the condition of venous flow. In consideration of these factors, most observers are agreed that the total volume of blood in the brain must be constant at all times (Monro and Kellie doctrine). Alteration of blood supply can, however, still be brought about by changes in the velocity with which the blood traverses the vessels. When more blood is required in the brain to supply the increased metabolism which we must presume accompanies heightened mental activity, it is not accomplished as in other parts of the body by an increase in the capacity of the vessels as compared with those of other vascular areas, but by a hurrying up of the circulation through vessels whose caliber remains unaltered. The main factors determining the velocity of bloodflow through the brain must, therefore, be dependent upon changes occurring elsewhere in the vascular system, a conclusion for which there is abundant experi- mental evidence. Of the many ingenious methods that have been de- vised to secure this evidence, we will cite but one in this place. Records are taken of changes in: (1) the venous blood pressure of the brain by connecting a cannula either with the vein immediately after leaving the skull or, better still, with the torcular Herophili;. (2) the brain volume, by connecting a very sensitive receiving tambour with a trephine hole in the cranium so that its open end lies against the pia mater.* Al- though, as we have seen, while incased in the rigid cranium the brain volume can not change to any degree, yet this will occur when a portion of the cranium is removed, so that pulsations correspond- ing to those in the blood vessels will be observed; (3) the circula- tory conditions elsewhere in the body, by taking arterial and venous pressures and plethysmograms. The results in a normal an- imal show the following points (see Fig. 80): (1) The tracings of the arterial blood pressure (A}, the brain volume (C) and the intra- cranial venous pressure (C) have exactly the same contour — that is, the respiratory and the cardiac waves in all three of them are identical. The venous blood as it flows into the jugular veins , also pulsates in *This receiving tambour really consists of a brass tube of the same diameter as the trephine hole, into which it is tightly fitted. The brass tube is closed at its inner end by thin rubber membrane, and its outer end is connected with the receiving tambour. PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA 251 unison with the artery. (2) Any change in the blood pressure of the systemic venous system is immediately reflected in the blood pressure of the sinuses of the brain and in the brain volume, (not well shown in accompanying tracing). (3) A change never occurs in the vessels of the brain which can not be accounted for by some change occurring Fig. 80. — To show simultaneous records of the arterial blood pressure (A), the venous pres- sure (B), the intracranial pressure (C), the pressure in the venous sinuses (D). The fall in ar- terial pressure produced by stimulation of the cerebral end of the vagus will be found to cause a fall of intracranial and cerebral venous pressure, accompanying that in the arteries, but a rise in that of the venous system. (From Leonard Hill.) elsewhere in the vascular system outside the cranial cavity. This re- sult is important because it shows that there can not be vasomotor nerve control of the brain vessels. Taking into consideration not only the results of such experiments, but also the peculiar physical conditions existing in the cranial cavity, 252 THE CIRCULATION OF THE BLOOD we must conclude that changes in blood supply depend on changes in the velocity of the bloodflow, and that such alterations in velocity are dependent upon changes occurring in the aortic and more especially in the vena-cava pressure. When the aortic pressure rises, more blood will flow into the cerebral arteries and move along them at an increased velocity, the increased pressure probably causing a moderate degree of passive dilatation, to allow extra room for which the numerous small cerebral veins become compressed. This compression of the veins probably does not obstruct the greater flow of blood through them, be- cause, taken as a whole, they are ordinarily much more capacious than need be. On the other hand, if the aortic pressure should remain con- stant, but that in the vena cava increase, then there would be obstruc- tion to the passage of blood in the intracranial arteries, and conse- quently a diminished velocity of flow. Vasomotor Nerves It might be inferred that, since the bloodflow through the cerebral vessels is mainly dependent on vascular conditions elsewhere in the body, there would be no need, as in the vessels of other vascular areas, for vasomotor fibers. Histologists have, however, discovered the pres- ence of such fibers, and it has become necessary for the physiologist to find out if they are really of importance in .connection with the regula- tion of the blood supply to the brain. Even if it is admitted that the arterioles could not contract or expand as a whole without producing local changes in venous pressure or cranial volume, it is yet of course always possible, as has already been pointed, out, that one set of arte- rioles might contract at the same moment that another set expanded. That the vessels can undergo a process of constriction has been shown by experiments in which the volume of outflow from the vessels of the brain was measured in perfused preparations of brain. When epinephrine was added to the perfusion fluid, curtailment of outflow was observed to occur (Wiggers). Since this drug causes constriction of vessels only when these are supplied with constrictor fibers (see page 736), the conclusion may be drawn that the cerebral blood vessels do contain such nerve fibers. Nevertheless, the local vasomotor control of the cerebral blood vessels can not have the significance in connection with changes in blood supply that it has for other vascular areas (Hill and Macleod20). No doubt nerve fibers are present in the cerebral blood vessels, and presumably under certain conditions they are capable of causing the blood vessels to undergo alterations in caliber, but it is impossible to see of what real value this can ~be under normal conditions. PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA 253 Intracranial Pressure One word more with regard to what is known as intracranial pressure, that is, the pressure in the space between the skull and the brain. Under ordinary conditions it must be equal to that in the cerebral capil- laries, and may be measured by connecting a sensitive manometer with a tube screwed into the cranium as described above. It has been found to vary from 0 mm. Hg in a man standing erect to 50-60 mm. Hg in a dog poisoned by strychnine. It becomes increased, not only by com- pression of the veins of the neck and by an increase in general arterial pressure, but also in pathologic conditions, such as hydrocephalus. A new growth in the brain, if it occupies more space than the tissue which is destroyed, exerts pressure on all parts. of that region of the cranial cavity, but this pressure may not be transmitted equally throughout the cranial contents, for the falciform ligaments and the tentorium sup- port a part of it, thus directing the spread of pressure along certain pathways. The structures at the base of the brain, the optic nerves, the veins of Galen and the Sylvian aqueduct are most affected in this way. If the pressure is rapidly applied, however, it may rise through- out the cranial contents. In such cases the pressure is, of course, cir- culatory in origin, since immediately after death from cerebral tumor the intracranial pressure is not found to be raised. The major symptoms of cerebral compression are no doubt due to anemia of the medulla oblongata, which may be the result either of pressure applied locally in the bulbar region, where the presence of a very small foreign body or only trivial tumor formation is sufficient to destroy life, or of pressure transmitted from the cerebral cavity, in which case, on account of the support offered by the tentorium, a much larger growth is required to affect the medulla. Internal hydrocephalus produced by blocking of the aqueduct of Sylvius and the veins of Galen causes the greatest rise in intracranial tension, and may affect the me- dulla, because the brain is driven downwards so as to pinch the bulb against the occipital bone. .It must be emphasized that it is not the pressure per se that causes the symptoms, but the attendant anemia, the symptoms of acute cerebral anemia and of compression being iden- tical (Leonard Hill19). To relieve the compression, trephining is the common practice. The trephine hole should be as large and as near to the source of compression (tumor, etc.) as possible. CIRCULATION THROUGH THE LUNGS The pulmonary or lesser circulation, as it is called, is quite different from the systemic circulation. In the first place, because the pressure 254 THE CIRCULATION OP THE BLOOD in the pulmonary arteries does not amount to more than about 20 mm. Hg, or about one-sixth of that of the systemic arteries, the peripheral resistance in the blood vessels of the lungs is much less than that of the body in general. This lower resistance is owing partly to the large diameter of the arterioles and the small amount of muscular fibers in their walls, and partly to the fact that the capillaries are held con- stantly in a somewhat dilated condition on account of the subatmos- pheric pressure in the thorax (see page 306). Another peculiarity of the pulmonary circulation is that the caliber of the vessels is to a very large extent dependent upon the changes that occur in the intrathoracic pressure with each inspiration and ex- piration. They become dilated on inspiration and contracted on ex- piration. The extent to which these respiratory changes affect the amount of blood contained in the lungs, is very considerable. At the height of inspiration it is computed that a little more than eight per cent of the whole blood in the body is contained in the lungs, whereas on expiration it diminishes to between five and seven per cent. A third peculiarity is that the pulmonic blood vessels are not sup- plied with vasomotor nerve fibers — at least with such as can readily be demonstrated. It is said that, when the pulmonary vessels are per- fused and the outflow measured, a diminution in the latter is found to occur when epinephrine is added to the injection fluid — a result which is, however, denied by certain investigators. Changes in the bloodflow have not been observed to occur when the vagus or sympathetic nerve fibers running to the lungs are stimulated. In short, the conclusion which we must draw is much the same as that for the blood vessels of the brain — namely, that although, as a result of the epinephrine ex- periment, we must admit that a vasomotor supply may possibly be present, yet it is one which can be of no significance under normal conditions. When there is obstruction to the outflow of blood from the' left ven- tricle, as, for example, in cases of high aortic pressure, the blood is not entirely discharged with each beat of the left ventricle, and therefore dams back through the left auricle into the lungs. On account of the marked distensibility of the pulmonary capillaries, a large amount of this blood may collect there and thus make the lungs serve as a kind of reservoir of the heart. When the capacity of this reservoir has, how- ever, been overstepped, an increased peripheral resistance will come to be" offered to the movement of blood in the pulmonary arteries, the pressure in which will consequently rise and sooner or later interfere with the discharge from the right ventricle, causing as a result a stag- nation of blood in the systemic veins, and a consequent increase in vol- PECULIARITIES OF BLOOD SUPPLY IN CERTAIN VISCERA 255 ume of such viscera as the liver and kidneys. The same changes will obviously also supervene when there is regurgitation of blood from the left ventricle to the left auricle, as in cases of mitral insufficiency. CIRCULATION THROUGH THE LIVER .The liver is the only gland in the body receiving both venous and arterial blood, the former being supplied to it at a very low pressure by way of the capacious portal vein, and the latter at very high pressure by the strikingly narrow hepatic artery. Except for the relatively small amount of blood which is supplied to the walls of the blood vessels and the biliary ducts, none of the hepatic artery blood mixes with that of the portal vein until the vessels enter the hepatic lobules. Beyond this point the two blood streams mix and the combined stream is drained away by the sublobular and hepatic veins. Methods of Investigation To study the relative importance of these two sources of blood sup- ply, and also to investigate the manner in which the latter is controlled, the most satisfactory method has consisted in measurements of changes in volume flow rather than in those of changes in pressure. The vol- ume-flow measurement has been made either by connecting stromuhrs (page 207) to the hepatic artery or portal vein, or by measuring the out- flow of blood from the hepatic vein into the vena cava, first with both inflow vessels intact, and then with one of them ligated. An objec- tion to the first (the stromuhr) method is the possible interference with bloodflow or blood pressure produced by inserting the stromuhr into the entering vessels, and also the fact that simultaneous measurement of the flow in both vessels can not be made satisfactorily. To measure the outflow from the hepatic veins, the aorta is ligated below the celiac axis and a wide cannula is inserted into the central end of the vena cava below the level of the liver, a loose thread being placed around this vessel just above the diaphragm. By pulling on this thread the vena cava becomes obliterated, and the blood from the hepatic veins is therefore diverted into the cannula, through which it flows into one end of a vessel shaped somewhat like a sputum cup (the receiver), the other end being connected by tubing with a piston re- corder, from the movement of which the volume of blood flowing into the receiver can readily be computed. To measure the flow of blood, a clip on the tube of the receiver is removed af the same moment that the thread around the vena cava above the diaphragm is tightened, and when the receiver has filled with blood, this thread is again loosened 256 THE CIRCULATION OP THE BLOOD and the receiver tilted up so that the blood flows at low pressure back into the circulation. The receiver being of known capacity, the length of time it takes the blood to fill it as determined by the piston recorder, furnishes us with the necessary data from which to calculate the rate of flow. The receiver is chosen of such a size that it takes only a few seconds to fill, the diversion of blood into it not causing any material fall in arterial pressure. The observations are repeated frequently. Results. — By the use of these methods it has been found that the total mass movement of blood to the liver of the dog varies between 1.46 and 2.40 c.c. per second for 100 grams of liver. Considerable changes may occur in the arterial pressure without affecting the liver flow. When the hepatic artery is occluded, the flow diminishes by about 30 per cent, or conversely, when the portal vein is obstructed but the hepatic artery left intact, by about 60 per cent, indicating that about one-third of the total bloodflow through the liver is contributed by the hepatic artery and two-thirds by the portal vein. Some blood, however, gains the liver through anastomotic channels between it and the diaphrag- matic veins. The relative supply by the two vessels is subject to various condi- tions. That through the hepatic artery, for example, may be very con- siderably altered on account of vasoconstriction in this vessel, for its walls can easily be shown to be liberally supplied with vasoconstrictor fibers carried by the hepatic plexus. This can be demonstrated by the rise in blood pressure which occurs in a branch of the hepatic artery during stimulation of the plexus. On the other hand, alterations in the bloodflow in the portal vein can not be brought about by active con- striction or dilatation of the intrahepatic branches of this vessel, no active vasomotor fibers having been demonstrated by stimulation of the hepatic nerves, although, as in the case of the brain and lung blood vessels, a certain amount of constriction may occur under the influence of epinephrine. The bloodflow through the portal vein is dependent on changes oc- curring at either end of the distribution of the vessel, that is, changes occurring in the liver itself or in the intestine. Of these factors the lat- ter is no doubt the more important, an increase not only in portal blood pressure but also in portal bloodflow being readily produced by dila- tation of the splanchnic blood vessels; for example, as the result of sec- tion of the splanchnic nerve. Alterations in portal bloodflow brought about by changes in the caliber of the vessels in the liver itself are partly dependent upon changes in the branches of the hepatic artery. Let us consider briefly how this may be brought about. At the point where the portal and hepatic arteries come together — that is, at the in- PECULIARITIES OP BLOOD SUPPLY IN CERTAIN VISCERA 257 trahepatic capillaries — the pressure of the blood in them must become equal, which means that in its course through the interlobular connec- tive tissue, the branches of the hepatic artery must offer much resistance to the blood flowing through them. This frictional resistance resides in the hepatic arterioles, and since these are richly supplied with constric- tor nerves, great variation in hepatic inflow becomes possible. These changes will affect the degree of tension of the interlobular connective tissue in which the arterioles lie. In this tissue, however, also lie the thin-walled branches of the portal vein. When therefore the tension of this tissue becomes greater, as a result, for example, of vasodilatation in the hepatic artery, the portal vein radicles will become compressed and the bloodflow along them impeded. Conversely, when vasocon- striction occurs in the hepatic arteries, the congestion of the connective tissue becomes diminished, the veins dilate, and the blood flows through them more readily (Macleod and R. G. Pearce21). Experimental evi- dence in support of the above view is furnished by observing the out- flow of blood from the liver before and during stimulation of the he- patic plexus. The first effect is an increase in the outflow, which very soon returns to its original amount, even though the stimulation of the plexus is kept up during the experiment. This return to the normal flow must indicate either that the constriction of the hepatic artery has not been maintained, or that it has been maintained but is accompanied by a compensatory increase in the flow through the portal vein. As a matter of fact, we know that the hepatic artery remains constricted as long as the hepatic plexus is stimulated, indicating that the conges- tion of the connective tissue in which the venules lie has become reduced to such»an extent, as a result of the constriction, that these open up and permit the blood to flow through them more readily. The initial in- crease in outflow immediately following upon stimulation of the hepatic plexus, is no doubt caused by the squeezing out of the blood already in the hepatic vessels, and it is a result which is often observed in other organs during stimulation of vasoconstrictor nerve fibers. THE CORONARY CIRCULATION We have already studied the effect produced on the heartbeat by in- terfering with the flow of blood in the coronary vessels, and it remains for us to study: (1) peculiarities in the bloodflow through them, and (2) whether this bloodflow can be altered by dilatation or constriction of the vessels brought about through nerves. With regard to the pecu- liarities of bloodflow, it may be stated that there is said to be two periods in each cardiac cycle during which an increase takes place in the mass 258 THE CIRCULATION OF THE BLOOD movement of blood in the coronary vessels — namely, at the beginning of systole, and again at the beginning of diastole. Nevertheless the pressure pulse has the same contour in the coronary as in the systemic circulation. (W. T. Porter.22) During systole the intramural branches of the coronary artery are ^compressed and the blood pressed out of them. This emptying of the vessels favors the flow of blood through the heart walls. Regarding the presence of coronary vasomotor nerves, there is at pres- ent a certain amount of doubt. When strips of the coronary artery are suspended in a solution of epinephrine, they undergo relaxation instead of contraction. On the assumption that the action of epinephrine on blood vessels is the same as that of stimulation of the vasoconstrictor fibers, this result has been taken as evidence of the absence of such fibers and the possible presence of vasodilator fibers. A somewhat similar type of experiment has been performed by injecting epineph- rine into the fluid used to perfuse the excised mammalian heart, with the result that, when such injections are made into a heart that is not beating, evidence of vasoconstriction is obtained, whereas when injected into a beating heart, dilatation occurs. This latter result may, however, be owing to the action of the epinephrine in stimulating the cardiac contractions. Other observers, however, deny that the in- jection of epinephrine into the coronary circulation has any influence upon the outflow of the perfusion fluid. Taking the result of these observations as a whole, we may at least conclude that epinephrine does not produce the same marked vasoconstriction that it produces in other blood vessels — a fact, which, as already stated, may be taken advantage -of in bringing about the rise in coronary pressure .that is necessary for successful resuscitation of the heart. Attempts to demonstrate the presence of vasomotor 'fibers by electrical stimulation of the vagus or sympathetic nerve have yielded results. which are quite inconclusive, although some observers assert that the vagus nerve carries vasoconstrictor fibers to the coronary vessels, and that the sympathetic carries vasodilator. CHAPTER XXIX CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS* In the following chapters a brief account will be offered of the clinical use of the electrocardiogram, of polysphygmograms, and of bloodflow measurements. This is done to show how physiologic technic is being employed for the accurate investigation of cardiovascular disease. ELECTROCARDIOGRAMS To observe the electrical change produced by the spread of the excita- tion wave over the heart from auricles to ventricles, it is not necessary to place the electrodes directly on the heart, but, as already hinted, we may follow the electrical change by leading off from electrodes applied to the surface of the body. From such electrocardiographic tracings extremely important facts concerning the propagation of the heartbeat may be ascertained. In order to make an observation the hands and the left foot are each placed in a solution of sodium chloride contained in porous jars, immersed in larger vessels, containing a saturated solution of ZnS04 and zinc terminals, t An arrangement like that in Fig. 81 may also be used. By manipulation of suitable keys, the extremities may then be connected with the electrocardiograph in the following manner: Lead 1, right arm and left arm; lead 2, right arm and left leg; lead 3, left arm and left leg. Through lead 1, the current acting on the galvanometer will be that produced more especially at the base of the heart. Through lead 2, the current will pass through the long axis of the heart, and through lead 3, it will pass mainly along its left border. When any pair of leads is connected with the galvanometer, it is ob- served that the string is deflected to one side owing to electrical cur- rents arising from the skin. Before taking a record of the cardiac movements of the string, it is necessary to compensate for this skin cur- rent by introducing into the circuit in the opposite direction the re- *A certain amount of repetition of. matter previously discussed has been found advisable in these chapters for which the indulgence of the reader is requested. tit is really unnecessary to use the so-called nonpolarizable electrodes. Glass vessels containing 20 per cent NaCl solution with the zinc plates dipping into them arc quite satisfactory. 259 260 THE CIRCULATION OP THE BLOOD quired amount of current, called the compensating current, to bring the strong shadow back to the zero or midposition. In order that the rec- ord obtained may be quantitative in character, it is further necessary that the movement of the string be standardized. This is done by as- certaining to what extent the string moves when a current of known voltage is sent through it and by altering the tension of the string so that one millivolt of current causes an excursion of one centimeter of the string shadow on the photographic plate. It would take us beyond the Fig. 81. — Etectrocardiographic apparatus as made by the Cambridge Scientific Materials Co. Con- tact electrodes are shown, but the immersion electrodes described in the context are preferable. confines of this volume to go in any greater detail into the technic in- volved in taking electrocardiograms, but it may be said that this is by no means difficult, provided the instructions which are supplied with the instrument are carefully followed. In practice the taking of elec- trocardiograms is indeed quite a simple matter, and the extremely im- portant information which they give us concerning the mechanism of the heartbeat and the evidence of myocardial disease should make their employment a universal practice in all cardiac clinics. Some of these clinical applications are described elsewhere (page 266). ELECTROCARDIOGRAMS 261 What particularly interests us here is the contour of the electrocardio- gram in a normal person (Fig. 82). It will be observed that there are three waves above the line of zero potential and two waves below it. They have been lettered from before backward, P, Q, R, S, and T, and in all such records when correctly obtained, the waves above the line of zero potential indicate that the base of the heart is negative to the apex. The exact cause of each wave has been ascertained by taking simultaneously with the electrocardiogram a record of the mechanical changes occurring in the heart during each cardiac cycle. Such records -T Fig. 82.— Normal electrocardiogram. Leads 1, 2, 3. Note that the height of the R deflection in lead 3 equals the difference between the height of RI and 7?2- have been secured by taking intracardiac pressure curves with the results as shown in Fig. 83. The top curve represents auricular and the second one ventricular pressure, whereas the lowest is an electrocardiogram. It will be observed: (1) that the P-wave occurs just antecedent to con- traction of the auricles; (2) that the small positive wave, Q, which is ab- sent in these tracings, must occur just before the beginning of the con- traction of the ventricles; (3) that the negative wave, R, occurs just be- fore and during the early part of ventricular systole — that is, during the presphygmic period; and (4) that the long upward wave, T, culmi- nates at the moment the ventricle begins relaxing. 262 THE CIRCULATION OF THE BLOOD Although such comparisons give us considerable insight into the cause of several of the waves, there yet remain certain peculiarities of the electrocardiogram to be considered. These are: (1) the cause of the slight positive wave, Q; (2) the cause of the- positive wave, S; (3) the cause for the period of equal potential at the base and apex during ven- tricular systole indicated by the portion of the curve between S and T ; (4) the cause for the negative wave, T. To solve these problems it is necessary to compare electrocardiograms taken from the surface of the body with those from electrodes placed directly on the base or apex of the ventricle of the exposed heart. Fig. 83. — Ivlectrocardiogram {dog) taken simultaneously with curves from auricle and ven- tricle. It will be observed that wave P slightly precedes auricular systole and that wave R occurs just before the presphygmic period starts in the ventricle. (From Lewis.) The Ventricular Complex In view of the nature of the electric change which occurs in a strip of denervated muscle when a wave of contraction passes along it (page 188), the simplest interpretation of the ventricular part of the above curve is that the contraction must pass into the ventricle at a little dis- tance from the base, thus causing the latter, for a moment of time, to be positive to the rest of the ventricle, and accounting for the slight down- ward wave, Q. Immediately after this the base of the ventricle becomes negative to the apex, giving us the marked upward wave, R, which however lasts for but a short period of time, being followed by an inter- val during which the base and apex are of the same electrical potential (horizontal part of wave between R and T). Finally the base again be- comes negative to the apex, thus accounting for the smaller upward ELECTROCARDIOGRAMS 263 wave, T. The cause of the occasionally observed downward wave, S, following R, is obscure. The most significant fact in the electrocardiogram is therefore that the base is negative to the apex at the beginning (It-wave) and again at the end (T-wave) of the ventricular contraction. How may this be ex- plained? When electrocardiograms are taken through electrodes placed directly on the base and apex of the ventricle of the exposed heart, it has been found that the contour of the electrocardiogram is like that which is obtained from a strip of muscle when a wave of contraction passes along it: it is diphasic in character (page 188), a result which may be interpreted as indicating that the wave of contraction starts at the base and ends at the apex. This rules out the explanation, at one time suggested for the T-wave, that the wave starts at the base, then proceeds to the apex, and finally ends at the base, following the disposi- tion of the muscular fibers of the ventricle in a folded or loop form, with the bend of the loop at the apex and the free ends at the base. Al- though the explanation seemed at first to conform with the embryo- logic fact that the heart is developed from a folded tube, it can not hold, as has been shown by observing the course of the excitation wave se- cured through electrodes placed at various points on the surface of the exposed ventricle (page 194). The explanation which is accepted by the majority of observers at the present time is to the effect that the T-wave is caused by the longer con- tinuance of the electric change at the base of the ventricle than at the apex. To test this hypothesis the crucial experiment would evidently be to see whether a T-wave could be induced in an electrocardiogram, such as that of the frog ventricle, in which no T-wave exists, by hurry- ing up the contraction process at the apex without affecting it at the base. This can be done by local warming of the apex, or by applying the ventricular electrode at varying parts' of the ventricle in an excised heart beating in Ringer's solution of relatively high H-ion concentra- tion. Mines showed that under these conditions a typical T-wave ap- pears in the electrocardiogram, as shown in Fig. 84.* The existence of the small Q-wave, indicating that the contraction does not really start from the base, conforms with the observation that the Purkinje system of fibers ends about the papillary muscles, which therefore would be the first to contract, and with the observations of Lewis, already alluded to above, on the appearance of the negative vari- ation on the surface of the exposed heart. The most important clinical application of the electrocardiogram is "This tracing was found among those left by Professor Mines of McGill University, and for permission to use it the author is indebted to the authorities of that institution. 264 THE CIRCULATION OF THE BLOOD A. — Normal B. — Apex cooled C. — Apex warmed Fig. 84. — Records of electrocardiogram and movement of ventricle of frog showing that when the apex is warmed a typical T-wave appears in place of a wave in the opposite direction appear- ing when the apex is cooled. (From Mines.) ELECTROCARDIOGRAMS 265 undoubtedly in connection with the determination of the rate of trans- mission of the excitation wave from auricle to ventricle; thus, the P-R interval, as it is called, indicates the time taken for the impulse to travel from the sinoauricular to the auriculoventricular node and bundle. In delayed transmission this interval becomes abnormally long. Obvi- ously also conditions of .heart-block, of auricular fibrillation, or of auric- ular flutter will be immediately revealed by the electrocardiogram. The interpretation of abnormalities in the contour of the ventricular portion of the curve is, however, not so easy a matter, and should never be undertaken unless curves from the three leads have been secured, for it will be found that the corresponding electrocardiograms differ from one another in detail; for example, the R-wave is usually most prominent in lead 2, although sometimes it is more prominent in lead 3. T is always upright in normal individuals in curves taken from lead 2, but it is not infrequently inverted in those of lead 3, and may show partial inversion in those from lead 1. The Q-R-S group is often of peculiar contour in curves from lead 3. These variations are possibly dependent upon the relative preponderance of the musculature in the left and right ven- tricles, for it is evident that the amount of muscle included in the path- way between the two leads will vary. CHAPTER XXX CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (Cont'd) CLINICAL APPLICATIONS OF ELECTROCARDIOGRAPHY The Electrocardiogram in the More Usual Forms of Cardiac Irregularities BY R. W. SCOTT The principle of the application of the string galvanometer to the study of cardiac irregularities has been indicated. It is our object here to outline some of the more common forms of irregular heart action, with a brief description of the abnormalities in the electrocardiogram resulting therefrom. For the sake of comparison a normal electrocar- diogram is shoAvn in Fig. 82. The cause and relationship of the various deflections have been explained (see page 262). Sinus Arrhythmia. — This irregularity is seen commonly in children and young adults, and is without pathologic significance. The electro- cardiogram presents the normal deflections and shows by the varying spaces between the P deflections that the cardiac impulse has been gen- erated at slightly irregular intervals. Sinus Bradycardia. — The electrocardiogram in a simple case of sinus bradycardia is usually normal, except that the deflections occur at an unusually slow rate (Fig. 85). This indicates that the cardiac impulse is built up at a slow rate, but when generated it evokes a normal auric- ular and ventricular contraction. The Extrasystole. — The extrasystole may be either auricular or ven- tricular in origin. Occasionally a rare type is seen in which the im- pulse arises in the junctional tissues between the auricle and ventricle. When the focus of impulse production is at or near the sinoauricular node, the resulting electrocardiogram complexes are practically normal. If, however, the seat of impulse formation is removed from the S-A node, the P deflection may be distorted or actually inverted, followed by a normal Q-R-S-T complex (Fig. 86). In the case of ventricular extrasystole, the cardiac impulse originates in either the right or the left ventricle. This abnormal site, together 266 CLINICAL APPLICATIONS OP ELECTROCARDIOGRAPHY 267 Fig. 85. — Sinus bradycardia. Rate 32 per minute. Note the normal appearance of the electro- cardiogram. P-R interval = .17 seconds. Fig. 86. — Auricular extrasystole. Two auricular extrasystoles following two normal complexes. Note the ectopic origin of the extrasystoles indicated by the inversion of P. Fig. 87. — Ventricular extrasystoles arising in the right ventricle. Fig. 88. — Ventricular extrasystole arising in the left ventricle. 268 THE CIRCULATION OF THE BLOOD with the path which the impulse takes, produces a much greater differ- ence of electric potential than is seen in the normal electrocardiogram. When the impulse arises in the right ventricle near the base, the prin- Fig. 89. — Paroxysmal tachycardia. Auricular origin. Note that the P deflection falls back on T. Rate 200 per minute. cipal R deflection is upwards in both leads 1 and 2. Arising near the apex, the principal R deflection is up in lead 1 and down in lead 2. Two extrasystoles both arising in the right ventricle are shown in Fig. 87. Fig. 90. — Auricular fibrillation. Leads 1, 2, 3. Note the coarse fibrillation waves between the R peaks, and the absence of any B deflections in relation to R. Also the unequal spacing of the R deflections. In the case of the left ventricle, a basal ' impulse gives a downward principal deflection in lead 1 and up in lead 2. When the aberrant fo- cus is located near the apex of the left ventricle, the principal deflec- CLINICAL APPLICATIONS OF ELECTROCARDIOGRAPHY 269 tion is down in both leads 1 and 2. Any one or several of the general types of extrasystole may occur in the same patient. Fig. 88 shows an extrasystole originating from the left ventricle. Paroxysmal Tachycardia. — Electrocardiographic records taken in the interval between the paroxysms may appear normal. During the tachy- cardia the records normally show only two deflections, R and a combina- tion of T and the succeeding P (Fig. 89). If the paroxysm is of auric- ular origin, the P deflection may be inverted, indicating that the new focus of impulse production is located at some other site than the sino- auricular node. Rarely the new focus may be in the ventricles. Records taken during the paroxysm may show a rapid succession of deflections, simulating isolated ventricular extrasystoles. Auricular Fibrillation. — The electrocardiogram in auricular fibrilla- tion shows three distinctive features: 1. Absence of the P deflections typical of auricular contractions. 2. The ventricular complexes (Q-R-S-T waves) occur in irregular se- quence and may vary in height. 3. The presence of small irregular oscillations best seen between the ventricular complexes. A typical tracing of this condition is shown in Fig. 90. The dependence of the P-wave upon auricular contraction has been indicated (page 261). Its absence in auricular fibrillation is accounted for by the fact that the individual muscle fibers of the auricles contract independently of one another, so that some fibers are in a state of con- traction while others are relaxed. This renders impossible a coordinate contraction of the auricle as a whole. The multiple impulses from the fibrillating auricles reach the ventri- cles and evoke a contraction provided the ventricle is not already in a state of contraction (refractory period, page 178). These irregular ventricular responses will of course produce unequal spacing of the ventricular complexes in the electrocardiogram. The variations in the height of the R deflections is thought to be due to the distortion caused by the superimposition of the small waves representing auricular ac- tivity. These small waves must occur throughout the whole cardiac cycle, but are more or less masked by the ventricular complexes, appear- ing as separate oscillations only during diastole. Auricular Flutter. — Auricular flutter was discovered by the electro- cardiograph, and it is practically impossible to make a diagnosis of this condition without the use of the string galvanometer. The auricular deflections are usually rhythmic and in the average case vary in rate from 200 to 350 per minute. The initial deflection of P may be base negative or apex negative — up or down — depending on the site of the 270 THE CIRCULATION OF THE BLOOD origin of the auricular impulse (when arising from some other source than the S-A node the impulse is said to be ectopic). Usually a regular succession of P deflections can be traced throughout the record (Fig. 91). Since it- is impossible for the ventricle to respond to all the impulses coming from the auricles, a condition of partial heart-block obtains (2:1 — 3:1 — 4:1, etc.). The ventricular complexes will occur regularly- except when a 3:2 rhythm exists. l?ig. 91. — Auricular flutter. Auricular rate 300. Ventricular rate 80. Note the inversion of the P deflections. Usually the ventricular complexes are such as to indicate that the stimulus arose in the auricle (supraventricular). The height of the individual deflections Q-R-S-T may vary, depending on the predominance of a right or left ventricular hypertrophy. Fig. 92.— Delayed conduction. Note the normal appearance of the electrocardiogram except for the prolongation of the P-R interval, which measures .23 seconds. Heart-block. — There are three degrees of severity in heart-block: (1) delayed conduction, (2) partial dissociation, and (3) complete dissocia- tion. Any one of these conditions may be present in the same patient at successive intervals. DELAYED CONDUCTION. — When the conducting tissues of the heart are so affected as to cause an abnormal prolongation of the P-R interval, the condition is called delayed conduction. The ventricles respond to each stimulus originating at the sinus node, but the time required for the impulse to pass through the conducting tissues is longer than normal. CLINICAL APPLICATIONS OF ELECTROCARDIOGRAPHY 271 In a simple case the electrocardiogram may appear perfectly normal, but when the P-R interval is measured accurately, it will be found to be lengthened beyond the extreme limits of the normal (0.20 seconds) (Fig. 92). PARTIAL DISSOCIATION. — In the typical case of partial dissociation the Fig. 93. — Partial dissociation. Note the failure of ventricular response following the second P, which has been preceded by two extrasystoles (x) of ventricular origin. ventricles respond to the impulse coming from the auricle most of the time, but occasionally fail to do so, when the condition is called "dropped beat." The electrocardiogram records a P deflection but no ventricular complex, showing that the auricles have contracted at their usual rate but that the ventricles failed to respond to the stimulus coming from the sinoauricular node (Fig. 93). Fig. 94. — Complete dissociation. Note that the P wave spaces regularly and bears no definite re- lation to the R wave of the ventricular complex. Auricular rate 72. Ventricular rate 40. COMPLETE DISSOCIATION. — In a simple case of complete dissociation the auricles beat independently of the ventricles; hence the P deflection of the electrocardiograms bears no relation to the ventricular complex (Q-R-S-T) (Fig. 94). The P deflections space regularly and are easily made out when they fall during diastole of the ventricle. Occasionally 272 THE CIRCULATION OF THE BLOOD the auricle will happen to contract during ventricular systole, causing a distortion of the ventricular complex by the superimposition of a P deflection. Except when this occurs the Q-R-S-T complex is the normal supraventricular type. The P deflections occur more frequently than the Q-R-S-T complex, showing that the auricles are beating more often than the ventricles. The auricular rate in the average case of complete heart-block is about 72, while the ventricular rate is much slower (35 to 40). CHAPTER XXXI CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (Cont'd) POLYSPHYGMOGRAMS In taking polysphygmograms, the following technic is usually followed; Venous Pulse Tracings. — Direct the observed person to lie down with his head slightly raised by a cushion and bent to the right side. Place the receiver (thistle funnel) over the jugular bulb on the right side of the neck. This lies immediately above the inner end of the clavicle. Bring the style of the recording tambour to write with a minimal amount of friction on the paper or drum. Since a venous pulse tracing can not be interpreted without a simultaneous tracing from an artery, now adjust the button of a receiving tambour over the radial artery and adjust the style of its recording tambour so as to write on the drum in the same perpendicular as the style of the venous tambour. Tracings should be taken with the recording surface at a moderate speed. Before disturbing, the relative positions of the writing points, allow them to make vertical marks (with recording surface stationary) at various parts of the tracings. These alignment marks permit of ac- curate comparisons between the curves. Repeat the above, using the carotid instead of the radial. A time tracing (l/5 sec.) should always be taken simultaneously. The polysphygmograph is shown in Fig. 95. To interpret the venous curve, make a vertical mark on the arterial pulse tracing corresponding to the beginning of the pulse upstroke. If this is done on the radial pulse tracing, measure one-tenth of a second in front of it, and make a vertical mark to allow for the time lost in propa- gation of the pulse from the heart to the radial artery. This line 3 (corrected in case of radial pulse) corresponds to the be- ginning of the sphygmic period of ventricular systole — i. e., to the open- ing of the semilunar valves. Measure the distance from it to the near- est vertical line that was made to indicate the relative position of the writing points. Then measure off the same distance in the venous trac- ing from the corresponding indicator line. This will fall at the begin- ning of the small wave (c), which is due to the bulging into the auricles of the closed auriculoventricular valves. (Fig. 96.) 273 274 THE CIRCULATION OF THE BLOOD The auricular wave (a) occurs one-fifth of a second in front of c, and may now be ascertained by measuring off this distance in front of c. This is line 1. Now measure the distance on the radial pulse tracing from the begin- ning of the upstroke to the dicrotic notch. The distance between these is the sphygmic period (E). Fig. 95. — Polysphygmograph. This instrument records in ink on glazed paper two simul- taneous tracings, i. e., radial pulse and one other, such as carotid, jugular, apex beat, etc., in addi- tion to the time tracing. The ink tracings are both more convenient and permanent than smoked paper tracings. The clockwork operates at variable speeds, permitting the taking of protracted records at different speeds. Measure off the same distance on the venous tracing from c. Line 5 will be found to fall just before a small wave (v), which is due to the sudden opening of the tricuspid valves. This practically coincides with the dicrotic notch on the radial pulse tracing. Sometimes a little wave Fig. 96. — Normal jugular tracing. The spacing below shows the duration of the a-c interval. (From E. P. Carter.) occurs on the upstroke of wave v just where line 5 falls. This co- incides with the closure of the semilunar valves. The distance between it and wave v corresponds to the postsphygmic period. The cause for the depression (marked x) following c will readily be understood by referring to the intraauricular curve (Fig. 97), to which, as already explained, the venous pulse tracing is qualitatively similar. POLYSPHYGMOGRAMR 275 The rise in the curve following depression x is caused by the filling of the auricle with blood. This goes on until v, when the tricuspid valves open, allowing the blood to fall into the ventricle. Fig. 97. — Reduced tracings from carotid, aorta, ventricle, auricle and jugular, to show the general relationships of the various waves. An electrocardiogram is also shown. Note that the jugular and auricular curves have the same contour, and that the depression (x) in them occurs during systole of the ventricles. (After Lewis.) To interpret the cardiogram, adjust receiving tambours to the radial and open beat with both writing styles in the same perpendicular, and following the other directions described under "venous pulse" mark on the cardiogram: (See Fig. 98.) Fig. 98. — Polysphygmograms including jugular, apex and radial tracings. Line 4 on the radial tracing is first of all located. It is then transferred (by measurement from the alignment mark on the right edge of the tracing) to the jugular and 1/10 second subtracted from it, giving line 3. When this is similarly transferred to the apex tracing, it falls somewhere 'on the upstroke the be- ginning of which is line 2. 276 THE CIRCULATION OF THE BLOOD 1. The beginning of the sphygmic period (E) (line 5). 2. The end of the sphygmic period (E) (line 5). 3. The auricular wave. 4. The beginning of ventricular systole (difference between 1 and 4 equals presphygmic interval). 5. The opening of auriculoventricular valves (lowest point in tracing). Mark also on the tracing the exact moment at which the heart sounds are heard. It is important to make certain that the button of the tambour is ac- curately over the apex beat, since otherwise a depressed or negative wave may be inscribed at ventricular systole. Simultaneous Arterial Pulse Tracings. — The velocity of the transmis- sion of the pulse wave is calculated by measuring the time between the systolic rise in the carotid and in the radial arteries, tracings of which are taken by applying one receiving tambour to the carotid artery and another to the radial artery. Abnormal Pulses The following is a brief description of the main character of abnormal pulses: The Ventricular Pulse. — In this no "a" waves are present in the jugular tracing, the heart action being either regular or irregular. In the former case, the absence of the "a" waves may depend on: (1) over- filling of the right auricle, (2) increase in the heart rate, or (3) complete heart-block associated with auricular fibrillation. When the heart is irregular, the absence of the "a" waves signifies auricular fibrillation. Delayed Conduction and Heart-block. — This causes a change in the time relationship of the "a" and "c" waves in the jugular curve. When the heart-block is of the -first degree, the "a-c" interval merely becomes lengthened, but when it is of such degree that the normal impulse some- times fails to be conveyed along the auriculoventricular bundle, isolated •"a" waves can be detected. In the higher degrees of heart-block there are regularly recurring "a" waves having no constant time relationship to the ''c" waves. For the purpose of exact analysis of the curves in suspected cases of delayed conduction, it is often advantageous to draw vertical lines below the tracing representing the beginning of auricular and ventricular systole. This has been done in the tracing reproduced in Fig. 99. The line joining these two verticals indicates the conduction time or "a-c" interval. When it exceeds one-fifth of a second, there is delay in the conduction time. POLYSPHYGMOGRAMS 277 A tracing showing a higher degree of heart-block is given in Fig. 100. Sinus Arrhythmia.— In this condition the radial pulse is markedly irregular, but the "a," "c" and "v" waves of the jugular tracing occur with the usual time relationship to one another, and there is no delay in the "a"-" c" interval. _ X pjg. 99. — Delayed conduction time. First stage of heart-block. The A-C intervals measure more than 0.2 second. (From E. P. Carter.) Sinus Bradycardia. — The beat originates at long intervals in the sinus; the "a-c" interval is normal, and the radial pulse very slow but practically regular. Premature Beats. — These may be either ventricular or auricular in origin. In the former case the "a" waves on the jugular tracing space regularly throughout, but the "c" waves at the point of disturbance Fig. 100. — Dropped beats. Second stage of heart-block. (From F*. P. Carter.)' coincide with the "a" waves, giving therefore a more pronounced wave. This is due to a premature contraction of the ventricle occurring about the time of the "a" wave, so that the latter finds the ventricle in a re- fractory state (see page 178). The premature contraction is therefore followed by a compensatory pause, which is evident on the tracing. An example of such a case is given in Fig. 101. In doubtful cases the exact 278 THE CIRCULATION OF THE BLOOD site of origin of the premature beats can be determined only by careful measurement of the distances between the various beats of the ventricle. Whenever an irregularity repeats itself and the duration of one cycle of the arrhythmia accurately corresponds to another, the irregularity 1 I I I I Fig. 101. — Premature beats (extrasystoles) ventricular in origin at PB. Compare the duration of the intervals marked A and B' with those marked C and D. (From E. P. Carter.) may be due to: (1) premature auricular or ventricular contractions; (2) the occasional occurrence of dropped beats (a failure of ventricular response) ; or (3) a high degree of heart-block with a wide variation in the ventricular response. The important point to note here is that, no matter how irregular such a tracing may appear, if the irregularity re- peats itself it can not be due to auricular fibrillation. Fig. 102. — Paroxysmal tachycardia. The paroxysms start at xx fallowing normal beats and lasting for seven beats. The clue to "a," which falls with "v" after the first premature contrac- tions, is found in the initial beat of the new rhythm. (From E. P. Carter.) Paroxysmal Tachycardia. — When the rate of a regular pulse is sud- denly altered but the change in rate bears a simple ratio to the previous rhythm, the change may be due to (1) premature ventricular contrac- tions which do not reach the radial, or (2) to the sudden development POLYSPHYGMOGRAMS 279 of a two-in-one heart-block. When on the other hand, there is no exact ratio between the slow and the rapid rate, the change is due to the sud- den appearance or disappearance of paroxysmal tachycardia. The paroxysms during which the auricle is beating very rapidly may last for a variable time, such attacks persisting off and on for hours or even days. The tracing in such a case is given in Fig. 102. Fig. 103. — Auricular flutter. In this case the ventricular rate varied from 82 to 98 per minute. (From E. P. Carter.) Auricular Flutter. — It is impossible to diagnose the not infrequent existence of this cardiac condition without the use of either the poly- sphygmogram or the electrocardiogram. The jugular curve may be of two types, one made up of rapid, more or less uniform waves, the other of waves that are paired with a constant time interval between the pairs. Fig. 104. — Auricular flutter. Note the relative rates of A and V, and also that the ventricular rate is regular. (From E. P. Carter.) All of the frequent beats of the auricle do not reach the ventricle in this condition, so that the ratio between auricular and ventricular beats may be 1:3 or 1:4. The condition must therefore not be confused with heart-block, the main point of distinction being that in the latter condi- tion the ventricular pulse is slow and the auricular about normal. The 280 THE CIRCULATION OF THE BLOOD radial pulse may be regular or irregular. The cause for the failure of every auricular beat to travel to the ventricle during auricular flutter is partly the refractory condition of the bundle, and partly the refrac- tory phase of ventricular contraction. The bundle may be considered as a narrow bridge which will transmit the impulses across it only at a certain rate. If the impulses arrive too rapidly, only some of them can cross the bridge, and even of those that do cross, a limited number only will find the ventricle in a condition of excitability because of the re- fractory period (see page 178). Tracings showing auricular flutter are ^ . _ I s: K\ I w h* K* ha K\ V Fig. 105. — Auricular fibrillation. Note the absence of all "a" waves from the jugular tracing, the marked irregularity of the radial pulse, and the occurrence of "c" and "v" during the sphygmic period. (From E. P. Carter.) given in Figs. 103 and 104. In one of them the radial pulse is regular; in the other, irregular. Auricular Fibrillation. — The contractions of the auricle, as already ex- plained, are entirely irregular, so that the jugular tracings show an en- tire absence of all "a" waves,. the radial tracing being characterized by the complete absence of a dominant rhythm and by great variation in the length of the individual beats from one cardiac cycle to the next. This irregularity does not repeat itself, and the long pauses are not simple multiples of the shortest pause. Tracings from a case of auricular fibrillation are shown in Fig. 105. CHAPTER XXXII CLINICAL APPLICATIONS OF CERTAIN PHYSIOLOGIC METHODS (Cont'd) THE MEASUREMENT OF THE MASS MOVEMENT OF THE BLOOD Method. — The apparatus used for this purpose consists essentially of a vessel containing a known quantity (3,000 c.c.) of water and a ther- mometer from which a change of temperature of a hundredth of a de- gree centigrade can be read. In order to diminish as much as possible the loss of heat between the vessel and the outside air, the walls are double, the space between being stuffed with broken cork. The top of the vessel is closed with a thick cork plate, having suitable openings in it for the hand or foot and for the thermometer and a stirrer (feather) with which to keep the water in constant motion. The apparatus is called a calorimeter. After the hand or foot has been in the calorimeter, with the water a few degrees below that of the body, for a certain time (ten minutes), the temperature of the water will of course become raised, and the degree to which this occurs, multiplied by the volume of the water in cubic centimeters, will give in calories the amount of heat dissipated. By the application of a very simple formula it is now an easy matter to calculate how much blood must have passed through the blood vessels of the part in order to give out the observed amount of heat; for, if we divide the calories by the difference in temperature between the inflowing and out- flowing blood of the part, the result must indicate the volume of blood, in cubic centimeters, that has passed through it (since by definition a calorie equals volume multiplied by difference in temperature). It remains to explain the equation by which the results are arrived at. If Q equals the amount of blood, H the calories of heat given out to the calorimeter, T the temperature of the arterial blood and T' the temperature of the H* venous blood, then we have the equation : Q=™ — 7^7. It has been shown *For the determination of H we must multiply the cubic centimeters of water plus the water equivalent of the hand and calorimeter (because both of these will absorb some heat) by the dif- ference in temperature plus the self-cooling of the calorimeter (because some heat is lost to the air during the observation). The water equivalent of the hand is equal to its volume multiplied by 0.8; that of the calorimeter must be determined for each instrument and is usually about 100 c.c. The self-cooling of the calorimeter is determined by observing the fall in temperature for a period equal to that of the actual observation without the hand in the calorimeter. 281 282 THE CIRCULATION OF THE BLOOD by Stewart that T may be taken as the same as that of the mouth, or 0.5° C. below that of the rectum, and T' as the average temperature of the water in the calorimeter during the observation. To allow for the specific heat of blood, the result is multiplied by-2 tension of blood is supposed to be 5.75. differs slightly from that just described in that a bubble of air is exposed to a relatively large quantity of blood, so that after a time actual equilibrium of gas tension becomes established between the bub- ble and the gases of the blood. This apparatus is shown in Figs. 122 and 123. It consists of a graduated tube of narrow bore sur- rounded by a water jacket. To the upper end of the graduated tube a small syringe is attached. The lower end of the graduated tube ex- pands into a thistle-shaped bulb, closed below by a cork, through which is inserted a tube (inflow tube) ending near the top of the bulb in a fine opening and connected outside with an artery. An outflow tube is also connected with the thistle-shaped bulb. At the beginning of the experiment the thistle-shaped bulb and the graduated tube are filled with physiologic saline. By means of the syringe a small bubble of air is then introduced, so that it lies at the THE CONTROL OP THE RESPIRATION 339 junction of the thistle-shaped bulb and the graduated tube. As the blood is allowed to enter through the inflow tube, it is ejected in a fine stream around the bubble of air, which moves about in the stream. The blood displaces the saline out of the bulb into the side tube. After the bub- ble has been subjected to the influence of the blood for some minutes, the gases in it come into perfect equilibrium with those in the blood. The percentage of 02 and C02 in the bubble will therefore correspond to the tension of these gases in the blood. The analysis is effected by drawing the bubble into the graduated tube by means, of the syringe, . Fig. 122. — The gas analysis pipette for the microtonometer sh see context. (From A. Krogh.) Fig. 123 "' bub ' Fig. 123. own in Fig. 123. For description context. (From A. Krogh.) Fig. 123. — Microtonometer, to be inserted into a blood vessel. The small circle represents the ble of air. For further description see context. (From A. Krogh.) measuring its capacity, transferring it into a bulb containing KOH, which absorbs the 002, then taking it back into the capillary tube and again measuring. The shrinkage obviously corresponds to the amount of C02. The bubble is then transferred into potassium pyrogallate solu- tion, where the 02 is absorbed.* The Tension of C02 and 02 in Alveolar Air. — Having seen how we may determine the tension of the gases in blood, we must now consider *Since thei above was written, a more efficient tonometer devised by the late T. G. Brodie has been described by O'Sullivan (Am. Jour. Physiol., Sept., 1918). 340 THE RESPIRATION the method by which the tensions of these gases in alveolar air can be determined. The simplest and until recently the most accurate method is that of Haldane and Priestley.19 This consists in having an individual, Avith his nostrils clamped', breathe quietly through a piece of hose pipe about a meter long, which has at the mouth end a short side-tube lead- ing to an evacuated gas-sampling bulb of about 50 c.c. capacity.* After the subject has become accustomed to breathing through the tube, he is asked to make a forced expiration and at the end of it to close the mouthpiece with his tongue. At this moment the operator opens the tap of the sampling tube, allowing the air from the tubing through which the individual has made the forced expiration to rush in and fill it. This sample represents the air from the alveoli (see page 302), and is analyzed for percentages of C02 and 02. Since each normal inspira- tion dilutes the alveolar air somewhat, it is necessary, for constant re- Fig. 124. — Apparatus for collection of a sample of alveolar air by Haldane's method. It is better to use a mouthpiece than a mask. suits, to make two analyses of alveolar air from each subject, one taken at the end of a normal inspiration and the other at the end of normal expiration. The average of the two results is taken as the composition of the alveolar air. On account of the difficulty in securing intelligent cooperation in the application of this method, particularly Avith children, others have been devised. One of the simplest is that of Fridericia, which is a modifica tion of the Haldane-Priestley method, the apparatus for which is shown in the figure (Fig. 125), and the manipulation of which is outlined in the legend. Another is to take a mixed sample of the very last portion of several normal expirations. On account of the extended use which is being made of measurements of alveolar air composition, both in lab- *In place of the gas-sampling tube it is much more convenient and equally accurate to employ one of the modern ground glass piston syringes (L,uer). The piston should, of course, be well smeared with a good mineral grease. THE CONTROL OF THE RESPIRATION 341 Fig. 125. — Fridericia's apparatus for measuring the CO™ in alveolar air. The person expires forcibly through the tube with the stopcocks as in I. A is closed and the tube placed in water to cool the air, after which B is turned as in II. The entrapped column of air equals 100 c.c. A solution of caustic alkali is now sucked into C with stopcocks as in II. B is then turned as in I but with A still closed, and the alkali solution allowed to enter b, after which B is turned off, the excess of alkali solution in C allowed to run out and the burette shaken. The burette is then submersed up to a in a cylinder of water, with B as in III. After allowing for cooling, the level at which the water stands gives the per cent of CO». 2093 in. inspired air Co* Fig. 126. — Curves to show the relationship between the O2 and CO2 tensions in alveolar air (dotted lines) and arterial blood (continuous lines). It will be observed that the tension of CO2 in blood is slightly above that in alveolar air, but that the reverse relationship obtains for Oa. In the upper part of the curve the O2 tension in the alveolar air was experimentally altered, causing a corresponding alteration in the O2 tension of the blood. This result is of practical significance in connection with O2 alterations in gas poisoning, pneumonia, etc. (From A. and M. Krogh.) 342 THE RESPIRATION oratory and in clinical work, a special chapter has been devoted to the subject, giving in detail the more recent methods devised by R. Gr. Pearce. Lastly, it should be noted that several observers believe that a more reliable estimate of the alveolar tension of C02 (and of 02) can be made by analyzing a sample of ordinary expired air and calculating the per- centages of C02 and 02 in the alveolar air by allowing a constant dead- space capacity of 140 c.c. (Krogh, etc.). If we compare the C02 tension of arterial blood, as measured by the Krogh method, with that of alveolar air, we shall find that there is a remarkable correspondence, indicating, therefore, that, when the arterial 2.9 0,05 Co, 2.9 fa. a spirettair 220 30 Fig. 127. — Same as Fig. 126, except that in this case the tension of CO2 in the alveolar air was experimentally altered. (From A. and M. Krogh.) blood leaves the alveoli, its partial pressure or tension of C02 is exactly equal to that in the alveolar air. This is shown in the accompanying curves of experiments performed by Krogh. The dotted line in these curves represents the tension of C02 or 02 in alveolar air, and the con- tinuous line, these tensions in arterial blood. Close correspondence will be observed between the C02 curves even when sudden changes in alveolar C02 were induced by artificial means. In the case of the 02 tensions, however, that of the blood is always lower than that of the alveolar air, the differences being especially marked when the 02 ten- sion in the alveoli is raised (Pigs. 126 and 127). Tension of C02 in Venous Blood. — If we examine the C02 tension of the venous blood coming to the lungs, we shall find that it is distinctly THE CONTROL OF THE RESPIRATION 343 higher than that in the alveolar air. The earliest method for measuring it consisted in passing a lung catheter into the right bronchus and then blocking the passage above the open end of the catheter by inflating a rubber collar or ampulla. The renewal of air in the right lung is thereby prevented, and a sample of the stagnant air can be removed and analyzed. In such a case, however, the blood will have circulated several times round the body, and with only one lung operating the risk is incurred that more C02 is being discharged into the blocked lung than cor- responds to the tension of C02 of venous blood under normal conditions. Much more practical methods are those of Haldane, Yandell Hender- son and R. G. Pearce, which are much the same in principle. In Pearce 's method, the person first of all inspires from a gas meter containing a gaseous mixture with about 10 per cent of C02. Immediately after fill- ing the lungs, he makes a rapid forced expiration into a tube provided with a valve having four openings. This valve is turned through a complete circuit during the expiration, so that four fractions of the ex- pired air can be collected in rubber bags connected with side tubes opening opposite the four openings in the valve. The first fraction will contain a little less than 10 per cent C02, the second distinctly less, while the fourth will contain the same as the third, indicating that equi- librium between the C02 of the alveolar air and the blood must have been attained. This figure therefore gives us the tension of C02 in the venous blood of the lungs. In Henderson's method the rebreathing is per- formed into gas receivers containing 6 per cent C02. These results then indicate that the whole process by which C02 is exchanged in the lungs is dependent on the law of gas diffusion ; the gas diffuses from a place of lower to a place of higher pressure, and does so until equilibrium is attained. CHAPTER XXXIX THE CONTROL OF RESPIRATION (Cont'd) THE ESTIMATION OF ALVEOLAR GASES BY R. G. PEABCE, B.A., M.D. Methods such as that of Haldane and Priestley, which calculate the mean percentage composition of the alveolar air by analysis of a sample taken from the end of -a prolonged forced expiration, give values which are too high for C02 and too low for 02. There are several reasons for this: (1) In the time taken for the prolonged deep expiration an appreciable amount of C02 will be given off by the blood to the alveolar air, and oxygen will be absorbed — that is, the sample will not contain the same percentages of C02 and 02 at different stages of expiration. (2) The portion of the tidal air which reaches the alveoli dilutes the alveolar air and thus causes the amount of C02 given off by the blood to vary during the different phases of respiration. If we bear in mind that the tensions of CO2 in the alveolar air and in the blood leaving the lungs are always the same (page 343), and that the entire fa.ll in C02 tension in the alveolar air occurs during inspiration, then it is clear that the blood in the pulmonary capillaries must have a maximum tension and load of C02 at the end of expiration, and a minimum tension and load of C02 at the end of inspiration. Accordingly, the average of the per- centage of C02 and 02 at the end of inspiration and expiration, as de- termined by the Haldane-Priestley method or by any of its modifications, must fail to give the correct mean tension of these gases in the alveolar air during expiration. The error which makes the C02 higher than it should be, makes the percentage of 02 less than it should be. These in- fluences taken along with the fact, which will be shown later, that the evolution of C02 from the blood is relatively more rapid at low than at high tension of C02, indicates that the blood in the pulmonary capil- laries during inspiration must contribute a greater part of the C02 excreted during a respiratory cycle than that in the pulmonary capil- laries during expiration, and moreover that a greater part of the C02 excreted must be evolved at a tension which is below the mean tension of the C02 present in the entire time of the expiration. We conclude, therefore, that the average tension of C02 in the alveolar air, determined 344 THE CONTROL OF THE RESPIRATION 345 by the actual tension under which the gas is evolved from the blood, is less than the average tension of C02 in the alveolar air during the time of a respiratory cycle. In the case of 02 the conditions are different. While the diluting effect of the alveolar tidal air is marked in altering the amount of C02 given off during the different phases of a respiration, it can have little influence on the amount of 02 taken up by the blood under normal con- ditions. This is evident from a study of the dissociation curve of hemo- globin (page 383), which shows that at tensions above 65 mm. Hg the hemoglobin is practically saturated with 02. .Since the tension of 02 in the alveolar air under normal conditions is greater than 65 mm. (95-100 mm.), the rate of absorption of 02 must be practically maximal during the respiratory cycle — that is, it will not change at different phases of it. While the relationship of the alveolar gases is continually changing at different stages of the respiratory cycle, their mean relationship for periods including several respirations or for complete respirations is more or less constant, being controlled by the type of the metabolism, and mathematically expressed by the respiratory quotient (page 547). The average relative percentages of the two gases in the alveolar air must therefore be the same as in the tidal air. In the alveolar air col- lected by the Haldane method, however, the above factors cause the respiratory quotient to be less than that in the tidal air. These points have been insisted upon because much of the knowledge of the gaseous exchange between the blood and the air in the lungs, as well as the control of respiration, has been built upon data obtained by the Haldane-Priestly method, and in considering this work, which we shall do in subsequent pages, it is advisable that we be aware of the limitations of the method employed. The method has been an invaluable one for opening up a hitherto entirely unexplored field of research, but now, the pioneer work having been done, we must employ methods which will enable us to explore 'more exactly. An Accurate Standard Method for Normal Subjects. — The most accu- rate method, and one free from many of the theoretic errors present in the others, depends on the relationship found to exist between the dilut- ing effect of the air in the dead spa.ce (see page 302) and the known per- centage composition of the alveolar air in expirations which are of vary- ing depths but of equal and normal duration and which follow normal inspirations (R. G. Pearce). In this method the subject is made to breathe through valves, which automatically separate the inspired from the expired air. The expired air is led into a tube con- nected with two spirometers by two three-way stopcocks. The spirometers are of the 346 THE RESPIRATION Gad-Krogh type, one being capable of holding ten liters, and the other one and a half. The exact time during which air enters is recorded by the small spirometer by means of a grooved dial on the axis of the lid, on, which a thread works over a system of pulleys, and any movement is accurately recorded by a writing point on the smoked paper of a drum. The spirometers are connected so that the air current may be directed in the three following ways: (1) through Cocks 1 and 2 outside; (2) directly through both cocks into the large spirometer for the purpose of collecting a series of expirations; and (3) through Cock 1 directly into the small spirometer for catching a single expira- tion. In all experiments the first filling of the spirometer is rejected, so that the dead space of the spirometers is filled with air of approximately the same composition as in the succeeding expirations. The time is marked in seconds by a time clock. The respira- tory movements are recorded by a pneumograph. (Fig. 128.) The subject is brought into respiratory equilibrium by having him breathe through the valves for a period of time before the observation. The respiratory movements during this time are recorded while the cocks are in Position 1. When the observation is started, the cocks are turned into Position 2 during the time an inspiration is being Fig. 128. — Arrangement of meters and connections of Pearce's method for measurement of COs of alveolar air in normal subjects. made, so that the expirations which follow may be collected in the large spirometer. After about ten respirations (a counted number) have been collected, the cocks are turned to Position 3 during an inspiration, and a single deep expiration is collected in the small spirometer. In order that the time of this may be the same as the normal expiration, it is necessary to quicken it. This is more or less a chance procedure, but with a little training, the operator can close the stopcock with sufficient accuracy to interrupt the deep expiration, at the end of the normal expiratory time. Should there be any gross variation from the normal expiratory time, the sample must be col- lected again. Not infrequently the inspiration immediately preceding the expiration into the small spirometer is varied involuntarily by the subject on account of his being aware that the following expiration has to be deepened and quickened; this can be partially overcome by giving1 him the signal to breathe out deeply after he has actually begun to expire. Determinations are made of the average volume of the tidal air (c.c. air in large spirometer divided by number of breaths), of the volume collected from the deep ex- piration, and of the percentage composition of the tidal air and that of the deep expiration. A criterion for determining whether or not the procedure has been carried THE CONTROL OF THE RESPIRATION 347 out correctly is the respiratory quotient (ratio of CO., excreted to O2 absorbed). For reasons which are set forth above, the quotients should be approximately equal in the air collected in the large and in the small spirometers; if they are not so, the condi- tions of the method have not been correctly carried out. Since the dead space and the average composition of the alveolar air under these conditions may be considered constant, the percentage composition of the deep expira- tion will differ from that of the mixed sample of several normal expirations in propor- tion as the dead space exerts a greater diluting effect in the small than in the large expiration. This being the case, the data obtained can be combined algebraically to give either the capacity of the air passages or the percentage composition of the alveolar air. Let A = amount of air in large expiration (small spirometer), Ai = amount of air in small or normal expiration (tidal air), B = the percentage of CO2 or O2 in the expired air of large expiration, Bi = the percentage of CO2 or O2 in the. expired air of small expiration, x = the capacity of the dead space, y = the average percentage of C(X or O2 in the alveolar air; then, A x B = (A - x)y and Ai x Bi = (Ai - x)y. Solving this for x, y remaining constant under the same physiologic conditions, we A x Ai x (B-Bi) J have: x — , the dead space. Or solving for y, we have: AxB-AixBi y = . * I 12^ i, the mean percentage of CO,, in the alveolar air. In case the A- Ai dead space for O2 is desired, B and Bi must be made to equal the O2 absorbed. Clinical Method. — The use of the kymograph and pneumograph, and other complicating factors, make the method as just described quite im- practicable for clinical procedure, but the use of the same apparatus with the following modification will yield satisfactory results for most clinical purposes. The patient is made to respire through the valves for a short time, after which the observer collects a single expiration in a small spirometer by turning the stopcock from Position 1 to 2. A sam- ple of this is taken for analysis, and the spirometer is again emptied and a series of successive samples of deeper expirations taken. This is done by directing the patient, after he has started to breathe normally into the spirometer, to breathe more deeply. The amount of air col- lected in each expiration is controlled by the observer by closing the stopcock when the desired volume is obtained. By this means one can collect several expirations differing from one another by increasing amounts but all occupying the same time. The samples of the various expirations are collected in a series of numbered sampling syringes, and the gaseous composition of each is determined. When the percentage of C02 or 02 in each expiration is plotted on cross section paper on the ordinates, with the volume of the expirations in c.c. on the abscissae, a hyperbolic curve should be obtained. Any marked deviation from such a curve indicates that some error has been made in taking a sample, and 348 THE RESPIRATION this observation should be discarded. The different observations are then combined in the formula given on page 347. The determination of the C02 percentage of expired air is so simple that a number of speci- mens of varying depths of expiration can be taken and thus many points on the curve determined. For the most accurate results it is in general best to compare only those expirations which differ from one another by at least 0.3 per cent in C02 and by at least 200 c.c. in volume. This depends on the fact that the diluting effect of the dead space in reduc- ing the percentage of C02 in the expired air from that in the alveolar air is greater in relatively small expirations. If more exact work is de- sired, the 02 content can be determined on each specimen, the respiratory quotient calculated, and only those expirations which show the same respiratory quotient combined. In the table each observation is compared with each of the others in all possible combinations. • PrR CENT ALVEOLAR CO., DKAD SPACE OBSERVA- EXPIRED CO., IN EXPIRED 1 2 3 1 2 3 TION AIR AIR 1 450 3.10 2 637 3.66 4.99 170 3 750 4.00 5.34 189 4 960 4.28 5.30 5.48 5.27 189 183 214 5 1120 4.30 5.11 5.15 4.92 161 140 184 6 1440 4.40 5.16 4.98 4.82 171 127 17"! General average for CO2 in alveolar air, 5.13. General average for dead space, 172. Dead space in valves in this experiment was about 30 c.c. Another method which has been suggested for clinical purposes is that of Plesch; this consists in having the subject breathe several times in and out of a small bag. It is assumed that after such respiration the composition of the air in the bag will become the same as that in the alveoli. Although this is no doubt true, it has been shown that the method is fallacious, because the C02 tension determined in this way is not that of the arterial blood alone, but is the average between it and that of the venous blood. THE CONTROL OF RESPIRATION (Cont'd) THE NATURE OF THE RESPIRATORY HORMONE The practical importance of the observations described in the foregoing chapters in the investigation of the relationship between CH of the blood and respiratory activity will now be plain, and it remains for us to consider the physiologic evidence that such a relationship exists. In the first place, let us consider the behavior of the acid-base equilibrium 'during conditions of abnormal breathing- — hyperpnea and dyspnea.* As C02 accumulates and 02 becomes used up in a confined space, the breathing becomes intensified. In searching for the exact cause of this effect, we must first of all ascertain whether the hyperpnea is due to the deficiency of 02 or to the accumulation of C02. Many of the experi- ments bearing on these problems can be more satisfactorily performed on man than on laboratory animals, because anesthesia is not necessary and the subjective symptoms experienced are of great value in the inter- pretation of the results. If an individual is placed in a large air-tight chamber (2000 liters' capacity), and the depth and rate of breathing ob- served as the C02 accumulates and the 02 becomes used up in the air of the chamber, no distinct change in respiration will be observed until the C02 percentage of the air has risen to almost 3. Above this point, how- ever, the hyperpnea becomes more and more pronounced, until finally, when the C02 percentage has risen to about 6 and the 02 percentage has fallen to 13.5, it becomes unbearable (dyspnea). From the results of the foregoing observation alone we could not, however, decide whether the excitation of the respiratory center is due to the deficiency of 02 or .to the increase of C02. If the experiment is repeated with the difference that the C02 as it accumulates is absorbed by soda lime, no hyperpnea will develop even when the 02 is as low as in the previous experiment. We may conclude, therefore, that in the first experiment C02 accumulation must have acted as the respiratory stimulus. The same conclusion is arrived at as a result of observations on indi- viduals caused to breathe in a more confined space as into a rubber bag of about 225 liters' capacity. Under these conditions hyperpnea de- *Hyperpnea means slightly increased breathing; dyspnea, labored breathing, but yet with suffi- cient ventilation to maintain life; asphyxia, the results of insufficient breathing. 349 350 THE RESPIRATION velops more rapidly than in the large cabinet, and a higher percentage (10 per cent) of C02 can be tolerated. That in this case also deficiency of 02 is not responsible for the hyperpnea can be shown by repetition of the experiment either with an excess of 02 in the bag or with absorption of the C02 by soda lime. In the former case hyperpnea will develop as usual, while in the latter it will not supervene until the percentage of 02 has fallen below 10, when cyanosis becomes marked. In fact, some people become cyanosed and unconscious, and collapse under these conditions before there is any respiratory disturbance. A peculiarity of the effect of 02 deficiency is that the person may be unaware of the seriousness of his condition ; indeed he may be somewhat stimulated. The conclusion may be drawn that deficiency of 02 per se can serve as a respiratory stimulus only when it is so extreme as to cause other serious symptoms. This conclusion does not rule out an important influence of 02 deficiency in increasing the excitability of 'the center towards C02. Under ordi- nary conditions, however, the center is far more sensitive towards plight changes in the C02 percentage. There is an obvious reason why the adjustment of pulmonic ventila- tion should not depend upon changes in 02 supply to the respiratory cen- ter. If it were so, many other tissue activities and other nerve centers would suffer from the 02 deficiency before there was time for the breath- ing to become stimulated sufficiently to make good the loss of 02. As a matter of fact, headache, dizziness, nausea and even fainting are almost certain to be caused whenever any muscular exercise is attempted in an atmosphere containing a deficiency of 02 but no excess of C02 (cf. moun- tain sickness). An adequate 02 supply of the body is, therefore, insured by changes in C02 tension of the blood. Quantitative Relationship between C02 of Inspired Air and Pulmonary Ventilation. — These results suggest, as the next step in the investigation of our problem, the determination of the quantitative relationship be- tween the C02 percentage of the respired air and the amount of air breathed (pulmonic ventilation).* That there is such a relationship has been most successfully demonstrated by R. W. Scott, who used for his purpose decerebrate cats.f The trachea was connected, through a T-tube provided with valves, with tubing leading to a large bottle and a Gad-Krogh spirometer, so that the animal breathed out of the bottle into the spirometer, these two being also connected together. The spirom- *A distinction is somewhere drawn between pulmonic ventilation and alveolar ventilation, the former being the total amount of air that enters and leaves the lungs, and the latter, that which en- ters and leaves the alveoli. This distinction is based on the assumption that the capacity of the dead space may vary considerably from time to time, which, as pointed out elsewhere, is erroneous. For practical purposes pulmonic ventilation is the s'afer value to give. fDecerebrate animals must be used in these experiments, since anesthetics markedly depress the activity of the respiratory center. THE CONTROL OP THE RESPIRATION 351 eter was made to record its movements on a drum, so that an accurate record of the depth and frequency of the respirations was secured. Sam- ples of air were removed from the bottle by ground-glass plunger syringes at frequent intervals during the time that the animal was respiring into the tubing. joe 400 300 /oo Fig. 129. — Composite curve obtained from the data on sixteen experiments, showing the re- spiratory response to COa in the decerebrate cat. Abscissae = percentage of COa in the inspired air. Ordinates = the percentile increase the tidal air per minute. (From R. W. Scott.) The results are given in the accompanying curve (Fig. 129), which shows that there is a perfect correspondence between the C02 percentage in the air of the bottle and the pulmonary ventilation. Moreover, when the bottle was filled with 02 instead of air to start with, the same results were obtained, showing that the C02 accumulation alone was responsible for the hyperpnea. In these cases the percentage of 02 remaining in the 352 THE RESPIRATION system after hyperpnea had become extreme, was far above that at which direct excitation of the center from 02 deficiency is possible. Experiments of a similar type had previously been performed by Por- ter and his pupils,21 but their object was not so much to show the close parallelism between the C02 content of the respired air and the pulmonic ventilation as to demonstrate the changes produced in the sensitivity of the respiratory center in pneumonia. Possibility that C02 Specifically Stimulates Center. — After showing that C02 acts as an excitant of the respiratory center, the question arises whether we are justified in the assumption that has been made tentatively that the action depends on the raising of the CH of the blood, or whether it may be a specific action of the HC03 anion itself. Many attempts have been made to decide this question experimentally, the general principle of the experiments being to determine whether CH of the blood runs parallel with the C02 content of the respired air and with the hyperpnea. Using the gas-chain method (page 31), Hasselbalch and Lundsgaard22 found that the hyperpnea produced in rabbits by breathing in C02-rich air runs approximately parallel with the increase in the CH of the blood, but on account of the experimental difficulties encountered they could not decide whether changes in CH are alone responsible for the effect. These authors had previously demonstrated that changes in CH can be induced in blood removed from the body by alterations in the C02 tension within the physiologic limits. An increase of one millimeter in C02 tension Avas found to cause an increase in CH of 0.0065 x 10 7 (see page 27). R. W. Scott's experiments, above referred to, have, however, yielded more definite results. By using the colorimetric method for determining CH of the blood (see page 32), it could be readily shown, as is evident from the table (col. 8 in table), that a marked rise in CH became evident when the inspired air contained 5 per cent or more of C02. That this rise was due to increase in the C02 tension was shown not only by finding a greater percentage of C02 (col. 15) in the blood, but also by being able to demonstrate that when C02-free air was bubbled through the blood removed during the dyspnea, CH immediately returned to the normal, which it also did when the blood removed after the animal had breathed for a few minutes in outside air (col. 16). The C02 content likewise re- turned (col. 17). Had the increase in acidity been caused by nonvolatile acids — lactic, for example — these results, particularly the latter, could not have been obtained. Although there is therefore no doubt that the CH of the blood ma" be raised because of an increase in C02 in solution in the blood plasma — a C02 acidosis, as we may call it (s*ee page 354) — this does not prove that the stimulation of the respiratory center is brought about solely by CH. THE CONTROL OF THE RESPIRATION 353 The increase in the carbonate ion — HC03 ion — itself might also serve as a stimulus. That such is actually the case was demonstrated by finding that, if the CH of the blood was first of all lowered by injecting alkali intravenously, hyperpnea still developed in proportion as the CO., accumulated in the inspired air; and that CH of the blood, when the hyperpnea was at its highest, was below that of normal blood. Some other factor than CH must obviously be responsible for this result. This must be the HCO, anion. THE EFFECT OF REBREATHING CARBON DIOXIDE ON THE MINUTE VOLUME AND ON THE H-ION CONCENTRATION AND TOTAL CARBONATE CONTENT OF THE ARTERIAL BLOOD IN THE DECEREBRATE CAT FIFTEEN MINUTES PRELIMINARY PERIOD* REBREATHING PERIOD AFTER RE- BREATHING PERIOD g • B 'E V B t- u B 1 B " "S «.s U o -> C o g u "* Time c 1 .*2 ojs B O, o. i-i "o, E 73 0^3 o3 M £ •M • .2 t. *c O "73 ^ .2 fa 'rt '|H ^ «~g V *" <£ 3 •a B o rt "-1 4) 3 ."2 "« 0 S 0 rt" 55 o rtl~1 M g > H « P PH H M H 0. O fL, H PH per per kg. c.c. c.c. c.c. cent cent c.c. c .c. 27 36 2.2 10.30 a.m. 38.5 48 864 7.45 30.3 64 3,968 540 6.45 7.25 53.4 7.4 34.2 28 33 2.2 1.50 p.m. 38.8 22 616 7.40 40.2 40 2,880 367 5.02 7.20 58.0 7.35 42.0 29 35 2.0 10.30 p.m. 38.6 28 784 7.40 38.1 40 3,840 389 5.35 7.25 51.1 7.4 40.0 30 32 2.3 11.00 p.m. 38.7 32 768 7.45 30.2 54 4,536 489 6.40 7.20 51.1 7:4 35.0 31 31 2.0 11.15 a.m. 38.8 32 896 7.40 32.0 60 4,800 436 5.95 7.20 50.0 7.4 35.2 Avera ge 32.4 785 7.42 34.1 *PH is the actual value given in the table. This is inversely proportional to CH. Further corroboration of the claim that the HC03 anion has a specific stimulating effect on the respiratory center that is independent of CH, has been furnished by Hooker, Wilson and Connett.23 These authors succeeded in keeping the centers of the medulla alive by perfusion with defibrinated blood through the blood vessels of the brain, and found that, although the respiratory movements of the diaphragm became de- pressed with a decrease and excited with an increase in CH of the per- fused fluid, a greater activity of the center was produced when this con- tained a high tension of C02 than with another fluid of the same CH but with a low tension of C02. We conclude that, although the CK is the important respiratory hormone, the carbonate ion (HC03 anion) also has a stimulating influence. 354 THE RESPIRATION Relationship Among Acidosis Conditions, Alveolar C02 and Respir- atory Activity. — It will be plain that variations in the respiratory hor- mone, whatever this may be, are associated with changes in the C02 content of the alveolar air. Closer examination has shown, however, that this relationship is by no means always so simple as in the instances just described, where increased respiration was found to be associated with an increase in alveolar CO,. There are many cases where the re- verse relationship obtains — namely, decreased alveolar C02 and hy- perpnea. The whole question is very closely linked with that of the con- trol of the reaction of the body fluids and with the etiologic factors in acidosis. When it is fully answered, many obscure clinical conditions in which respiratory disturbances occur will be much better understood than they are at present. On account of its great importance, considerable attention will be devoted in the next few pages to some of the researches which have been made bearing on the relationship between the C02 of the alveolar air and the various modified types of breathing that can be produced experimentally or become developed under altered physiologic conditions. We shall consider these conditions in the following order: (1) Con- stancy of the alveolar C02 under normal conditions and during moderate variations in barometric pressure. (2) The quantitative relationship between an artificially induced increase in alveolar C02 tension (as by breathing C02-rich air) and the increased respiration. (3) The results of these observations will demonstrate a very precise relationship to exist between alveolar C02 tension and respiration, but if we proceed to repeat the latter observations under conditions where the accumulation of C02 in the inspired air is accompanied by oxygen deficiency (as by breathing in a confined space), we shall see that the relationship no longer holds, indicating that the oxygen deficiency has caused something to happen which disturbs it. We shall find that the disturbing factor is accumulation of unoxidized acids in the blood, and this will naturally lead us to study the conditions in which such acids might develop; namely, (4) Breathing in rarefied air (mountain sickness). (5) Apnea. (6) Muscular exercise. In succeeding chapters, we intend to review the work in these fields in considerable detail, partly because of its very important bearing on the general question of the control of the respiratory center and partly be- cause of the light the observations throw on the nature of the mechanism involved in the adjustment of the CH of the blood and tissues. As we have seen, much work concerning the physicochemical principles involved in the control of the reaction of the blood has been contributed during recent years by physical and biological chemists, but much of this THE CONTROL OF THE RESPIRATION 355 work in our judgment fails to pay sufficient regard to the extraordinarily complicated conditions existing in the animal body, and more particu- larly, to correlate the purely physicochemical data with the numerous observations that have from time to time been recorded by physiologists regarding the behavior of the respiratory center. Physical chemists have recently, for example, gone so far as to define acidosis as a condition in which there is a diminution in the bicarbonate content of the blood in- duced by the discharge into it of fixed acids. This is going too far, for it fails to recognize acidosis due to an increase in the C02 of the blood. r TT CO ~\ It is the molecular ratio ,T *^A which determines the tension of C02. [NaHCOg J When C02 is added to the blood, either experimentally by respiring the gas, or naturally OAving to muscular exercise or to pathologic conditions in which there is a deficient excretion of C02, as in heart disease, the ten- dency of the equation to change, by increase of the numerator, is pre- vented partly by stimulation of the respiratory center, which gets rid of C02, and partly by an increase in the denominator. The respiratory center is so sensitive to slight increases in CH that it becomes excited before a sufficient increase in H2C03 has occurred to disturb the normal r H CO ~\ ratio HT tT^A When fixed acids are added to the blood the denom- L NaHCOg.J inator of the equation, NaHC03, is lowered and consequently CH rises, and increased respiration occurs, lowering H2C03 and thus reestablishing the ratio. CHAPTER XLI THE CONTROL OF RESPIRATION (Cont'd) THE CONSTANCY OF THE ALVEOLAR C02 TENSION UNDER NORMAL CONDITIONS Since a close relationship exists between the alveolar C02 tension and the respiratory activity, it is to be expected that the two would bear a strict proportionality to each other, and since the breathing under normal conditions does not vary much, the C02 tension should also be constant. Many observations show this to be the case. The tension is remarkably constant from day to day and even from month to month in the same individual, provided the physiologic conditions are the same. A slight seasonal variation is said to exist, a rise in the temperature of the en- vironment of the individual causing a slight depression in the C02 ten- sion, while a fall in temperature causes a slight rise (Haldane). These changes are independent of any demonstrable change in rectal temper- ature and, therefore, are probably due to the influence of the temperature on the skin. Since it is the number of molecules of C02 in a given volume of alve- olar air (i. e., the partial pressure or tension) that is of importance, it is only when the barometric pressure is the same that the percentage of C02 in the sample of alveolar air can be constant. To allow for this, all results are reduced to standard barometric pressure (760 mm. Hg). If the barometric pressure is lowered, there will have to be a higher percentage of C02 in the sample in order that there may be the same tension of this gas in the air of the alveoli ; and vice versa when the bar- ometric pressure is raised. The equation by which this tension, ex- pressed in millimeters of mercury, is determined is: 100:760: :a:p, where a is the percentage actually found in the air of the sampling tube and p the tension. A correction must also be introduced in this equation to allow for the vapor tension of the air in the alveoli, for of course H20 molecules will behave like C02 molecules in causing a partial pressure. When reduced to this standard, it has been found that the tension of C02 in the alveolar air remains constant under the different barometric conditions that obtain at the top of a mountain or at the foot of a deep mine. This is shown in the following table: 356 THE CONTROL OF THE RESPIRATION 357 (1) (2) (3) BAROMETRIC CO, ACTUALLY FOUND PARTIAL PRESSURE PRESSURE IN DRY ALVEOLAR OF CO, IN MOIST (MM. HO) AIR ALVEOLAR AIR AFTER (PERCENT) CALCULATING FOR BAROMETRIC PRESSURE Top of Ben Nevis 646.5 6.62 5.23* Oxford 755 5.95 5.53 Foot of Dolcoath Mine 832 5.29 5.48 Compressed air cabinet 1260 3.52 5.64 *The figures in this column are arrivpd at hv tf B' - A x P' P. when P — figures in B last column; B' — figures in first column; A = aqueous tension of alveolar air; P' = figures of second column; B = barometric pressure at sea level. A is obtained from tables giving the aqueous tension at different temperatures. Changes in the frequency of breathing that are within physiologic limits have no influence on the tension of alveolar C02, provided that exactly the same time is taken in performing the forced expirations during which the samples of alveolar air for analysis are removed. The Degree of Sensitivity of the Respiratory Center to Changes in the C02 Tension of the Alveolar Air This can be determined by observing the alterations produced in the volume of air that actually enters the alveoli (alveolar ventilation) dur- ing breathing in atmospheres containing different percentages of C02. In man an increase of from 0.2 to 0.3 per cent in the alveolar C02 is sufficient to double approximately the alveolar ventilation ; or, more pre- cisely, an increase of ten liters in the air entering and leaving the alve- oli per minute is caused by raising the alveolar C02 tension by from 2.2 to 3.1 mm. Hg (Douglas, etc.)24. THE ALVEOLAR C02 TENSION DURING BREATHING IN A CONFINED SPACE We have already employed similar experiments in ascertaining whether C02 accumulation or 02 depletion is responsible for the hyperpnea pro- duced under these conditions. We concluded for the former, but now on closer examination we shall see that, although our conclusion was correct, the deficiency in 02 also has an indirect effect on the respiratory center. This is revealed by the fact that the tension of the C02 in the alveolar air does not increase in proportion to the observed increase in pulmonary ventilation. We must conclude that the decrease in 02 has some effect. How may this be explained? Two possibilities exist: (1) that the 02 want has caused organic acids to accumulate in the blood and so raise the CH ; and (2) that in the absence of a certain tension of 358 THE RESPIRATION 02 the excitability of the center is raised (i. e., its "threshold" lowered), so that it becomes stimulated by CH, to which ordinarily it does not re- spond. We shall now proceed to examine the experimental evidence bearing on these possibilities. By examination of the alveolar air "of an individual confined in a pneu- matic cabinet in which the barometric pressure is gradually lowered, it has been found that although the C02 tension remains constant for a considerable range (cf. page 356), it begins to fall when the barometric pressure has reached about 550 mm. Hg. At this pressure the tension of 02 in the alveolar air will be 62 mm. instead of its normal of about 105 at atmospheric pressure. Below it the alveolar C02 tension quickly falls, and at the same time hyperpnea becomes evident, although the person himself may be unaware that he is breathing more deeply. If this experiment is repeated with the difference that, as the pressure is lowered, an excess of 02 is introduced into the chamber, the hyperpnea does not supervene until a barometric pressure has been reached that is distinctly lower than when no excess of 02 is present, and at the same time the C02 tension in the alveolar air remains unchanged. The ex- planation of this result is that by lowering the 02 tension in the alveolar air and, therefore, in the blood and tissues, oxidative processes become depressed so that unoxidized acids, such as lactic, accumulate in the blood and by adding their effect to that of the C02 serve to raise the CH of the blood. As a result, the respiratory center becomes excited, hy- perpnea supervenes, and the volatile CO, is removed from the blood into the alveolar air. On supplying 02 artificially, this failure of proper oxidation does not set in and breathing goes on normally- There should be a stage in the above experiment during which the CO, tension of the alveolar air is increased — namely, when the fixed acids first appear and decompose the carbonates of the blood. This stage has, however, not been detected. When a person is kept in such a chamber for some time at a pressure which causes a diminution in the alveolar C02 tension, the tension does not immediately return to its normal level when atmospheric air is again breathed, indicating that the fixed acids are only slowly got rid of. The second hypothesis — namely, that the 02 deficiency directly raises the excitability of the respiratory center — has many advocates, among them Lindhard,25 who found that, Avhen the percentage of 02 in the alve- olar air was raised, a higher percentage of C02 was necessary to cause an increase in the ventilation of the lungs, and conversely, that a distinct increase in the excitability of the center occurred when the inspired air contained less than the normal percentage of 02. Although it is ad- mitted by Haldane and his school that such alteration in the excitability THE CONTROL OF THE RESPIRATION 359 of the respiratory center to the CH of the blood may occur after long- continued exposure of the center to the changed tension of 02, yet they deny that such alteration can occur as a temporary condition. These workers found that, in order to raise the pulmonic ventilation by 100 per cent, the increase in the alveolar C02 tension required was practically the same (0.3 per cent) when the inspired air contained 20 per cent of O2 as when it contained 54 per cent. In the observations already referred to on the decerebrate cat, R. W. Scott20 has secured some evidence that would seem to support Haldane's contention. He found that the response of the respiratory center to the percentage of C02 in the respired air was exactly the same whether the latter contained a low (13-14) or a high (30 and over) percentage of 02. The possibility that the excitability of the respiratory center is affected directly by the 02 tension is to be considered as one of the most im- portant problems awaiting solution. Even if it may have a certain influence on the excitability of the re- spiratory center, 02 deficiency per se can serve as a direct stimulus of the center only when it is of extreme degree. Much light has been thrown on the relationship of 02 to respiratory activity by observing the respirations during breathing in and out of rubber bags through soda lime absorption bottles of sufficient size to remove the C02. We have already seen that even the general results of such observations (page 349) show clearly how much more potent a respiratory stimulant is accumulation of C02 than deficiency of 02. More particular investi- gation in which the alveolar air is analyzed bears out these conclusions and at the same time indicates the exact conditions under which organic acids become -developed. With a very small bag (a few liters' capacity) hyperpnea of a dis- tressing type but without cyanosis supervenes in a few minutes, and the alveolar air contains perhaps as low as 6 per cent 02 and 4 per cent C02. Of still greater interest and significance, however, is the fact that the ratio between the volume of C02 excreted and of 02 absorbed (respira- tory quotient) during the hyperpnea is raised considerably above unity, indicating that an. excessive excretion of C02 must be occurring. This result is explained by assuming that the deprivation of 02 causes large quantities of fixed acids to be produced, and that these expel C02 from the blood more quickly than the 02 is absorbed. In corroboration of this explanation, it has been observed that, after outside air is breathed for some time following the above experiment, the respiratory quotient becomes very low, so that C02 must now be accumulating in the blood. If the above experiment is repeated Avith a larger bag (about 200 liters), so that the 02 falls slowly, the breathing can be maintained for 360 THE RESPIRATION a much longer period without any evident symptoms of hyperpnea, even though the 02 percentage in the alveolar air may fall as low as in the previous experiment, and there are marked symptoms of 02 want, such as cyanosis, twitching of the muscles of the hands, lips, etc. The re- spiratory quotient does not become abnormal in this experiment indicat- ing that no expulsion of C02 from the blood can have occurred as in the previous experiment. The cause for the virtual absence of hyper- pnea in this experiment is no doubt that the more gradual reduction in 02 of the alveolar air and therefore of the blood did not bring about the accumulation of -lactic acid at a rate that was greater than that at which the C02 was got rid of into the alveolar air. BREATHING IN RAREFIED AIR; MOUNTAIN SICKNESS In considering the part played by fixed organic acid in the control of the CH of the blood, the most important results have been secured by observations on the condition of individuals living at high altitudes. As is well known, under these conditions certain symptoms are likely to develop, the condition being known as mountain sickness. The great interest which physiologists have taken in this subject has been owing, not so much to the importance of the observations in connection with the condition itself, as to the light which they throw on the mechanism of respiratory control and on the cause for abnormal types of breathing. More or less hyperpnea, especially on exertion, soon appears in a rarefied atmosphere, and the alveolar C02 tension assumes a value con- siderably below the normal. For example, at sea level the minute vol- ume of air breathed in one individual was 10.4 liters, and the alveolar C02 tension 39.6 mm. Hg. After being some time on Pike's Peak, where the barometer registers only 459 mm. Hg, Douglas26 found the minute volume of air to be 14.9 liters, and the alveolar C02 tension 27.1 mm. Hg. At first sight the above statement may seem to contradict one pre- viously made, to the effect that the alveolar C02 tension remains constant at different barometric pressures. This applies, however, to the imme- diate effects, whereas we are now considering the later effects. The im- portant point is: How are we to reconcile with the above hypothesis the fact that a diminution in the alveolar C02 tension should be accompanied by hyperpnea? A solution of the seeming contradiction will not only be of importance in connection with our present problem, but will assist us in the investigation of the clinical conditions of hyperpnea, in which likewise a diminished C02 alveolar tension is often observed. Mountain sickness may indeed be considered as an intermediate condition between the physiologic and the pathologic. From what we have learned we should expect the above result to be THE CONTROL OF THE RESPIRATION 361 dependent upon an increase in the nonvolatile acid content of the blood That such is really the case has been conclusively shown both by titra- tion methods and by observing the dissociation curve of hemoglobin, which, as will be explained later (see page 386), may be made to serve Fig. 130. — The horizontal interrupted lines represent the mean normal alveolar CO2 and O^ pressures at sea level (i.e., Oxford and New Haven); the thick line, alveolar CO2 pressure; and the thin line, alveolar O* pressure. (From Douglas, Haldane, Henderson, and Schneider.) as an index of the H-ion concentration of the blood. The exact chemical nature of the nonvolatile acids that accumulate in the blood is not as yet known. Two types of acid can be thought of, either unoxidized organic 362 THE RESPIRATION acids, of which lactic acid may be taken as the representative, or inor- ganic substances, like the acid phosphates. That it is not lactic acid is shown by both direct and indirect evidence. The direct evidence has been furnished by Ryffel, who was unable to find any increased per- centage of this substance either in the urine or in the blood of persons who had been living for some time in the famous Regina Margherita hut on Monte Rosa.27 The indirect evidence has been furnished by ob- serving the time that it takes after the individual has started breathing the rarefied air for the alveolar C02 tension to fall, as well as that re- quired to bring about the recovery to the normal when he descends to sea level. The following curve, which is self-explanatory, will illustrate these points. Thus, on Pike's Peak, where the barometric pressure is 459 mm. Hg, the C02 tension after an initial fall took about seven days before it came to its permanent level for that barometric pressure, and fourteen days elapsed after descending from the mountain before the sea-level tension had been regained. The slow nature of these changes, when com- pared with, the rapid changes observed in the experiment with the bags already alluded to (page 358), shows clearly tliat lactic acid can not be responsible for the increase in H-ion concentration in mountain sickness. By exclusion it would appear that the increase in CH is the result of an excess of fixed inorganic acid (H3P04) in the blood dependent on a dis- proportionate excretion of bases by the kidneys during the period of acclimatization to the rarefied air. Other observers aver that the acidosis does not really exist, but that the excitability of the respiratory center itself becomes raised (its threshold lowered), so that it responds more readily to the normal CH of the blood. It has been stated that the increase in excitability of the center is dependent upon the action of the intense light rays at high altitudes — the erythema of the skin, etc., being evidence of this excit- ing action of light. The constant irritation of the skin, these authors say, serves by stimulation of afferent nerves to maintain a hyperexcit- ability of the respiratory center. Others believe that the hyperexcit- ability of the center is a direct result of the maintained 02 deficiency. The balance of evidence, however, stands in favor of the view that the phenomena of mountain sickness depend on changes occurring in the in- organic nonvolatile acids of the blood. The other phenomena of this interesting condition will be discussed elsewhere (page 399). APNEA If a man breathes forcibly and quickly for about two minutes, he will experience no desire to breathe for a further period of about the THE CONTROL OF THE RESPIRATION 363 same duration — he becomes apneic. "When the desire to breathe re- turns, the breathing is at first very shallow, but gradually becomes more marked, until at last normal respiration'is reestablished. If a sample of alveolar air is removed at the time when the desire to breathe returns, it will be found to contain a very small percentage of 02 indicating that for some time previous to the onset of breathing there had been in the alveolar air, and therefore in the blood, so low a percentage of 02 that if 02 deficiency could stimulate breathing, this would have started much earlier than it actually did. A curve showing the results of such an experiment by Haldane is given in Fig. 131. The person may begin to show symptoms of 0, want, such as cyanosis, before the desire to breathe returns, which furnishes strong proof that 0, want itself can not serve as a stimulus to the respiratory center. The failure of the center to act must rather be due to the lowering of the CH consequent upon the removal of CO, from the blood by the forced respiration which preceded the apnea — washing out of the CO,, as it is called. That this has really occurred can readily be shown by estimating the C02 con- tent of a sample of alveolar air collected by having the subject make a forced expiration early in apnea. Extremely low values along with a respiratory quotient (page 547) of about 0.2 are often found, whereas, during the preceding forced breathing while the C02 is being washed out, the quotient is often ten times as great — viz., 2.0. As would be expected, the low 02 percentage present in the alveolar air toward the end of the apneic pause is not without some effect, indi- rect though it may be, 011 the excitability of the respiratory center. This accounts for the fact that the alveolar air, at the moment the de- sire to breathe returns, usually- contains a lower percentage of C02 than the normal, indicating that some nonvolatile acid must have accumulated in the organism so as to raise the CH of the blood, and thus require a lower tension of CO, to overstep the threshold of excitability of the re- spiratory center. In agreement with this explanation it has been found that, if the last two or three forced respirations preceding the apnea are made in an atmosphere of 0, instead of air, so as to fill the alveoli with 02, the apnea can be maintained for a very much longer period; and when the natural desire to breathe returns, the C02 tension of the alveolar air, instead of being below the normal, is above it. The effect of 02 in prolonging apnea must, therefore, be dependent on the fact that it prevents the accumulation in the organism of the unoxidized acids, leaving to C02 alone the function of raising the CH in the blood to the level required to excite the respiratory center. By this means the period during which the breath can be held after breathing 02 is sometimes 364 THE RESPIRATION tFi&txvuA .c % a •f -S o g f I 1 1 5 a fS -c 0 SI 1 ) 1 0 4( ' 1OO J mm. pressure Percentage saturation with oxygen 0 10 20 30 40 50 QO 70 80 90 100 Oxygen pressure Fig. 137. — Upper left hand, percentage saturation of hemoglobin with oxygen at 37" C. cor- responding to oxygen pressures of 0, 10, 20, 40 and 100 mm. of oxygen, respectively. Upper right hand, the same spaced with the oxygen pressure as the abscissae. Lower figure, dissociation curve representing the equilibrium between oxygen, oxyhemoglobin (red) and reduced hemoglobin (purple). (From Joseph Barcroft.) RESPIRATION BEYOND THE LUNGS 383 conditions readjusted, the manometer closed off from the ouside air, and the apparatus again shaken so that the ferricyanide mixes with the hemoglobin solution. This drives off all the 02 from the oxyhemoglobin solution, and, therefore, raises the pressure in the atmosphere of that bottle so that the clove oil moves to the opposite side of the manom- eter, the degree of displacement being proportional to the amount of oxyhemoglobin. We have now all the necessary data for estimating the relative amounts of reduced hemoglobin in the hemoglobin solution as removed from the tonometers, for it is plain that the second estimation, as described above, tells us how much oxyhemoglobin might have been formed had all the hemoglobin been saturated and the first one, how much 02 had yet to be taken up by the original hemoglobin solution to produce saturation. The Dissociation Curve. — The next step is to plot the results obtained from the various hemoglobin solutions in the form of a curve. This is known as the dissociation curve of hemoglobin. It is plotted with the relative percentages of reduced and oxyhemoglobin in each of the solu- tions along the ordinates, and the partial pressures of 0, in millimeters of mercury to which they were exposed along the abscissae. The -curve thus drawn is exactly of the same shape as that Avhich would be pro- duced if we were to place the tonometers in a row at distances from one another corresponding to the partial pressure of 02 which each con- tained, and then to mark on each tonometer the. relative amounts of reduced and oxyhemoglobin found in the solutions after shaking. A line joining these marks on the tonometers would then exactly corre- spond to the curve drawn by the method described above. This will be clear from the accompanying figure from Barcroft's book (Fig. 137). In such a chart the space below the curve can be taken to represent the percentage of oxyhemoglobin (red in chart), and that above it of reduced hemoglobin (blue in chart), at the varying partial pressures of 02 which are indicated along the abscissas as being contained in the at- mosphere of the tonometers, and which must be proportional to the partial pressure of 02 in the solution in which the hemoglobin is dis- solved. Difference between Curves of Blood and Hemoglobin Solutions. — The curve obtained from pure hemoglobin solutions is very far, however, from clearing up the problem as to how the blood absorbs and discharges 02. On the contrary, it makes this problem appear all the more difficult, for, according to the curve (Fig. 137) the hemo- globin is already more than half combined with 02 at a partial pressure of this gas of no more than 10 mm. Hg, which means that in the low partial pressure of 02 existing in the capillaries the oxyhemoglobin, in- 384 THE RESPIRATION stead of readily yielding up its load of 02, would greedily retain prac- tically the whole of it. The curve, in other words, would satisfactorily explain why hemoglobin should readily absorb 02 from the alveolar air, but would fall far short of explaining how this 02 is readily released when it is required in the tissues. Obviously there is some artificial con- dition present in the above experiment which can not obtain in the nat- ural environment of the blood. 10 $.0 30 W 50 60 70 90 in Fig. 138). The results are very different. At a partial pressure of 02 of about 60 mm. Hg— that is, a lower pressure than exists in the lung alveoli (100 mm.)— the blood becomes nearly saturated with 02, whereas at pressures below 50 mm. it -readily loses 02, so that at 10 mm! there is nearly complete reduction. The question is: What are the environmental conditions under which the hemoglobin in the blood so alters its combining power for 02 as to '00 90 80 70 60 40 30 20 20 30 40 50 60 70 80 Fig. 139. — Dissociation ' curves of hemoglobin. 90 TOO Ordinates — Percentage saturation of hemoglobin. Abscissa— Tension of oxygen in mm. of mercury. I. Dissociation curve of hemoglobin dissolved in water. II. Dissociation curve of hemoglobin dissolved in 7% NaCl III. D^sociation curve of hemoglobin dissolved in 9% KC1 Temperature 37-38° C. (From Joseph Barcroft.) produce such a difference in the dissociation curve? By experimenting with hemoglobin solutions, three such factors have been found to come into play: (1) the presence of inorganic salts, (2) the hydrogen-ion con- centration (C02 tension) of the solution, and (3) the temperature. If hemoglobin is dissolved in water containing the various salts of plasma in the same proportion as in blood (artificial plasma), the dissociation curve will be found to change so as to resemble that of blood (Fig. 139)'. 386 THE RESPIRATION Since the plasmas of different animals contain different proportions of salts, the artificial plasma required to secure the result is not always the same. It differs, for example, for the dog and man. Potassium salts are particularly efficient in causing hemoglobin to absorb 02. The in- fluence of varying hydrogen-ion concentrations of the solution may be conveniently studied by adding varying percentages of C02 to the gas mixture in the tonometers, when it will be found that the curve be- comes lowered in proportion to the amount. of C02 present. This is shown in Fig. 140. The effect of temperature on the dissociation curve is twofold: (1) on the rate with which equilibrium is established at the given partial pres- 0 10 20 30 40 00 CO 70 80 90 100 Fig. 140 — Dissociation curves of human blood, exposed to 0, 3, 20, 40 and 90 mm. CO2. Ordinate, •percentage saturation. Abscissa, oxygen pressure. (From Joseph Barcroft.) sure of 02, and (2) on the position of the curve; the lower the tempera- ture, the higher the curve. The Rate of Dissociation. — Though it is now clear that the three con- ditions— namely, saline content, CH, and temperature — are capable of altering the dissociation curve of a pure hemoglobin solution so as to make it correspond with that of blood, this does not entirely solve our problem, for we have yet to show how the cooperation of these forces renders it possible for the rate at which hemoglobin takes up 02 in the lungs to correspond exactly with that at which it gives up its 02 to the tissues. To study this problem a somewhat different kind of experiment must be undertaken. The hemoglobin solution is placed in a tube and the gas mixture slowly bubbled through it, samples of the solution being removed at intervals for analysis in the differential blood- KESPIRATION BEYOND THE LUNGS 387 gas apparatus. To obtain the rate of oxidation, a mixture of N2 or H2 and 02 is bubbled through the blood with the partial pressure of the 02 the same as that which obtains in alveolar air — namely, about 95-100 mm. Hg; and to obtain the rate of reduction pure N2 or H2 gas is bub- bled through. The rates of reduction or of oxidation as thus determined are then plotted in curves constructed with the percentage saturation of the 100, 1?° Oxidation 17-5° C. no CO, Reduction 100 Oxidation 37-5° C. no C02 Reduction 80 60 20 V Oxidation 37-5° C. + 40 mm. pressure ofC02 Reduction Fig. 141. — Curves showing relative rates of oxidation and reduction of blood as influenced by temperature and tension of CO2. Ordinates — Percentage saturation. Abscissae. — Time in minutes. Reducing gas, hydrogen. Oxidizing gas, oxygen. A, temperature 17.5° C., with no CO2. B, temperature 37.5° C., with no CO2. C, temperature 37.5° C., but the O2 and H contained 40 mm. Hg pressure of CO2. (From Joseph Barcroft.) 388 THE RESPIRATION hemoglobin on the ordinates and the time in minutes along the abscissas (Fig. 141). Even if we use blood in this experiment and therefore make certain that the hemoglobin is acting in the presence of the proper pro- portion of salts, we shall find, as Fig. A shows, that at room temperature the rate of oxidation is very much greater than the rate of reduction. If now we repeat the observation at a temperature of 37° C., the two curves come more nearly to correspond, but still the rate of reduction is slower than that of oxidation. If in a third experiment, besides having proper temperature and chemical conditions, we produce the oxidation and reduction in the presence of a partial pressure of C02 of 40 mm., which corresponds to that of the arterial blood, we shall find that oxida- tion becomes a little slower, whereas reduction is further quickened. Indeed the two curves, as seen in C in the figure, come practically to correspond, indicating that the environmental conditions under which hemoglobin combines and gives off 02 in the blood are exactly adjusted. One word more with regard to the influence of CH. Its effect in flat- tening out the curve, especially at the lower partial pressures of 02, indicates that when a high CH is present, the blood will very readily part writh its 02 supply. Now, the most significant application of this fact is that high concentrations of H ion will occur just exactly where it will be of benefit — namely, in the capillaries (because of the C02 and lactic acid produced by the tissues) . Some doubt has, however, recently been thrown on the importance of this factor. Since, as we have seen, hemoglobin absorbs 02 according to chemical laws, it will naturally be asked not only why the dissociation curve flat- tens out while yet maintaining the shape of a right-angled hyperbola, as by the action of acids or an increase in temperature, but also why it should change its shape when salts are also present. The explanation offered by Barcroft and his pupils is that the changes depend on the fact that hemoglobin being a colloidal substance, its molecules undergo processes of aggregation under the conditions referred to above, and therefore cause the reaction to become of a different type from that represented by the equation Hb02 ^± Hb + 02. As has been pointed out by Bayliss, although such an explanation might suffice to explain the flattening out of the curve, it fails to explain the change in its shape; for, according to the laws of mass action, such a change could occur only if molecules of a different type came to take part in the reaction. Dissociation Constant. — Notwithstanding these criticisms, it is of con- siderable practical importance to know that an equation exists from which the entire dissociation curve can be plotted by making only one determination of the relative amounts of oxy- and reduced hemoglobin at a particular tension or partial pressure of oxygen. This equation is as RESPIRATION BEYOND THE LUNGS 389 y Kxn follows: = -TJ — -_ n , where y equals the percentage saturation of hemoglobin with 02, x the 02 tension, and K and n are constants, K being the equilibrium constant and n the average number of molecules of hemoglobin supposed to exist in each aggregate. When this equation is applied to human blood, the value of n remains unchanged and is given as 2.5, so that by transposition we are enabled y to find the value of K as follows: K = — -r^ r . If we find the value xn(100-y) of K by measuring the relative saturation of the blood with 02 at one pres- sure of this gas, then by changing the value of x to correspond to other 02 pressures, we can find all positions of the curve for a given sample of blood. An important practical application of this method is found in the determination of the (7H of blood, for, as we have seen, the dissociation curve becomes lowered in proportion to the concentration of hydrogen ions. The acidity of a sample of blood can therefore be found by com- parison of its dissociation curve, as plotted from the values found for K, with that of normal blood to which known quantities of acid have been added. When the curves correspond, the bloods must contain the same amounts of acid, other things being equal. In brief, then, the re- action of the blood is proportional to the value of K. When this is low, it indicates that the blood is taking up an abnormally low percentage of its possible load of 02 at a given pressure of 02, and that the acidity is greater than normal ; when K is high, for the same reason the acidity must be low. In determining K for the blood as it exists in the body, it is necessary that it should be subjected to the same tension of C02 as obtains in the blood vessels. K will then be proportional to the CH of the living blood. This condition would be impossible to fulfil in drawn samples were it not for the fact that we can place in the tonometer an atmosphere con- taining the same partial pressure of C02 as is found in the alveolar air. Since this value varies in different individuals, it must be separately ascertained in each case (see page 344). As determined with these modifications, K has been found to vary in healthy men between 0.000212 and 0.000363 (ten individuals). When acid substances appear in the blood, as in acidosis, K becomes extremely low; thus, in one case suffering from acidosis with dyspnea, it was found a few hours before death to be only from 0.000082 to 0.00011. Similarly K becomes low in the acidosis associated with mountain sickness, and it is said to be raised after taking food that is rich in alkali.* *When K is found to be normal, the blood is said to be mesectlc; where K is low, it is said to be myonectic; and when K is high and the acidity is therefore small, it is said to be pleonectic. CHAPTER XLV RESPIRATION BEYOND THE LUNGS— Cont 'd THE MEANS BY WHICH THE BLOOD CARRIES THE GASES In the foregoing account of the physiology of the blood gases, empha- sis is placed on the tension under which the gases exist rather than on the total amount of each gas present in the blood. This has been done because the exchange of gases between alveolar air and blood and be- tween blood and tissues proceeds according to the laws of gas diffusion, which are of course dependent upon differences in gas pressure or tension. Something must now be said regarding the amount of the gases. This may be measured either by physical or by chemical methods. In the former, a measured quantity of blood is received into an evacuated glass vessel, which is then attached to a mercury pump, by which the gases are -sucked out of the blood and transferred, by suitable manipulations of stopcocks, to a graduated tube, in which they are then analyzed by chemical means. The principle of the chemical method has already been described in connection with the measurement of oxygen in hemoglobin solutions (see page 382). A measured quantity of blood, kept free from contact with the air, is transferred under some weak ammonia solution to one of the blood-gas bottles of the blood-gas differential manometer, and a few drops of a saturated solution of potassium ferricyanide is placed in the pocket of the bottle. After the blood has been laked and temperature conditions adjusted, the ferricyanide is mixed with the blood solution, thus causing the 02 to be quantitatively displaced. From the increased pressure produced in the manometer the amount of 02 can readily be computed. To determine the C02 of the blood, the bottle is now removed from the manometer and a few drops of a saturated solu- tion of tartaric acid placed in the pocket. When this is mixed with the deoxygenated blood mixture, after the usual adjustment for tempera- ture, the pressure caused by the evolved C02 is recorded and the amount present calculated. The results of the analysis are expressed as the number of cubic centi- meters of gas present in 100 c.c. of blood — the volume percentage, as it is called. The following are approximate percentage values: 390 RESPIRATION BEYOND THE LUNGS 391 OXYGEN CARBON DIOXIDE TOTAL GAS Venous blood 12 48 60 Arterial blood 20 ' 40 60 The estimation of the amounts of the gases, although of little value in connection with the physiology of gas exchange, is very important in supplying information regarding the respiratory activities of the various organs and tissues. Just as we determine the total respiratory exchange of an animal by measuring the differences in 02 and C02 in inspired and expired air, so may we determine the degree of tissue respiratory ex- change by analysis of the gases in blood removed from the artery and vein of the tissue. It should be clearly understood, however, that it is not the percentage but the total amount of the gases that must be con- sidered, and that it is therefore necessary to know the volumes of blood- flow as well as the percentage of the gases. Something will be said later of the results of such investigations (see page 393). At present we are concerned with the manner in which gases are carried in the blood. The 02, as we have seen, is carried by the hemo- globin, some being also in a state of simple solution in the plasma. The C02, which it will be noted is present even in arterial blood in con- siderably greater amount than the 02, is partly combined with alkali to form bicarbonates. The alkali available for this purpose varies from time to time according to the amount of other acid substances present. Since these are stronger acids than carbonic, any increase in their amount (acidosis) causes displacement of some of the C02, thus bring- ing about, as we have seen, a relative increase in free C02 in the blood and therefore raising the CH. What particularly interests us here is the agency by which the com- bined C02 is carried in the blood. If blood is exposed to a full atmos- phere of C02, it will take up as much as 150 per cent of the gas — that is, between two and three times the amount ordinarily present in it. It has therefore a great reserve capacity for C02. A greater propor- tion of the CO2 is carried in the plasma than in the corpuscles; but if plasma (or serum) is exposed in a vacuum, all of the C02 present in it will not be evolved. When blood itself is similarly exposed, on the other hand, all the C02 is given off. To liberate all of the C02 from plasma in vacua, some acid must be added, from which it has been in- ferred that blood corpuscles act like weak acids. It is commonly stated that hemoglobin or some constituent of blood is capable of freeing C02 from solutions of sodium carbonate, but the recent work of Buckmaster32 shows that this is not the case. The decomposing power of blood is caused by the development of acidity in the shed blood and any similar power that the corpuscles may exhibit is due to a discharge from 392 THE RESPIRATION them into the plasma of acid radicles. At least it has been found that the alkalinity of the plasma increases when C02 is bubbled through blood, this increase in alkalinity being interpreted as the result of the migration of acid radicles into the corpuscles. This would lead us to expect that under the opposite conditions (i. e., in vacuo) acids would leave the corpuscles. Proteins are amphoteric substances — that is, they combine with acids or alkalies — which would lead us to expect that they would be capable of absorbing some C02. That this is the case, particularly for hemo- 75 70 65 60 50 40 A 30 40 60 70 vn. rtvm. 80 90 Fig. 142. — Curve of CO2 tension in blood. For description, see text. (From Christiansen, ' Doug- las and Haldane.) globin, has been shown by comparing the C02-combining powers of water and a solution of pure hemoglobin. Attempts have been made to determine the relative amounts of C02 carried by these various agencies in the blood. The following is an ex- ample of such a table: In simple solution in plasma and corpuscles ] a in corpuscles 6 As sodium bicarbonate } ^ in plagma 12 In combination with hemoglobin 7.5 In combination with proteins of plasma 11 1.9 c.c. 8 " 8 1 .0 } 18' "I } 19.3 " .8 J (Loewy.) 40.0 RESPIRATION BEYOND THE LUNGS 393 The power of blood to absorb C02 at various tensions of this gas, as determined in the same way as for 02 (see page 380), has shown that saturation of the hemoglobin with 02 distinctly diminishes the C02- carrying power of the blood. This is shown in the accompanying curves The various tensions of C02 are given along the abscissae and the volume per cents of C02 taken up by the blood on the ordinates. The upper curve is drawn from results obtained when the blood was shaken with C02 in the presence of hydrogen, and the lower, when in the presence of air. (The dotted curve may be disregarded.) The line AB drawn between the two curves represents the absorption of C02 by the blood within the body. At a tension of 40 mm. C02 — that present in alveolar air (see page 356) — A stands in arterial blood at about 52 vols. per cent; and at a pressure of 62 mm. — possibly present in the tissues — B stands in venous blood at about 67 vols. per cent. The C02-containing power would be 7 per cent lower (i. e., 60 vols. per cent) in blood saturated with 02 at the latter pressure. The oxygenation of blood in the lungs, therefore, helps to drive out the C02; and conversely, its deoxygenation in the tissues enhances its power of absorbing this gas. Having shown how the blood transports its charge of 02 from the lungs to the tissues, we may now proceed to study the call for 02 by the tissues, and in this connection we have to consider (1) the amount of .02 which they require under varying conditions of rest and activity, and (2) the mechanisms by which their varying demands are met. THE OXYGEN REQUIREMENT OF THE TISSUES In order to ascertain the average 02 requirement of the different tis- sues of the body, it is necessary to adopt as a standard of measurement the amount of 02 in c.c. absorbed per gram of tissue per minute. To ob- tain it we must know: (1) the weight of the particular organ or tissue under investigation; (2) the bloodflow through the vessels of the organ in c.c. per minute; and (3) the different percentages of 02 in the arterial and venous blood of the tissue. It would be beyond the scope of this book to review in any detail the many experimental investigations which have been undertaken in this connection. A few of the most recent and important results are given in the accompanying table from Halli- burton's Physiology: In the order of their oxygen requirements, or the coefficient of oxida- tion, as it is called, the tissues may be divided into four groups ; glandular, muscular, connective, and nervous. The nervous tissues should possibly stand above the connective, but very little is known regarding their oxygen consumption, although it appears that this is quite low (Hill and 394 THE RESPIRATION ORGAN CONDITION OP REST OXYGEN USED PER MINUTE PER GRAM OF ORGAN CONDITION OF ACTIVITY OXYGEN USED PER MINUTE PER GRAM OF ORGAN Voluntary muscle Nerves cut. Tone absent 0.003 c.c. Tone existing in rest Gentle contraction 0.006 c.c. 0.020 c.c. Active contraction 0.080 c.c. Unstriped muscle Besting 0.004 c.c. Contracting 0.007 c.c. Heart Very slow and feeble contractions 0.007 c.c. Normal contractions Very active 0.05 c.c. 0.08 c.c. Submaxillary gland Nerves cut 0.03 c.c. Chorda stimulations 0.10 c.c. Pancreas Not secreting 0.03 c.c. Secretion after injec- tion of secretin 0.10 c.c. Kidney Scanty secretion 0.03 c.c. After injection of diuretic 0.10 c.c. Intestines Not absorbing 0.02 c.c. Absorbing peptone 0.03 c.c. Liver Suprarenal gland In fasting animal Normal 0.01 to 0.02 c.c. 0.045 c.c. In fed animals '0.03 to 0.05 c.c. Nabarro). It is of course necessary in making these comparisons to secure the coefficient of oxidation both when the tissue is at rest and when it is thrown into varying degrees of activity. Special attention has been devoted to the requirements of skeletal muscle, heart muscle and the salivary glands. Skeletal Muscle.^In observations on skeletal muscle, Verzar (cf. 27) isolated the gastrocnemius muscle of the cat, and without disturbing its blood supply collected samples of blood by introducing a 1 c.c. pipette into a branch of the saphenous vein. Activity was produced by throw- ing the muscle into tetanus by the application of an electrical stimulus to the sciatic nerve. During its contraction the muscle lifted a weight, so that it did about 70 gram-centimeters of work at the beginning of each period of tetanus. The velocity of bloodflow was determined by the rate at which the blood flowed along the pipette, and the 02 consump- tion, by the difference in percentage of 02 in the venous and the arterial blood. These measurements were made: (1) before contraction, (2) dur- ing contraction, and (3) after contraction. It was found that during the tetanus the 02 consumption in some cases was greater than during rest, while in others it was actually less,, but in every instance a great increase in 02 consumption followed the tetanus — that is, the call for 02 continues for some time after the actual work has been performed. This result RESPIRATION BEYOND THE LUNGS 395 shows that the contraction is not dependent upon oxidation, but that the oxidation occurs after the contraction is over. The mechanism involved in muscular contraction can not therefore be analogous with that by which energy is liberated in a steam engine by the oxidation of the coal. The mechanism must rather be like that of a spring, which becomes un- wound during the muscular contraction and requires 02 for its rewinding. Interesting results corroborative of these conclusions have been se- cured by observations on the heat production of isolated muscles. It was found that heat production occurred after a single shock to the muscles, not only during the contraction, but for a considerable period after it, provided 02 was present. In the absence of 02 this recovery was either greatly delayed or entirely abolished. Such results favor the view that 02 is used largely in the processes whereby the muscles, "like an engine charging an accumulator, synthesize substances con- taining a considerable amount of potential energy, which again, like the accumulator, it discharges when appropriate stimuli are applied" — (L. V. Hill, cf. 27). One immediately thinks of lactic acid in connection with these interesting results, for, as has already been stated, Hopkins and Fletcher29 have shown that this acid is produced in the absence of 02 in excised frog muscles, but when 02 is present, it is either not pro- duced or, if so, quickly disappears. Heart Muscle. — Another muscle that has been thoroughly investigated in this connection is that of the heart. The gaseous exchange has been studied both on isolated heart preparations and by examining the ex- change in the lungs of a combined lung and heart preparation. The most important investigations by the first of these methods are those of Rohde (cf. 27), who arrived at the very important conclusion that the 02 taken in by the heart muscle varies directly with the maximal ten- sion set up in the heart by the contraction. This tension was measured by placing a rubber bag in the ventricle and distending it with water at a known pressure. By altering the initial pressure and by observing the pulse rate, it was found that the 02 used by the heart depends on the product of the pulse frequency and the maximal increase in pressure produced by each cardiac contraction; or, in the form of an equation: Q = a constant quantity ; where Q is the oxygen used, T the maximal NT increase of pressure at each beat, and N the frequency of the pulse. It should be pointed out, however, that constancy in the product of the above equation does not hold under abnormal conditions of the heart- beat. For example, when the pressure in the heart is very high, the amount of 02 required begins to go up out of proportion, indicating that 396 THE RESPIRATION • the heart is becoming overtaxed — that it is losing its efficiency. The same result occurs when the heart is dying, and when depressing drugs are used, such as chloral hydrate, potassium cyanide, veratrine, etc.. Some other drugs, however, such as epinephrine, do not cause altera- tion in the ratio, nor does vagus stimulation. Of course when the vagus is stimulated, the 02 consumption in a given period decreases because the heartbeats are slowed ; but the absorption of 02 is not increased rela- tively to the slowing of the heart. Glands. — Most work has naturally been done on the most accessible gland — the submaxillary. By stimulating the secretory nerve of this gland (the chorda tympani) in the dog, it has been found that, whereas the more abundant secretion lasts only so long as the stimulus is ap- plied to the nerve, the 02 consumption is increased to several times that of rest, and remains increased for a considerable period after the stimulus has been removed. Accompanying the increased functional activity in such structures as muscles, there is a very marked increase in bloodflow due to vasodilatation, which, in part at least, is dependent upon the secretion into the blood of some substances resulting from the glandular activities, and is not entirely due to the action of vasodilator nerve fibers. Similar results have been obtained in the case of the pancreas when excited to secrete by the injection of secretin (see page 425). Under such conditions, the oxygen consumption has been observed to increase about fourfold and to be accompanied by a dilatation of the gland. The work on the kidney has been especially interesting, because it has been found that increased activity, which of course is measured by the rate of urine excretion, is not always accompanied by increased consumption of oxygen. When diuresis is produced by injecting Ring- er's solution into the circulation, a great increase in urine outflow may occur without any change in oxygen consumption; whereas, on the other hand, when a diuretic such as sodium sulphate or caffeine is used, the oxygen consumption increases enormously. Regarding the other tissues and organs, the 02 consumption of the lungs and brain appears to be small. It is a very significant fact, how- ever, that the higher cerebral centers are extremely sensitive to depri- vation of 02. The Blood. — In the blood itself, a certain amount of oxidation goes on because of the presence of leucocytes. This oxidation becomes con- siderable in the blood of animals rendered anemic by the injection of phenyl hydrazin. A thorough investigation of the cause of this greater oxidation has shown it to be owing, not to an increase in nucleated corpuscles, but to the presence of the young unnucleated red blood RESPIRATION BEYOND THE LUNGS 397 corpuscles, which appear in large numbers in the blood under these con- ditions. A similar increase in blood oxidation occurs during posthemor- rhagic anemia, the rate of oxidation running parallel with the rate of regeneration of the red corpuscles. The Mechanism by Which the Demands of the Tissues for Oxygen Are Met There are two possible methods by which this may be brought about: (1) by a change in the CH or the saline constituents or the temperature of the plasma, so that the hemoglobin more readily delivers up its load of 02; and (2) by an increase in the mass movement of blood through the vessels of the acting tissue. Kegarding the first of these possibilities, there is no doubt that acids are produced during metabolism of acting tissues. As we have seen, when muscles contract in the presence of an abundance of 02, C02 is produced in large amounts, and when they contract in a deficiency of 02, sarcolactic acid. In the submaxillary gland, too, it has been possible to show that the CH of the venous blood, as measured by the value of K of the dissociation curve of hemoglobin, becomes distinctly increased dur- ing glandular activity. That this increase in CH will dislodge 02 we have already seen (page 386). As to the possible influence of local changes in temperature and in saline constituents of the plasma, nothing can at present be said. Regarding the second possibility, vasodilatation may be dependent either upon the action on the blood vessels of nerve impulses coming along vasomotor nerves, or upon the production by the active tissue of vasodilating or depressor substances (see page 243). Much evidence has been accumulating in recent years which tends to show that such depressor substances are produced, and they may be either (1) acids, or (2) organic bases of a similar nature to /3-imidazolylethylamine (hista- mine). This latter substance is of considerable physiologic interest be- cause of its close relationship to one of the main amino acids of the protein molecule — namely, histidine (see page 604). Its effect in pro- ducing vasodilatation is extraordinary. Thus, half a milligram of the drug injected intravenously into a monkey will lower the mean arterial pressure by fifty per cent. But before such an hypothesis can be entertained, it is necessary to show that, independently of nerve impulses, the blood vessels of an acting organ may dilate. The best evidence has been secured by studying the effects of stimulating with epinephrine the cervical sympathetic nerve to the submaxillary gland of a cat. The gland cells become more active, 398 THE RESPIRATION and dilatation of the artery occurs, although on blood vessels alone epinephrine in similar dosage produces constriction. Of course in show- ing that local chemical products of activity serve as the excitant of local dilatation, we do not mean to imply that the vasodilator fibers going to the blood vessels are of no use. Indeed we know that such fibers do be- come active in the case of a salivary gland whose cells have been para- lyzed by atropine, but it is a significant fact that this dilatation is of rela- tively short duration, whereas that produced by glandular activity lasts for some time. The suggestion seems therefore not out of place that un- der normal conditions the initial dilatation of an acting gland may be brought about through nervous stimuli, but the later dilatation is main- tained by metabolic products. CHAPTER XLVI THE PHYSIOLOGY OF BREATHING IN COMPRESSED AIR AND IN RAREFIED AIR In the application of a knowledge of the physiology of respiration to the investigation of disease, a group of conditions arises in which con- siderable interference with physiologic mechanisms occurs, not as a result of disease, but of changes in the atmospheric environment. The regula- tion of the functions of respiration depends very largely on changes in the physical and chemical properties of the alveolar air, so that it is to be expected that similar changes in the atmosphere will have a marked influence on the respiratory activity and on the general well-being of the animal. The most thoroughly investigated of these conditions are those which develop in rarefied and compressed air. Either condition can be pro- duced experimentally in the laboratory by the use of air-tight chambers (pneumatic cabinets) and suitable pumps, although most of the im- portant work on the effects of rarefied air has been conducted at high altitudes, where the barometric pressure is low. MOUNTAIN SICKNESS This condition depends primarily on disturbances in the control of the respiratory function, and it is on account of the useful information con- cerning the nature of these functions, rather than because of the so-called disease itself, that so much attention has been devoted to its investiga- tion during recent years. The disturbances produced by the rarefied atmosphere develop rather quickly, but after some time they gradually disappear, indicating that the organism has acclimated itself — that is, the compensatory mechanisms have come into play to bring the respira- tory control back to normal. When animals are placed in pneumatic cabinets from which some of the air is pumped out, most of the imme- diate symptoms observed in mountain sickness occur, but it is usually impracticable to continue the observations for a sufficient length of time to allow the compensating mechanisms to develop. Because of their great value in revealing the nature of the respiratory hormone, many of the results of the recent investigations on mountain 399 400 THE RESPIRATION sickness have been given elsewhere in this volume (page 360), where the general symptoms are also described. In this place we shall consider very briefly some of the more general aspects of the condition, and, mare particularly, the nature of the adaptation that occurs. All of the symp- toms are essentialy dependent upon lack of oxygen. Cyanosis is com- mon and the symptoms are much the same as those of coal-gas poisoning. Not only, does this deficiency of oxygen cause acid substances to appear in the blood, thus raising the CH and stimulating the respiratory center, but it allows other poisonous materials to accumulate. These act on the various nerve centers, producing symptoms which vary in different in- dividuals according to their relative susceptibilities. In some, the diges- tive centers are affected and nausea and vomiting occur; in others, the higher cerebral centers are affected, causing depression and general men- tal apathy, great drowsiness, muscular weakness, or it may be mental excitement and loss of self-control. The susceptibility of different individuals also varies according to the amount of previous experience in mountaineering and the type of breath- ing. Much of the value of previous experience and training depends on the ability to perform muscular effort economically; to adjust the effort to the available oxygen supply without permitting unoxidized harmful products to accumulate in the body. It often happens that no symptoms appear so long as the person is at rest, but immediately do so whenever any muscular effort demands a much more abundant oxygen supply. The type of breathing that best withstands the rarefied air is slow and deep, rather than rapid and shallow. The reason for this is of course that much more of the outside oxygen gets into the alveoli in the former case than in the latter, the dead space being practically constant. The following figures taken from observations on three different individuals will illustrate the importance of this factor. C.C. PER NO. OF RES- HEIGHT IN METERS RESPIRATION PIRATIONS AT WHICH SYMP- PER MINUTE TOMS OCCURRED Subject 1 270 20 3300 " 2 440 14 6000 " 3 700 8 6500 (From Halliburton.) After living for some time in the rarefied air and quite independently of training in the efficient performance of muscular work, adaptation occurs, so that the symptoms pass off. The essential feature of this adap- tation is increased absorption of 02 into the blood. Three mechanisms have been described as responsible for this effect: (1) increase in the ten- sion of 02 in the alveolar air; (2) assumption by the pulmonary epithelium BREATHING IN COMPRESSED AND IN RAREFIED AIR 401 of the power of secreting 02 into the blood; (3) increase in the erythrocytes and hemoglobin of the blood. The increased alveolar 02 tension is a result of the more rapid breathing brought about by the increased CH of the blood. If no adaptation occurred, the 02 tension at 10,000 feet would be 59 mm. and at 15,000 feet, 33.8 mm. Actual observations on men, how- ever, gave at 10,000 feet a tension of 65 mm. and at 15,000 feet, 52 mm. The evidence for an increased secretory activity of the pulmonary epithelium depends on observations made by Haldane and his cowork- ers,33 who found that blood collected from the finger of a man living on a high mountain is brightly arterial, whereas if this same blood is shaken in a flask with alveolar air from the man from whom it was taken, it will become darkly venous. To account for this difference it is believed that the pulmonary epithelium forces 02 into the blood contrary to the laws of diffusion. A more exact proof was sought for by comparing the relative amounts of O2 and CO that blood would take up (1) when exposed outside the body and (2) while in the blood vessels. Carbon monoxide has a very great avidity for hemoglobin, so that if blood is shaken in a flask with air containing 0.07 per cent of this gas, colorimetric measurement will show an equal mixture of oxy- and carboxy-hemoglobin. Since carbon monoxide is destroyed with extreme slowness in the body, it is possible by causing a man to breathe a mixture of it in air to determine, in a sample of drawn blood, whether as much carboxy-hemoglobin has been formed as in vitro. If so, the O2 tension in the blood must equal that in the alveoli; if less carboxy-hemoglobin should be formed, it would indi- cate that a higher tension of 02 exists in the blood. This latter is the re- sult which Haldane states he has secured. In one experiment, for ex- ample, when blood was shaken outside the body with 0.04 per cent CO2, the amount of carboxy-hemoglobin formed was 31 per cent of the whole hemoglobin. >When the same mixture was inhaled for three or four hours the percentage of carboxy-hemoglobin in the blood rose only to 26 per cent, which would correspond to an O2 tension of 25 per cent of an atmos- phere, whereas even at sea level the tension of 0.2 in the alveolar air can not be above 15 per cent of an atmosphere. The constant low tension of 02 in the plasma stimulates the red blood corpuscles and the percentage of hemoglobin to become markedly in- creased after residence for some time in high altitudes. At first this is due to a concentration of the blood by a diminution in plasma, but grad- ually the blood-forming organs become excited and an actual increase in the total amount of hemoglobin occurs. In the light of these facts it is interesting to compare the average number of red corpuscles in the blood of inhabitants living at different altitudes. 402 THE RESPIRATION HEIGHT ABOVE SEA RED CORPUSCLES (METERS) (PER C.MM. BLOOD) • Christiania Zurich Davos Arosa Cordilleras 0 412 1560 1800 4392 4,970,000 5,752,000 6,551,000 7,000,000 8,000,000 (From Starling.) COMPRESSED-AIR SICKNESS; CAISSON DISEASE; DIVER'S PALSY Divers and caisson workers are susceptible to peculiar symptoms. These are frequently of sufficient severity to cause death, but may be so mild as almost to escape notice. They first appear, not when the worker is subjected to the high pressure, but after he has come back to atmos- pheric pressure.* While in the compressed air the worker as a rule suffers no discom- fort. A stuffiness may be felt in the ears and temporary giddiness ; the respiration and pulse rate may become slow and frequency of micturition may be noticed, but none of the symptoms of disease appear until after the caissonier or diver has been decompressed (after he has returned to atmospheric pressure), the exact time of their onset being either imme- diately after decompression or af the end of several hours. The worker may have returned home and spent the evening feeling perfectly well until he went to bed, when symptoms supervened which may include mus- cular and joint pains, vertigo, embarrassed breathing, subcutaneous em- physema and hemorrhages, pains in the ears and deafness, vomiting, perhaps hemoptysis and epigastric pain. These symptoms usually pass off after some hours but the arthralgia and myalgia sometimes persist for a considerable time. In the more severe cases the first symptom is severe pain in the mus- cles and joints, quickly followed by motor paralysis, so that the patient falls and is likely to become unconscious. The pulse is almost imper- ceptible, the respiration is labored, sometimes even asphyxial, the face cyanosed, and the surface of the body cold. Many of the cases are fatal ; indeed, death may be almost instantaneous. Such cases are common in careless diving when the divers, to return the more quickly, screw up the outlet valve in their helmets so as to fill their suits with air, which car- *A caisson is a steel or wooden chamber sunk in water and prevented from filling by means of compressed air. For the passage of the workmen and of material, into and out of the caisson, the latter is connected with a second smaller chamber fitted with air-locks and decompressing cocks. ' A diver works in a waterproof suit, the head being enclosed in a copper helmet connected by hose with air pumps. Every 10 meters or 33 feet of water corresponds to one atmosphere pressure (15 pounds to the square inch), so that at this depth the total air pressure in a caisson, or in a diver's helmet, would amount to 30 pounds to the square inch, that is, + 1 atmosphere. BREATHING IN COMPRESSED AND IN RAREFIED AIR 403 ries them to the surface, where they .decompress themselves by opening the valve. Autopsies of persons dead of caisson disease have shown, as a rule, intense congestion of the viscera, hemorrhages in the spinal cord and brain, and ecchymoses on the pleura and pericardium. In some cases interlobar emphysema of the lungs and laceration of the spinal cord and brain have been noted. The Cause of the Symptoms The cause for the symptoms is not, as was at one time supposed, that the pressure drives the blood from the peripheral into the deep regions of the body, including the nerve centers. Such a process is impossible, because the fluids of the body — and all tissues, even the bones, are full of fluid — are incompressible. Pressure applied to any part of the body will be immediately distributed equally to every other part. If this were not so, life would be impossible during any variation of atmospheric pres- sure. It is now clearly established that all the symptoms of caisson disease are due to decompression, and not, in the slightest degree, to the mechan- ical effect of the pressure itself (Paul Bert, Leonard Hill and Macleod34). When an animal is under pressure, its tissue fluids dissolve a large amount of gas. They absorb it in obedience to the law of solution of a gas in a fluid, which -states that the amount of gas dissolved in water is directly proportional to the partial pressure of that gas in the atmos- phere; at two atmospheric pressures twice as much gas will pass into solution as at zero pressure (Dalton's law). So long as the gas is in simple solution, it does not in any way change the physical condition of the blood and tissue fluids. If, however, the animal is suddenly decom- pressed (i. e., the pressure of air surrounding it is reduced to zero), the dissolved gas will be so quickly thrown out of solution that bubbles of it are set free. These bubbles act as air emboli, sticking in the pulmonic capillaries or blocking up a terminal artery in the brain; or they may be large and tear the capillary wall and so lead to hemorrhage. If these bubbles are produced in the posterior spinal roots, intense pain results; if in the anterior, motor paralysis. Frothing of the blood in the heart im- pedes the action of the organ and death soon follows. The following experiments furnish proof of this explanation: A frog was placed in a small steel chamber connected with a cylinder of com- pressed air and provided with two windows by which a strong arc light could be passed through the chamber. The web of the foot was stretched on a wire and fixed so that the small blood-vessels could be seen by apply- ing a microscope to the outside of the window. After carefully observing the circulation of the blood in the vessels at atmospheric pressure, a posi- 404 THE RESPIRATION tive pressure, amounting in some experiments to + 50 atmospheres, was introduced but no effect could be noted on the circulating blood. By opening a tap in the chamber, decompression to zero pressure was quickly effected and, immediately, large bubbles were seen to develop in the blood, blocking the vessels and producing stasis. The bubbles were de- rived from the gas that had gone into solution under pressure. On re- applying the pressure the bubbles of gas again went into solution and the blood circulated normally. When the pressure was subsequently very gradually lowered to zero, the circulation went on undisturbed, and the frog was removed from the chamber in normal condition. The process involved in causing caisson disease is evidently the same as that which can be observed in a bottle of aerated water; if the cork in such a bottle is drawn, the dissolved gas escapes as bubbles and effer- vescence results ; if the bottle is recorked, the gas reenters solution and the fluid becomes quiet. If a pin hole is made in the cork, the gas will gradually escape and no effervescence will result. Confirmatory results have been secured by observations on mammals. The arterial blood pressure of rabbits was not found to become altered by exposure to compressed air, and various animals placed in a large, strong steel chamber at pressures far in excess of those to which man ever subjects himself did not show any symptoms like those of caisson sickness, unless the pressure was suddenly lowered.* Many times also, if symptoms had appeared they could be removed by again subjecting the animals to the compressed air. Investigations were also carried out to determine exactly how much gas the blood of an animal subjected to high pressures contains, and how long it takes to absorb the maximal amount of gas and to release it. It was found that the gases that increased in amount were nitrogen and oxygen, and that these become dissolved in the blood according to Dai- ton's law. The Prevention of the Symptoms The most important practical application of these observations con- cerns the length of time required for the saturation and desaturation to occur, for the results serve as a basis upon which the safe regulation of work in compressed air by man can be conducted. The most significant outcome of the above experiments from this standpoint is that it takes considerable time for the blood to absorb its full quota of gas at a given atmospheric pressure and to liberate it again when the animal is decom- pressed. The cause of delay is that the tissue fluids other than the blood take much longer than would be expected to reach equilibrium with the partial pressure of gas in the blood plasma. BREATHING IN COMPRESSED AND IN RAREFIED AIR 405 To understand why this delay should occur, let us suppose that the only gas concerned is nitrogen. As the pressure rises, the blood in the capillaries of the lungs must dissolve nitrogen in proportion to the pres- sure of this gas in the alveoli; the blood carries the dissolved gas to the tissues and these dissolve it until the pressure is again equalized between them and the blood. The blood, after giving up its excess of dissolved nitrogen, returns to the lungs and again becomes saturated and this goes on until blood and tissue have become saturated with gas at the external pressure. The tissues are two-thirds water and they contain (in man) from 15 to 20 per cent of fat. Fat, however, dissolves five times more nitrogen than water (Vernon) ; consequently, it takes longer for a given' volume of tissue than of blood to become saturated at a given pressure. The blood in man constitutes one-twentieth of the body weight; so that if the tissues were all liquid they would dissolve 20 times as much nitrogen as the blood. On account of the fat which they contain, however, the tissues take up more than this proportion — namely, in an average man about 35 times more than the blood. All the blood in the body takes about one minute to complete a round of the circulation, so that in this time, after being suddenly subjected to an increased pressure — assuming that the blood circulates equally throughout the body — the tissues will be one-thirty-fifth saturated; in the next minute another thirty-fifth of thirty-four thirty-fifths will be saturated, and so on. After five minutes the body will be about 22 per cent, and in 25 minutes about one-half, saturated; but it will take about two hours before saturation is complete. These calculations assume that the blood is evenly distributed through- out the body; but this is not the case, for its mass movement varies considerably in different parts, being much greater in the active muscles and in the glands than in passive structures, such as fat. These less vas- cular parts will therefore lag behind the others in taking up their full quota of gas, and therefore prolong the time necessary for complete saturation of the body as a whole. We see therefore that, after some time in compressed air, the blood and active tissues will be saturated and contain volumes of dissolved gas in proportion to their relative bulks ; the fat, although not saturated, will yet contain up to five times more gas than an equal volume of blood, and the passive tissues will be incompletely saturated. These considerations regarding the saturation of the different parts of the body apply also in its desaturation. Suppose, for example, that the external pressure is suddenly lowered: the blood, on leaving the lungs, will contain no excess of gas; when it reaches the tissues it will remove gas until the pressure is equalized, discharge this into the alveoli and return again for more. Other things being equal, it will take the 406 THE RESPIRATION same number of minutes to desaturate that it took to saturate, and the parts of the body that will lag behind the others, in being desaturated, are those with a sluggish circulation. When the mass movement of the blood is increased by muscular exer- cise, the rate of saturation and desaturation with nitrogen is increased in proportion. During active work the increase in movement of the blood may be four or five times over the normal, so that the tissues of the caisson worker become much more quickly desaturated during decom- pression than the above figures would lead one to expect. Application of Foregoing Laws in Practice With regard to the application of these principles in the decompression of caisson workers, it is impracticable to occupy as much time as it takes to saturate the body even at comparatively low pressures. If the great dangers attending work in compressed air are to be avoided, we must either insist on very gradual decompression or we must show how the dissolved gases may be got rid of by some modification in the decom- pression procedure. With this object in view, we must determine what difference of pressure may be allowed between the external air and the body without the formation of bubbles. Actual experience shows that there is no risk of bubble-formation, however quick the decompression, after exposure to + 15 pounds pressure ( i. e., 2 atmospheres absolute) . "Now, the volume of gas capable of being liberated on decompression to any given pressure is the same, if the relative diminution of pressure is the same" — (Haldane35). On reduction from 4 to 2 atmospheres, the same volume of gas will tend to be liberated as on reduction from 2 to 1 atmospheres — that is to say, no bubbles will form. The practical conclusion is "that the absolute air pressure can always be reduced to half the absolute pressure at which the tissues are saturated without risk." Thus, after saturation at 90 pounds absolute pressure (H- 5 atmos- pheres), a man can be immediately decompressed to 45 pounds (+ 2 atmospheres) in a few minutes without risk, but from this point on the decompression must be conducted slowly, so as to insure that the nitrogen pressure in the tissues is never more than twice the air pressure. The great advantage of this method is that it makes the greatest possible use of difference of pressure between tissues and blood in order to get rid of the gas that these contain. When the decompression from the start is gradual, the desaturation of the tissues will progressively lag behind that of the blood, and the tendency to the liberation of free gas will become greater. In such a case the decompression is far too slow at first and far too rapid later. BREATHING IN COMPRESSED AND IN RAREFIED AIR 407 Theoretically, therefore, tlie decompression should be rapid at first and very slow later. Before recommending the adoption of this principle of stage de- compression in caisson work, Haldane and his coworkers made numerous observations on the incidence of decompression symptoms in laboratory animals. They assert that the stage method is decidedly safer than the uniform method, the advantage being particularly after short exposures. On the other hand, Leonard Hill could make out no definite advantage for the stage method. The two methods have also been compared in actual caisson work at the Elbe Tunnel, where the pressure was + 2 atmospheres. Very little advantage could be demonstrated for the stage as compared with the uniform method at this comparatively low pressure. The general conclusion which we may draw is that the stage method should be employed, although it is not to be expected that it will absolutely insure absence of decompression symptoms. Of course the great advantage of the stage method is the saving of time, making it possible to persuade the workmen to adopt it. There are two other factors that are to be considered in hastening the desaturation of the tissues; these are muscular exercise, and the breath- ing of an indifferent gas. It is clear, from what has already been said, that the gas dissolved in the tissues will become removed in proportion to the mass movement of the blood, and it is probably true that muscular exercise, performed in the decompression chamber, is of as great importance in preventing the subsequent development of symptoms as a much prolonged decom- pression. In a man at rest, the circulation through the central nervous system and the viscera is constantly influenced by the pumping action of the respiratory movements, but in the capillaries of the muscles, joints, fat, etc., this influence is not felt and the blood flows more slowly. It is consequently in these parts that bubble formation is likely to oc- cur, especially some time after decompression. The bubbles cause the neuralgic pains — the ' ' bends ' ' and ' ' screws ' ' so. well known to caisson workers. These could no doubt be entirely prevented by muscular exercise and massage of the limbs during decompression. In illustration of these facts the following experiment by Greenwood may be cited: During decompression from + 75 pounds pressure in 95 minutes "Green- wood flexed and extended all the limb joints at frequent intervals, with the exception of the knees. Subsequently pain and stiffness were ex- perienced in the knees and nowhere else." In another experiment the knees also were flexed and no pain was felt. But even in the parts with active circulation, the gas in the tissues may lag considerably behind that in the blood, although the decompres- 408 THE RESPIRATION sion has been properly controlled. This has been shown by Leonard Hill in the case of the kidney. The "tissue" gas in this case can be taken as the gas dissolved in the urine, by analyzing which, therefore, at different stages of decompression, the excess of nitrogen over what it should be at the external pressure, can be ascertained. On decom- pression from + 30 pounds by two stages to zero, a considerable super- saturation was found to exist. The excess of nitrogen can, however, be cleared out of the kidneys rapidly and completely by breathing oxygen, which should therefore be administered during decompression in cases where great care has to be exercised (Leonard Hill). When symptoms do appear, they can, in most cases, be relieved by recompression, and all modern caisson works are provided with a special chamber for this purpose. We need scarcely say anything about this treatment here, as its value is so well known. Suffice it to say that, although it is most likely to afford relief when applied as soon as pos- sible after the appearance of the symptoms, yet it is often efficacious when applied several days after their onset. Quite apart from the dangers of decompression, it must of course be remembered that the working conditions in a caisson are somewhat dif- ferent from those at atmospheric pressure, as the air, owing to its com- pression, is warmer and is loaded to saturation point with moisture. This hot, wet air interferes with the heat-regulating mechanism of the body, making hard muscular work very uncomfortable because of the tendency of the body temperature to rise. The reaction of the body against this tendency to hyperthermia consists in dilatation of the su- perficial capillaries and increased heart action. When such working conditions are repeated day by day, the appetite is likely to fail, partly because of the tendency of the body to suppress the activity of the metabolic processes, so as to keep down heat produc- tion, and partly, no doubt, because the' digestive processes are working below par on account of there being less blood circulating through the visceral blood vessels, it having been sent to the surface of the body to be cooled off. The worker therefore tends to take less food, his metabo- lism becomes depressed, and his factors of safety against bacterial infections become lessened. The risk of the appearance of symptoms on decompression is also greater when the air in the caisson has been moist and hot, for the heart has been overworking to maintain the bloodflow in the dilated vessels; it gets fatigued and is consequently unable to maintain, during decom- pression, a rate of bloodflow that is adequate for carrying the gas- saturated blood to the lungs, where the excess of gas becomes dissi- pated. BREATHING IN COMPRESSED AND IN RAREFIED AIR 409 The criterion of proper working conditions in the caisson is there- fore the wet-bulb temperature. This should stand below 75° F. To maintain this condition it is necessary to ventilate the caisson, pref- erably with air that has been cooled by cold-water radiators; in any case, the ventilation should be adequate to keep down the wet-bulb temperature. The increased expense of ventilation with cooled air would soon be balanced by the greater working efficiency of the men. Constant circulation of the air in the caissons by means of fans assists also in improving the conditions, for it helps to increase dissipation of heat from the body. CHAPTER XLVII THE CIRCULATORY AND RESPIRATORY CHANGES ACCOM- PANYING MUSCULAR EXERCISE* During activity the muscles require many times more blood than dur- ing rest. When the activity is widespread the greater blood supply is provided by increased heart action accompanied by dilatation of the muscular arterioles and constriction of those of the splanchnic area, so that the entire available blood supply of the body is made to circulate more rapidly. When, on the other hand, the activity is confined to a limited group of muscles, the increased blood supply is mainly provided by a local dilatation of the blood vessels of the active muscles accom- panied by a reciprocal constriction of those of inactive parts. Under these conditions there may therefore be no quickening of the bloodflow as a whole. In order that this accurate adjustment of blood supply to tissue demands may be promptly and adequately brought about, all available types of coordinating mechanism are called into play; that is to say, mechanical, nervous and hormone factors cooperate to an extent which is dependent upon the type of work being performed. Besides the changes in pulse rate and blood pressure which are evi- dently designed to supply more blood to the acting muscles, changes dependent upon a secondary effect of the muscular movements have also to be considered. Although the various factors work together and are more or less interdependent, the final effect can be understood only after we have studied the relative influence of each separately. The Mechanical Factor. — It is particularly with regard to this factor that the circulatory changes may be an unavoidable consequence of, rather than a useful adjustment to, the muscular effort. The effects vary with the type of exercise performed. In repeatedly lifting and lowering dumbbells from the floor to above the head, the contracting muscles of the back and extremities and of the abdomen compress the veins and cause the blood to flow more rapidly into the hear,t, so that the arterial pressure suddenly rises. So long as this compression exists, the veins remain relatively empty and the arteries overfilled, but whenever it ceases and the muscles relax, the veins fill up again and the arterial pres- *This chapter is placed here rather than following circulation because of the interdependence of the circulatory and respiratory adjustments. 410 CHANGES ACCOMPANYING MUSCULAR EXERCISE 411 sure markedly falls, until the extra space in the veins has been occupied by blood. It is for this reason that the arterial blood pressure is always found to be little, if any, above normal when taken within a few seconds after such exercise. It subsequently rises because the other factors responsible for the increased pressure (quick heart and arteriole constric- tion) are still in operation at the time the veins again become filled with blood. The purely mechanical influence outlasts the exercise for a com- paratively short time, whereas the nervous and hormone influences con- tinue acting. This interpretation is supported by the observation that the fall of blood pressure is greater when the subject is left standing after a given amount of dumbbell exercise than when he is allowed to sit with his elbows resting on his knees. In the standing position the pres- sure on the abdominal veins is less and the hydrostatic effect of gravity causes more blood to collect in the large veins (Cotton, Rapport and Lewis36). Being purely mechanical in its causation, the preliminary fall following dumbbell exercise can always be demonstrated if the observa- tions are made at close enough intervals of time. The mechanical response of the circulation to exercise acts therefore through the rate of filling of the right heart with blood, and if this organ is in a healthy condition, it will respond to the greater inflow by correspond- ingly increased discharge. Like every other physiologic mechanism, the heart works with a large factor of safety — a reserve power — and it is the rate of venous filling that determines how much of this reserve must be called upon to maintain the circulation. In isolated heart-lung prep- arations Starling and his coworkers have very clearly demonstrated the close dependence of cardiac output upon rate of venous filling and the enormous range through which the systolic discharge can be made to vary by altering this factor. As explained elsewhere, when the reserve *power of the heart is lessened, the rise in blood pressure following exer- cise is longer in attaining its maximum, which is set at a higher level and persists for a longer time. Observation of the extent of these changes furnishes a most useful functional test of cardiac efficiency. Other mechanical factors that augment the cardiac output depend on the increased respiratory movements. During each respiration the in- crease in capacity in the thorax causes both an opening up of the thin- walled veins, so that blood is aspirated towards them from the extra- thoracic venous system, and a dilatation of the blood vessels of the lungs, so that the blood finds its way from right to left heart more readily. Although this dilatation will at first tend to cause more blood to collect in the intrathoracic vessels and less to be pumped out of them, the expira- tory act when it supervenes will, by compressing the veins, cause the extra blood to be expelled into the left ventricle and thence into the 412 THE RESPIRATION arteries. It is obvious that increased depth and frequency of the respira- tory movements will accelerate the bloodflow and tend to raise the arte- rial blood pressure. The above factors will come into play during most kinds of muscular exercise such as walking, running, or swinging dumbbells, etc. There are certain types of muscular effort, however, in which the mechanical factors produce decidedly disturbing effects on the circulation. During a sustained effort as, for example, in pulling against a resistance or in attempting to lift a heavy load, the respirations are suspended, often after a deep inspiration, and the contracted abdominal muscles press the dia- phragm up into the thoracic cavity. After a preliminary squeezing out of blood first of all from the veins of the abdomen into the thorax and then from those of the latter into the systemic arteries, with a consequent rise in arterial pressure, there comes to be a damming back of blood into the peripheral veins, causing them to swell and, if continued, marked cyanosis may develop. When such efforts are maintained for long, the arterial pressure begins to fall, and this fall is very pronounced indeed at the end of the effort, because, the compression being removed from the abdominal and thoracic veins, these open up and form a large unfilled blood reservoir. A similar mechanism comes into play during expulsive acts such as defecation, parturition, etc. In these the glottis is closed, usually after a preliminary inspiration, and a powerful expiratory movement is per- formed, with the consequence that the intrathoracic and intraabdominal pressures rise considerably, greatly augmenting the systolic discharge and causing the blood pressure to rise. Because of the obstruction to the bloodflow in the large veins of the abdomen and thorax, however, the later effect of the effort is to diminish the systolic discharge, but the fall in blood pressure which this would be expected to occasion is masked,' The pressure remains high because other factors increasing the peripheral resistance come into play. The fall in blood pressure following these acts may be very marked indeed. Similar mechanical effects are produced in the acts of coughing, sneezing, etc. The capacity of the veins varies considerably with the position of the body, and it is in order that we may cause alterations in this capacity and therefore encourage a more rapid bloodflow that we stretch the body after sitting for some time in a cramped position. The Nervous Factor. — The vagus, vasoconstrictor and respiratory cen- ters are all excited during muscular effort. In the earlier stages the excitation depends entirely on nervous impulses transmitted to the cen- ters, but later it depends on changes in the composition and temperature of the blood flowing through them — the hormone factor. The initial CHANGES ACCOMPANYING MUSCULAR EXERCISE 413 stimulation of the centers must be due to cerebral impulses independ- ently transmitted to the above centers, since the quickening of the pulse and respirations may be observed to begin before the actual muscular contractions. The Hormone Factor. — We have to consider first the nature of the hormone, and secondly the mode of its action. The Nature of the Hormones. — The most important hormone is car- bonic acid, but when the exercise is strenuous and continued, or from the very start is of such a nature that it uses up oxygen more quickly than the blood can supply it to the muscles, lactic acid also appears. Evidence for these statements can readily be supplied in man by analy- sis of the expired air (for carbon dioxide) and of the urine (for lactic acid) before and during muscular work. The real hormone in both cases is believed to be an increase in the H-ion concentration of the blood. There is, however, no direct proof of this assertion — that is to say, no one has actually shown that a measurable change in the H-ion concentra- tion of the arterial blood (for of course a change in the venous blood would be of no significance) does occur before the changes believed to be dependent upon acid production make their appearance. The well- known buffer action of the blood (that is, its ability to take up con- siderable quantities of acid or of alkali before any perceptible change occurs in H-ion concentration) furnishes another reason why doubt must be cast upon the H-ion hypothesis. The most delicate means for demonstrating a change in H-ion concentration of the blood consists in finding the dissociation constant for hemoglobin and the results have shown that acidosis develops during exereise at least at high altitudes (Barcroft1). So far as we are aware, however, it has not been possible by direct measurement (page 29) to detect a rise in H-ion concentration. Of course it may well be that the sensitiveness of the various nerve centers and other structures towards the H-ion concentration is very much greater than our most refined and sensitive laboratory methods can reveal. Such is at least commonly believed to be the case for the respiratory center (see page 351), and it may also be so for those of vascular tone and cardiac action. It is nevertheless possible that an increase in the free carbonic acid itself — the carbonate anion (-HC03), in other words — is the effective hormone. In the first stages of muscular work, this increase would be due to greater production of C02, .whereas later, especially when the work is strenuous, lactic acid would decom- pose the NaHC03 of the blood, liberating -HC03, which would become added to that still being produced by the active muscles, and as the NaHC03 (buffer substance) became gradually used up, would cause a relatively greater and greater proportion of -HC03 to exist in a free . 414 THE RESPIRATION state. That the C02 tension of the alveolar air should be found to be lowered by prolonged muscular exercise in no way detracts from this explanation, for it is dependent upon the greatly increased rate of movement of air into and out of the alveoli (see also page 366). One serious difficulty in accepting the HC03 ion as the exciting hor- mone of the nerve centers during muscular exercise depends on the ob- servation that the alveolar C02 after some time is lower than normal. If we accept Haldane's teaching that there is accurate correspondence between the tensions of C02 in arterial blood and alveolar air not only during rest but also during muscular activity, then obviously we must discard the HC03 hypothesis. Leonard Hill and Flack,37 however, have shown quite clearly both in experimental animals and in man that equi- librium between the blood and alveolar tensions of C02 may fail to occur. "When blood with excess of C02 is injected into the jugular vein of dogs, the respiratory center is stimulated, as shown by the increased breathing, which indicates that the C02-rich blood must have passed through the lungs without the excess of C02 being removed from it. Hill believes that the diffusion of C02 out of the blood into the alveolar air may be depressed in muscular exercise, and that this rather than the appearance of lactic acid in the blood is responsible for the low C02 ten- sions usually found present (see page 369). He points out in support of this view that a person after exercise can hold his breath for a much shorter time than is usual, and the C02 meanwhile mounts in the alveolar air very rapidly. The only way by which progress may be made in a problem like that under discussion is, however, to adopt some hypothesis and then to gather evidence for or against it. At the present stage of our knowl- edge, the hypothesis usually adopted is that a slight change in H-ion concentration of the blood is the effectual hormone. It is an hypothe- sis which is supported by the parallelism between the effects observed during muscular exercise and those produced by experimental increase in H-ion concentration. The Effects of the Hormone. — 'These may be classified as follows: (1) strictly local effects on the muscles themselves; (2) effects on the heart; and (3) effects on the nerve centers. The local production of acids in the muscles will cause dilatation of the arterioles, for it has been shown by various observers that acids cause relaxation of vascular muscle. Even the capillaries themselves are said to be dilated by carbonic acid (Severini). The effects produced on the heapt by changes in H-ion con- centration of the blood have been particularly studied by Starling and Patterson,38 who, working on isolated heart-lung preparations, have shown that the heart relaxes more and more and discharges less blood CHANGES ACCOMPANYING MUSCULAR EXERCISE 415 » as the H-ion concentration of the perfusion fluid is increased by adding C02 to the air ventilating the lungs. The influence of changes in H-ion concentration of the blood on the vagus and vasomotor centers is usually believed to be stimulatory. There is no doubt that an increase in CH stimulates the vasoconstrictor centers, not only of the medulla, but also, although much more feebly, of the spinal cord. But it is a question whether any part of the rise in systolic pressure during muscular exercise can be attributed to this cause, for the enormously increased bloodflow which is known to occur makes it problematical whether any vasoconstriction really occurs. If it does so, it must be confined to the splanchnic area, where it would have the effect of bringing about a redistribution of the total available blood by expressing it from the viscera and sending it to the active muscles. The effect of increased H-ion concentration on the vagus center must, be insignificant. It is commonly believed that it would cause not what is actually observed, a quickening, but rather a slowing of the heart rate. But even this is doubtful. The slowing of the heart that is observed in asphyxia, for example, is in part at least due to the increased intra- . cranial pressure, for when the carotid artery is connected with a mer- cury valve so- that the blood escapes as the pressure rises above the normal level, no slowing of the heart is said to occur in asphyxia. As Leonard Hill and Flack37 have shown, however, a part of the slowing is due to the direct effect of C02. If increase in the H-ion concentration does affect the heart during muscular exercise, it must act by inhibiting the vagus tone, which is opposite to the action which it is usually be- lieved to have. The activity of .the respiratory center is of course ex- cited" by increase in H-ion concentration, and this, as we have seen, will cause important changes in the circulation because of the mechanical effects which follow. Along with hormones we must consider the effect of change in the temperature of the Hood. That this rises during muscular exercise is well known, but that it should be responsible for many of the cardio- vascular adjustments that occur is quite commonly overlooked. It is, for example, very likely that rise in blood temperature is responsible for the acceleration of the heart that occurs during exercise when both vagi have been severed, and it no doubt is responsible for a part ' at least of the vasodilatation and respiratory acceleration. Finally, it is interesting to speculate as to the nature of the changes that occur when the " second wind" is acquired during strenuous mus- cular exercise. In running, for example, considerably more distress is experienced a short time after the start than some time later. Three 416 THE RESPIRATION very definite changes occur at the time the relief is experienced — namely, a slowing and steadying of the previously much quickened and irregu- lar pulse, sweating, and a marked fall in the respiratory quotient. The last mentioned change possibly gives a clue to the cause of the others. In the early stages R. Q. is raised, which indicates that relatively more C02 is being expelled from the blood into the alveolar air than oxygen is being absorbed, perhaps because of inadequate movement of blood through the lungs. At the time of the adjustment it is possible that a pronounced vasodilatation occurs in the muscles and coronary arteries. The former change by lowering the arterial blood pressure will relieve the pumping action of the heart, and the latter will improve its power of contraction by supplying it with more oxygen. % RESPIRATION REFERENCES (Monographs) Barcroft, J.: The Respiratory Function of the Blood, University Press, Cambridge, 1914. Borrutau, H.: Nagel's Handbueh der Physiologic, 1905, i, 29. Douglas, C. G.: Die Regulation der Atmung beim Menschen, Ergebnisse der Physiol- ogie, 1914, p. 338. Hill, Leonard: Caisson Sickness, International Medical Monographs, E. Arnold, London, 1912. Keith, Arthur: The Mechanism of Respiration in Man, Further Advances in Physi- ology, E. Arnold, London, 1909. Schenck, F.: Innervation der Atmung, Ergebnisse der Physiologic, 1908, p. 65. (Original Articles) iKeith, Arthur: Cf. Further Advances. sHoover, C. F.: Arch. Int. Med., 1913, xii, 214; ibid., 1917, xx, 701. 3Lee. F. S., Guenther, A. E., and Meleney, H. F.: Am. Jour. Physiol., 1916, xl, 446. *Meltzer, S. J.: Jour. Physiol., 1892, xiii, 218. sHaldane, J. S., and Priestley, J. G.: Jour. Physiol., 1905, xxxii, 225. Haldane and Douglas: Ibid., 1913, xlv, 235. 6Henderson, Y., Chillingworth and Whitney: Am. Jour. Physiol., 1915, xxxviii, 1. Henderson and Morriss: Jour. Biol. Chem., 1917, xxx, 217. 7 Krogh, A., and Lindhard: Jour. Physiol., 1913, xlvii, 30; ibid., 1917, li, 59. sPearce, R. G.: Am. Jour. Physiol., 1917, xliii, 73; ibid., 1917, xliv, 369. eSiebeck, R.: Skand. Arch. f. Physiol., 1911, xxv, 87; Carter, E. P.: Jour. Exper. Med., 1914, xx, 21. loPeabody, F. W., and Wentworth, J. A.: Arch. Int. Med., 1917, xx, 443. "Lewis, T.: Jour. Physiol., 1908, xxxiv, 213, 233. isPorter, W. T.: Jour. Physiol., 1895, xvii, 455. ^Christiansen and Haldane, J.: Jour. Physiol., 1914, xlviii, 272. "Boothby, W. M., and Berry, F. B.: Am. Jour. Physio!., 1915, xxxvii, 433; also Boothby, W. M., and Shamoff, V. K: Ibid., p. 418. isAlcock, N. H., and Seemann, J. : Jour. Physiol., 1905, xxxii, 30. leScott, F. H.: Jour. Physiol., 1908, xxxvii, 301. "Stewart, G. N., and Pike, F. H.: Jour. Physiol., 1907, xx, 61. i7aCoombs, H. C., and Pike, F. H. : Proc. Soc. Exper. Biol. Med., 1918, xv, 55. isKrogh, A.: Skand. Arch. f. Physiol., 1910, xxiii, 248; and A. Krogh with Marie Krogh, ibid., 179. CHANGES ACCOMPANYING MUSCULAR EXERCISE 417 isHaldane, J. S., and Priestley, J. G.: Jour. Physiol., 1905, xxxii, 225. zoScott, E. W.: Am. Jour. Physiol., 1917, xliv, 196. 2iNewburg, Means, and Porter, W. T. : Jour. Exper. Med., 1916, xxiv, 583. 22Hasselbalch, K. A., and Lundsgaard, Chr.: Biochem. Ztschr., 1912, xxxviii, 77, and Skand. Arch. f. Physiol., 1912, xxvii, 13. 23Hooker, D. E., Wilson, D. W., and Connett, H.: Am. Jour. Physiol., 1917, xliii, 357. 24Campbell, J. M. H., Douglas, C. G., and Hobson, F. G.: Jour. Physiol., 1914, xlviii, 303. 25Lindhard, J.: Jour. Physiol., 1911, xxxviii, 337; Haldane, J. S., and Douglas, C. G.: Ibid., 1913, xlvi. 26Douglas, C. G. : Art, Ergebnisse der Physiologic, see Monographs. 2?Barcroft, J. : see Respiratory Function of Blood. 28Milroy, T. H.: Quart. Jour. Physiol., 1913, vi, 373. 29Fletcher, W. M., and Hopkins, F. G.: Jour. Physiol., 1907, xxxv, 247; also Fletcher, W. M.: Jour. Physiol., 1913, xlvii, 361. soEyffel, J. H.: Proc. Physiol. Soc. in Jour. Physiol., 1909, xxxix, 29. siPembrey, M. S., and Allen, E. W.: Jour. Physiol., 1909, xxxii, 18. 32Buckmaster, G. A.: Jour. Physiol., 1917, li, 105. 33Douglas, C. G., Haldane, J. S., Henderson, Y., and Schneider, E. C. : Phil. Trans. Eoy. Soc., 1913, 203, B, 185. - 3*Hill, Leonard, Macleod, J. J. E.: Jour. Physiol., 1903, xxix, 507; Hill, Leonard, Greenwood, M., Flack, M., etc.: see Hill's Caisson Sickness. ssHaldane, J. S. : Deep Water Diving, Committee of the Admiralty (British), see Hill's Caisson Sickness. seCotton, T. F., Eapport, and Lewis, T.: Heart, 1918. 3'Hill, Leonard, and Macleod, J. J. E. : Jour. Physiol., 1908, xxxvii, 77. 3spatterson, S. W., Piper, H., and Starling, E. H.: Jour. Physiol., 1914, xlviii, 465. PART V DIGESTION CHAPTEK XLVIII GENERAL PHYSIOLOGY OF THE DIGESTIVE GLANDS The function of digestion is to bring the food into such a condition that it can be absorbed through the intestinal epithelium into the blood and lymph. Carbohydrates are broken down as far as monosaccharides ; neutral fats are split into fatty acids and glycerine; and proteins are broken down into the amino acids. The agencies which effect these decompositions are the digestive enzymes, or ferments, contained in the various digestive fluids or juices. The digestive juices are produced by glands, which are most numerous in the upper levels of the gastro- intestinal tract, the lower levels having as their main function that of absorption of the digested products. In order that the masses of food may be kept in a state of proper consistency, and that they may move readily along the digestive canal, numerous mucous glands are also scattered along the whole extent of the canal. Some of the digestive glands, such as the main salivary glands, the pancreas, and the liver, discharge their secretions into the digestive canal by special ducts, whereas others, such as the isolated salivary gland follicles in the mouth, the gastric glands and the crypts of Lieberkiihn in the intestine, do not have an anatomically distinct duct, but discharge their secretions directly into the digestive tube. It will be convenient to consider, first of all, certain properties that are common to the digestive glands, and then, the conditions under which each gland functionates during digestion. MICROSCOPIC CHANGES DURING ACTIVITY Structurally the active part of the glands, represented by the acinus or tubule, is composed of a basement membrane lined internally with the secreting epithelium. Outside the basal membrane are the lymph spaces and blood capillaries. After the gland has been at rest, the cells become 418 PHYSIOLOGY OF THE DIGESTIVE GLANDS 419 filled with granules or small globules, which are often so numerous as almost entirely to obliterate the nucleus. When the gland becomes active, on the other hand, the granules or globules leave the cells, except for a few which remain toward the lumen border. (Figs. 143 and 144.) A. Fig. 143.— Cells of parotid gland showing zymogen granules: A, after prolonged rest; B, after a moderate secretion; C, after prolonged secretion. (From Langley.) These observations indicate that the granular or globular material must represent part at least of the secretion of the glands. Sometimes, even before they are extruded, the granules become changed into some differ- ent material, as is indicated by the fact that they stain differently from D. Fig. 144. — Parotid gland of rabbit in varying states of activity examined in fresh state. The upper left-hand acini are resting. The upper right-hand acini are from a gland stimulated to activity by injecting pilocarpine, and the two lower acini from one after stimulation of its sym- pathetic nerve. (After Langley.) those of the resting gland. It must not be thought, however, that an extrusion of granules necessarily accompanies secretory activity, for under certain conditions a copious secretion of water and inorganic salts, as well as a certain amount of organic material, may be produced with- 420 DIGESTION out any change in the arrangement of the granules. In such cases it has been observed, as in the pancreas, that fine channels develop in the protoplasm of the cell (see page 429). From this histologic evidence it would appear that the gland cell during rest is endowed with the property of building up out of the pro- toplasm, as granules or globules, the material which is to serve as one of the main organic constituents of the secretion. It is commonly believed that this is the precursor of the active ferment of the secretion ; hence its name, zymogen. It has been shown that the process of separation of the zymogen granules starts around the nucleus with the production of a basophile substance, which in hardened specimens sometimes takes the form of filaments. From this basophilic ergastoplasm, as it is called, the granules are gradually formed, and then for some time continue to undergo slight further changes, as is evidenced by the fact that the staining reaction of those near the base of the cells differs from that of those at the free margin. When the gland cell is excited to secrete, the granules before being extruded, as noted above, often undergo a definite change, becoming swollen and more globular in shape. MECHANISM OF SECRETION These histologic studies merely tell us that active changes, associated with the production and liberation of certain of the constituents of its secretion, are occurring in the gland cell, but they throw no light on the mechanism whereby the gland cells secrete water and inorganic salts. This may be dependent, to a certain extent at least, on differences in osmotic pressure (see page 11). A possible explanation of the flow of water is as follows: If a watery solution of some osmotically active sub- stance is put in a tube, which is closed at one end" by a membrane impermeable to this substance and at the other by one permeable to it, and the tube immersed in water, a continuous current will be found to issue from the permeable end so long as there remains any osmotically active substance in the tube. If we assume, then, that the membranes at the two ends of the secreting cell are of such a nature that the one next the basement membrane is impermeable to some osmotically active substance manufactured by the cell, and the other toward the lumen is permeable, it will be clear that, so long as this substance exists in the cell, it will attract water from the blood, and the water together with the osmotically active substance will be discharged into the lumen. It is possible that when anything excites the cell to secretory activity, such as a nerve impulse or' hormone, it does so by causing a change in PHYSIOLOGY OF THE DIGESTIVE GLANDS 421 the permeability of the lumen border of the cell. This change in permea- bility may be dependent upon alterations in surface tension brought about by the migration of electrolytes to the border. That such a migra- tion of electrolytes does actually occur has been demonstrated by A. B. Macallum8 who developed a microchemical test for potassium, and was able to show that this electrolyte accumulates at the lumen border of the cell during secretory activity. Potassium may be taken as a prototype of electrolytes in general. Support is given to this view by the fact that potassium always accumulates at the border of the cell through which the secretion takes place. In the epithelium of the small intes- tine, for instance, where the current goes in the opposite direction to that in gland cells, the accumulation of potassium occurs at the portion of the cell next the basement membrane. Another possibility is that, when the gland becomes more active, the molecules present in the cell become broken down into smaller molecules and so raise the osmotic pressure of the cell content, with the result that water is attracted from the blood. When the gland is excited so that the zymogen granules, as well as water and salts, are secreted, the primary change appears to involve the granules only. Those near the lumen swell up by absorbing water, and become converted into spheres in which salts are dissolved in smaller proportions than exist in the lymph bathing the cells. These swollen structures are then ruptured at the periphery of the cell and discharged into the lumen. This discharge of a fluid containing fewer saline con- stituents than the cell or surrounding blood plasma brings about in- creased concentration in the remaining parts of the cell, a process which possibly is assisted by a breaking up of molecules in the protoplasm itself, and which causes an increase in osmotic pressure with a consequent flow of water from the lymph to the cells and therefore from the blood to the lymph. OTHER CHANGES DURING ACTIVITY Whatever may be the nature of the physiological changes that are responsible for the secretory activity of the cell, the fact stands out prominently that a considerable expenditure of energy is entailed. This is indicated by the fact that considerably larger quantities of oxygen are taken up by the gland when it is in an active state than when at rest. Thus, the oxygen consumption of the resting submaxillary gland of the cat may be increased five times during active secretion. On account of this increased oxygen consumption it is not surprising that it should be found that the secretory activity of the cell is greatly im- paired by a deficiency in oxygen. 422 DIGESTION These active processes occurring in the gland when it is excited to secrete are associated with changes in electric reaction and in the volume of the gland. The electric changes have been most extensively studied in connection with the salivary gland. Cannon and Cattel,6 by connecting a galvanometer with nonpolarizable electrodes, one placed on the gland and the other on neighboring connective tissue, were able to show that with each period of active secretion a current of action was set up. This was first discovered by Rose Bradford and Bayliss, and has been carefully studied by Gesell.Ga That the electric current is definitely associated with the secretion of saliva and is not caused by the vascular changes which usually accompany this act was shown by its occurrence when the blood supply was shut off from the gland, and by its absence when there was no secretion even though the vascular changes were brought about; neither is the electric change due to the movement of fluid along the duct, as evidenced by its persistence after ligation of the duct. With regard to change in volume, it might be expected, on account of the greater vascularity of the gland accompanying activity, that this would increase. On the contrary, however, it has been shown to de- crease, because of the large quantity of fluid secreted from the gland cells. The action of two drugs on the gland cells is of considerable physio- logic importance: that of atropine, which paralyzes the secretion, and that of pilocarpine, which stimulates it. We shall see later how this information may be used in working out the exact mechanism of the different glands. Important observations concerning the relationship of glandular activ- ity to the blood supply have been made by experiments in which glands were artificially perfused outside the body. When the submaxillary gland of the dog is perfused with oxygenated Ringer's solution, stimula- tion of its nerve supply does not produce the usual secretion, but if the Ringer's solution is mixed with blood plasma, the nerve stimulation has its usual effect for a short time. Although no secretion occurs when oxygenated Ringer's solution is perfused alone, the usual vascular changes still occur in the gland. The results seem to indicate that the presence of some constituent of the blood plasma is essential for the change in the permeability of the cell wall necessary for the process of secretion. Similar results have been obtained during artificial perfusion of the pancreas when secretin was used as the stimulus. CONTROL OF GLANDULAR ACTIVITY Having outlined the general nature of the changes occurring in gland cells during their activity, we may now proceed to study the nature of Center for for Facial nerveJNML Cerebellui (N. Medulla oblo'ngataj Parotid gland- Cord Thoracic nerves it fympani nerve wall superficial ^^ petrosal nerve'^Parotid duct Inf. max. div. N.V^ (Sten son's) •rS>-— -* , ^^, /Submaxillary I U^^^ff^PJl5 / duct(Wharron's) 'ublingual ducf (Bartholin's) Lingual nerve Chordo-linguaJ triangle nerve /,)Chorda tympani branches. /Superior cervical gang. Electrodes fiffeetf . (Small amount of thkk saliva vaso-constnction ) , Vaso constrictor fibers sympathetic secretory fibers Outgoing sympathetic rami communicantes Electrodes ( Large amount of thin 'Siibmaxillary \ vase-dilatation gland \Sublingual gland Post- qanglionic fibers are dotted thus — - Fig. 145. — Diagrammatic representation of the innervation of the salivary glands in the dog. (From Jackson.) PHYSIOLOGY OP THE DIGESTIVE GLANDS 423 the process by wh'ich this glandular activity is controlled. Two mechan- isms of control are known: (1) by the nervous system, and (2) by means of hormones. Nervous Control. — Control through the nervous system is most marked — indeed it may be the only means of control — in glands which have to produce their secretion promptly, whereas hormone control pre- dominates in those in which prompt changes in secretory activity are not required. Thus, nervous control alone is present in the salivary glands, whereas hormone control is predominant in the pancreas, intestinal glands and liver. The gastric glands are partly under nervous control, and partly under hormone control. It should be pointed out here that the glands of the body other than the digestive glands are also subject to nervous or hormone control according to the promptness with which they are required to secrete. The lachrymal and sweat glands, and the venom glands of reptiles, for example, are practically entirely under nervous control, whereas most of the ductless glands, with the exception of the adrenals, are mainly under the influence of hormones. The exact nature of the nervous control of glandular function has, therefore, been most extensively studied in the salivary glands, and that of the hormonic in the pancreas. With regard to the salivary glands, the following points are of importance: Their nerve supply comes from two sources: the cerebral autonomic, and the sympathetic autonomic (see page 877). These two nerve supplies have usually an opposite influ- ence on the secretory activity of the glands, and very frequently also on the vascular changes that accompany secretory activity. On account of its ready accessibility, the submaxillary gland in the dog and cat has been most thoroughly investigated. The cerebral auto- nomic nerve in this case is represented by the chorda tympani, and the sympathetic autonomic by postganglionic fibers that run from the superior cervical ganglion to the gland along its blood vessels (Fig. 145). After tying a cannula into the duct of the gland, it will be found in the dog that stimulation of the chorda tympani produces an immediate and abundant secretion of thin watery saliva accompanied by a marked dilatation of the blood vessels of the gland. That this secretion is not dependent on the vasodilatation is easily shown by repeating the experiment after administering a sufficient dose of atropine to paralyze the secreting cells. Stimulation of the nerve then produces a vasodilatation but no secretion. " The same conclusion is arrived at by an experiment of an entirely different nature ; namely, by observing the pressure produced in the duct when the chorda tympani is stimulated. This pressure rises considerably above that in the arteries, so that no such physical process as mere diffusion can be held accountable 424 DIGESTION for the secretion, and therefore vasodilatation alone can not be respon- sible for it. If the sympathetic nerve supply is stimulated, a very scanty, thick secretion takes place accompanied by vasoconstriction. Repetition of these experiments in the cat yields different results, particularly with regard to the influence of the sympathetic, a copious secretion being produced by such stimulation. The histologic changes produced in the gland cells are marked after sympathetic stimulation, but very slight, if present at all, after chorda stimulation. The outstanding conclusion which may be drawn from these results is that two kinds of secretory activity are mediated through the nerves; one causing a thin watery secretion, containing only a small percentage of organic matter, and the other, a thick viscid secretion with a large amount of organic material. To explain these differences the hypothe- sis has been advanced that! there are really two kinds of secretory fibers, called secretory and trophic, the former having to do with the secretion of water and inorganic salts, and the latter with the secretion of organic matter ; i. e., with the extrusion of the zymogen granules. Certain authors (Langley) believe that such an hypothesis is unneces- sary, and that the different results are dependent upon the concomitant changes in the blood supply produced by stimulating one or other nerve. That there are really different kinds of true secretory fibers is, however, evident from the following experiment. If the duct of the gland is made to open through a fistula in the cheek, secretion of saliva through the fistula can be induced by placing various substances in the mouth, such as meat powder or hydrochloric! acid. If the experiment is performed in such a way that the bloodflow through the gland can be observed, it will be found that the saliva produced by the stimulation with the meat powder contains a very much higher percentage of organic material than does that produced when hydrochloric acid is the stimulant, whereas the vascular changes in the gland and the inorganic constituents of the saliva are the same in both cases. Since stimulation of the chorda tympani causes the secretion of a watery saliva, while that caused by stimulation of the sympathetic is thick, it might be thought that the secretory fibers were contained in the former and the trophic fibers in the latter nerve ; that this is not the case can be shown by a repetition of the above experiment in animals from which the superior cervical ganglion has been removed. The same results are obtained, indicating that the chorda tympani con- tains both secretory and trophic fibers. CHAPTER XLlX PHYSIOLOGY OF THE DIGESTIVE GLANDS (Cont'd) THE HORMONE CONTROL This is exhibited best in the case of the pancreas. The crucial experi- ment demonstrating that this gland is not primarily dependent upon nervous impulses for the control of its activity was performed by Bay- liss and Starling.2 Starting with the well-known fact that the application of weak acid to the duodenal mucous membrane excites secretion of pan- creatic juice, these workers carefully severed all the nerve connections of a portion of the duodenum, and found on again applying acid to the mucous membrane that the secretion persisted. To explain this result they postu- lated that the acid must cause some substance to be liberated into the blood stream, which carries it to the pancreas, the cells of which it then excites to activity. To test this hypothesis they scraped off the mucous membrane of the duodenum and ground it in a mortar with weak hydro- chloric acid (0.6 per cent), and, after boiling the solution so as to remove the protein and nearly neutralizing it, they obtained a fluid which, when injected intravenously, immediately caused a copious secretion of pan- creatic juice. Accompanying the secretion, however, a marked fall in arterial blood pressure was observed, making it possible that the secretion might have been due to a vasodilatation occurring in the pancreatic blood vessels. To eliminate this possibility they prepared an extract that was free of the depressor substances by extracting intestinal epithelium without any of the submucous tissue. The resulting extract had merely the secretory effect and produced no fall in blood pressure. This secretagoguary substance they named secretin. Further evidence that the action of secretin is independent of the depressor substances has been obtained by taking advantage of the fact that the depressor substance is more soluble in alcohol than the secretin. If an acid decoction of duodenal mucous membrane is poured into abso- lute alcohol, a precipitate is formed. If this precipitate is redissolved in water and reprecipitated several times by absolute alcohol, then after drying a white powder is obtained, which is easily soluble in water. The resulting solution injected intravenously has a powerful secretory action, but produces no effect on blood pressure. The concentrated alcoholic 425 426 DIGESTION liquor, on the other hand, when similarly injected produces a marked fall in blood pressure. It is believed that this effect is due to the action of /3-imidazolylethylamine. A very strong preparation of secretin can also be prepared by the method of Dale and Laidlaw,7 which depends on pre- cipitation by mercuric chloride. Secretin does not exist preformed in the epithelial cells, as is shown by the fact that an extract, made with neutral saline solution, does not as a rule, have any secretory action when injected intravenously. Sometimes a slight secretion may be produced, but this is probably to be explained by the fact that some secretin remains behind in the cells as a result of a preceding phase of activity. If, on the other hand, the above neutral or slightly alkaline opalescent solution of the mucous membrane is boiled with acid, secretin may become developed in it. The interpretation put upon these results is that a substance, called prosecretin, exists in the epithelial cells, and that this becomes converted into secretin by the action of acid on the cells. The secretin thus produced is then taken up by the blood, none of it passing into the intestinal canal, because the free borders of the cells are impervious to secretin. That this is actually the case has been shown by finding that the introduction of neutralized secretin solu- tion into the duodenum, or other parts of the small intestine, does not cause a secretion of pancreatic juice. We know practically nothing concerning the chemical nature of secretin. Being soluble in about 90 per cent alcohol and in fairly weak acids, it can not belong to any of the better known groups of proteins. As it is readily diffusible through parchment membrane, it can not be of very complex structure, and as it withstands heat, it can not be an enzyme. It rapidly deteriorates in strength in the presence of alkalies. Any acid when applied to the mucous membrane is capable of producing secretin, and so are certain other substances, such as mustard oil. Watery solutions of saccharose or urea, when rubbed up with the duodenal mucosa in a mortar, produce secretin solutions of varying activity, but they do not in the living animal excite pancreatic secretion when applied to the duodenum. Secretin is very susceptible to destruction by such digestive enzymes as those present in the pancreatic, gastric, and intestinal juices. That secretin is present in the blood when acid is in contact with the duodenal mucosa has been shown by the fact that injection into a normal dog of blood from one in which secretin formation is going on (as a result of acid in the duodenum), excites pancreatic secretion. The pancreatic juice produced by the injection of secretin, like that which is produced under normal conditions, does not contain any active trypsin, but instead contains its precursor, trypsinogen. This becomes converted into trypsin in the intestine, being activated by contact with PHYSIOLOGY OF THE DIGESTIVE GLANDS 427 enterokinase, an enzyme present in the intestinal juice. By such a mechan- ism the mucosa of the pancreatic duct is protected against autodigestion by trypsin. NERVOUS CONTROL OF PANCREAS Prior to the discovery of secretin, Pavlov1 and his pupils had published numerous experiments purporting to show that the secretion of pancreatic Fig. 146. — Pancreatic acini stained with hematoxylin. The acini at the top and to the left of the figure are from a resting- gland, those to the right being from one that had been secreting for over three hours as a result of acid in the duodenum. The lowermost figure is from a gland the vagus nerve supply of which had been stimulated off and on for several hours. Note that the zymogen granules are extruded only after vagus activity but not after secretin activity. (From Babkin, Rubaschkin and Ssawitsch.) juice is controlled through the vagus nerve. The amount of secretion produced by nervous stimulation was, however, never found to be so large as that produced by secretin, and for several years after the discovery of 428 DIGESTION the latter hormone, much doubt existed as to the correctness of Pavlov's claim. As in many other fields of physiologic science,, investigators at- tempted to show that one or the other mechanism obtained, and they were not inclined to consider the possibility that both mechanisms might exist side by side. That such is the case, however, is clear from the most recent work, in which it has been found that if proper precautions are taken, repeated stimulation of the vagus nerve does call forth a secretion of pancreatic juice which, besides being less copious than that following II. III. 'Fig. 147. — Three preparations of pancreatic acini stained by eosin orange toluidin blue. The acini of Fig. I were from a gland after vagus stimulation, and it is noted that besides free ex- trusion of the granules, globules staining with orange (and appearing in deep black in the photo- graph) have formed and may be present in the ductules. Some of the globules, however, change in their staining properties, becoming light red (dark gray in photograph). The acini in II and III were from glands excited by secretin. No globules appear; the granules remain, and fine canaliculi appear in the clear protoplasm. (From Babkin, Rubaschkin and Ssawitsch.) secretin injection, differs from it in the important fact that it contains not trypsinogen but active trypsin. Since the normal pancreatic juice contains trypsinogen, this last mentioned fact would appear to indicate that vagus control of the normal secretion can not be an important affair. The vagus secretion of pancreatic juice is, moreover, paralyzed by atro- pine, which has no action on the secretin mechanism (cf. Bayliss). PHYSIOLOGY OF THE DIGESTIVE GLANDS 429 The copious secretion of pancreatic juice produced by secretin, on the one hand, and the scanty, thick secretion produced by vagus stimula- tion, on the other, calls to mind similar differences observed in the secre- tion of saliva as the result of chorda-tympani or sympathetic stimulation. It will be remembered that from these latter results it was concluded that there must be secretory and trophic fibers concerned in the control of the activities of gland cells. Interesting corroboration of this conclusion has recently been obtained by histologic examination of the pancreas fol- lowing secretin or vagus activity. After the repeated injection of secre- tin, it is difficult to observe any signs of fatigue in the cells ; the zymogen granules remain practically as numerous as in a resting gland, but in the clear protopfasm of the outer third of the cell, it is said that fine channels of fluid can be seen. Through these channels water is believed to pass from the blood towards the lumen and in its course to carry with it some of the zymogen granules, without, however, changing them. Thus, when the gland cells are stained with eosin and orange, after secretin activity some of the zymogen granules can occasionally be seen! in the lumen of the acini stained with eosin like those in the cell itself. After vagus stimulation the appearances are different ; not only are the granules more freely extruded from the cells, but they undergo a preliminary change; they lose the property of staining with eosin and become stained with the orange, at the same time increasing in size so as to form vacuoles. These vacuoles may wander into the ductules, and when they are present here they are stained by orange (Figs. 146 and 147) (Babkin, etc.7a). Why there should be both a nervous and a hormone control of the pan- creatic secretion is not clear. This- gland, unlike the gastric and salivary glands, is not called upon to become active all of a sudden, and it is dif- ficult to see what could serve as the normal stimulus operating through the nervous pathway. Taking it all in all, it is probably safe to con- clude that the nervous mechanism is relatively unimportant, and that under normal conditions it seldom if ever is called into operation. Cor- roboration for this view is afforded by the fact, above mentioned, that the pancreatic juice produced by vagus stimulation contains active tryp- sin, which is not the case with normal pancreatic juice. CHAPTER L PHYSIOLOGY OF THE DIGESTIVE GLANDS (Cont'd) Up to the present we have been concerned with the physiologic activi- ties of digestive glands in general, but now we must study each of them separately in order to find out the conditions under which .they become stimulated to activity in the normal process of digestion. The secretion of each gland has a definite role assigned to it in the complex and lengthy process of digestion. It takes up its work where the preceding secre- tion left off ; e. g., the pepsin of gastric juice digests protein so far as proteoses and peptone; the trypsin of pancreatic juice then attacks the proteoses and peptone, and the resulting lower degradation products are finally attacked by the erepsin of the intestinal juice. The secre- tions of the various glands are, therefore, required in a certain definite order — they are correlated; and we must now give some attention to the precise condition upon which the activity and correlation depend. To make possible such observations on the normal activities of the glands, a preliminary operation has to be performed so as to bring the duct of the gland to the surface of the body and permit of the observa- tion of its secretory activity after the animal has recovered from the immediate effects of the operation. We owe to Pavlov1 the surgical technic by which these conditions can be fulfilled. The general principle of the operation, in the case of glands provided with ducts, consists in making a circular cut through the mucous membrane surrounding the opening of the duct and then, after dissecting the duct free, stitching the edges of the cut to the skin wound. Healing then takes place without the formation in the duct of any stricture due to the cicatricial tissue. After the wound has healed, the secretion can readily be collected in a receiver attached over the duct fistula, the animal being in every other way in a perfectly normal condition. In the case of glands not provided with a duct, other methods must be adopted to collect the secretions. These will be described elsewhere. 430 PHYSIOLOGY OP THE DIGESTIVE GLANDS 431 THE NORMAL SECRETION OF SALIVA The duct fistula can in this case be made either for the submaxillary gland, representing a mucous gland, or for the parotid, representing a serous gland. Under ordinary conditions there is very little secretion from either duct. When secretion occurs, it is, of course, caused by influences acting on a nerve center or centers in the medulla oblongata, the exact location of which for the different glands has* been worked out in recent years by Miller.9 The impulses acting on these centers may be transmitted along afferent nerves coming from the mucous membrane of the mouth, nares, etc., or by impulses which we may call psychic, trans- mitted from the higher nerve centers. The reflex secretions caused by impulses traveling by the afferent nerve from the mouth, etc., have been called unconditioned, and those from the higher nerve centers, condi- tioned. With regard to the former, there is considerable discrimination in the type of stimulus that will be effective. Thus, if the dog — for most of the experiments have been performed on this animal — is given meat, a secretion of thick, mucous saliva will be observed to occur (submaxil- lary gland). On the other hand, if the meat is dried and pulverized, the secretion which it calls forth will be very copious and watery (par- otid gland). There is, then, an obvious association between the nature of the secretion and the function it will be called upon to perform when it becomes mixed with the food. The mucous secretion called forth by meat will serve to lubricate the bolus of food and thus facilitate its swallowing, whereas the thin watery secretion produced by the dry powder will have the effect of washing the powder from the mouth. It is evident that the mechanical condition of the food partly deter- mines its exciting quality. Mechanical stimulation of the mucosa alone is, however, not an adequate stimulus, for if pebbles are placed in the mouth, little secretion occurs, but if sand is placed in the mouth, secretion immedi- ately becomes copious. The nerve endings also respond to chemical stimuli. Thus, weak acid causes a copious secretion, while alkali has no effect; disagreeable, nauseous substances also excite secretion. The above dif- ferences in the response of the glands according to the mechanical condi- tion of the food has been observed only in the case of the parotid gland, increase in the submaxillary secretion being obtained only when actual foodstuffs are placed in the mouth. The investigations that have been made on the conditions of psychic secretion of saliva are still more interesting and important. Their im- portance depends not so much on the information they give us concern- ing the secretion of saliva as such, as on the methods they furnish us for investigating the various conditions that affect the psychic processes 432 DIGESTION associated with the taking of food. It is from the psychic rather than from the physiologic standpoint, therefore, that these observations are of importance, for they permit us, by objective methods, to study on dumb animals problems that would otherwise be beyond our powers of investigation. Many of the results, with their bearing on the functions of the higher nerve centers, have been discussed elsewhere. Meanwhile, however, even at the risk of repetition it may not be out of place to cite a few of the most interesting experiments. If we tease a hungry animal with food for which he has a great appe- tite, a copious secretion of saliva immediately occurs. If we go on teas- ing him without giving him food, and repeat this procedure on several succeeding days, it will be found that gradually he no longer responds to the teasing by increased salivation. Evidently, therefore, the reflex is conditioned upon the animal's afterward receiving the food. The experiment may be performed in another way. If, for example, we offer the animal some food for which he has no appetite, no secre- tion of saliva will occur; but, if at the end of the process we give him appetizing food, it will be found after repeating this procedure on several successive days that the presentation of the unappetizing food calls forth a secretion. He has learned to associate the presentation of unappetizing food with the subsequent gratification of his appetite. The experiment can even be performed so that a definite interval of time elapses between the application of the stimulus and the salivation: if the animal is teased on successive days with food for which he has an appetite but is not given the food until after ten or twenty minutes, presentation of this food will come to be followed by salivation — not immediately, but after the exact interval of time that had been allowed to intervene in the training process. During this interval there must be an inhibition of psychic stimulation of the salivary centers by other nerve centers. It is of great interest that this inhibition may itself be inhib- ited by various forms of stimulation of the nervous system (see page 858). THE SECRETION OF GASTRIC JUICE Methods of Investigation There being no common duct, the secretion of the gastric glands is a much more difficult problem to investigate than is that of glands which, like the salivary, are supplied with ducts. One of the most interesting chapters in the history of physiology concerns the methods which from time to time have been evolved for the collection of this juice and for studying the digestive processes in the stomach. Prominent among the problems confronting the earlier investigators was the question whether PHYSIOLOGY OF THE DIGESTIVE GLANDS 433 the main function of the stomach is to crush or triturate the food or to act on it chemically. The great French scientist Reaumur and a little later the Italian Abbe Spallanzani (1729-1799) attacked this problem by methods that anticipate those of Rehfuss and Einhorn. Spallanzani ulti- mately devised the method of swallowing small perforated wooden tubes containing foodstuffs and covered by small linen bags. After the bags were passed per rectum, he found that considerable erosion or digestion of the food had occurred, but that the wooden tubes, however thin- walled they might be, were not crushed. In order to secure samples of the gastric juice free from food, the only method available to the older investigators consisted in swallowing sponges attached to threads, which after being for some time in the stomach were withdrawn and squeezed dry of juice. The next great contribution came from this country, where, in 1833, Dr. Beaumont, while a surgeon in the service of the American troops located at Mackinaw, made observations on a Canadian voyageur by the name of Alexis St. Martin, who by the premature discharge of his gun had wounded himself in the stomach, the wound never healing but leav- ing a permanent gastric fistula. Beaumont arranged to keep Alexis St. Martin in his service for several years, during which time he made numerous observations on the process of digestion in the stomach — observations many of which are of great value even at the' present day. By none of these methods, however, could a sample of pure gastric juice be secured while the digestive process was actually in progress. To make the collection of such a sample possible, Heidenhain devised a method of isolating portions of the stomach wall as pouches opening through fistulae on the abdominal wall. The results of Heidenhain 's experiments are, however open to the objection that the secretion in the isolated pouches may not really correspond to that occurring in the main stomach, since the connections of the pouches with the central nervous system must have been severed. In order that these connec- tions might remain as nearly intact as possible, the Russian physiologist, Pavlov,1 devised an ingenious operation in which the pouch, or "minia- ture stomach," remains connected with the main stomach through a con- siderable width of mucous and submucous tissue. The essential nature of this operation will be evident from the accompanying diagram. (Fig. 148). The most recent investigations have been made by Cannon3 and by Carlson.4 The former fed animals food impregnated with bismuth sub- nitrate, and then exposed the animal to the x-rays. A shadow is produced by the food mass in the stomach, and from the changes in the outline of this shadow facts have been collected, not only concerning the 434 DIGESTION movements of the viscus, but also concerning the rate of discharge of food into the intestine and therefore the duration of the gastric digestive process. Carlson's contribution has been rendered possible by his good fortune in having in his service a second Alexis St. Martin, a man with complete closure of the esophagus and a gastric fistula large enough to permit of direct inspection of the interior of the stomach. Seizing the opportunity thus presented, Carlson during the last four or five years has devoted his attention exclusively to a thorough investigation, not only of the movements of the stomach, but also of the rate of secretion of the gastric juice under different conditions. He has also, with praise- worthy enthusiasm and keen scientific spirit, extended his observations both on laboratory animals and on himself and his coworkers, so as not Fig. 148. — Diagram of stomach showing miniature stomach (S) separated from the main stomach {V) by a double layer of mucous membrane. A. A., is the opening of the pouch on the abdominal wall. (Pavlov.) to incur the error, which is all too frequently made, of confining the observations to one animal. The Nervous Element in Gastric Secretion The first stimulus to the secretion of gastric juice is nervous in origin, and is dependent on the gratification of the appetite and the pleasure of taking food. This fact, after having been suggested by observations made in the clinic, was first thoroughly investigated by Pavlov, who for this purpose observed the gastric secretion flowing either from a fistula of the stomach itself, or from a "miniature stomach," in dogs in which also an esophageal fistula had been established. When food was given by mouth to these animals, it was chewed and swallowed in the usual manner, but before reaching the stomach, it escaped through the esopha- PHYSIOLOGY OF THE DIGESTIVE GLANDS 435 geal fistula. This experiment is known as that of "sham feeding." Within a few minutes after giving food the gastric juice was found to be secreted actively, and if the feeding process was kept up, which could be done almost indefinitely since the animal never became satisfied, the secretion continued to flow. Thus, in one instance Pavlov succeeded in collecting about 700 c.c. of gastric juice after sham feeding an animal for five or six hours in the manner above described. After the stomach has emptied itself of the food taken with the pre- vious meal, it is said by Pavlov to contain only a little alkaline mucus. The more recent work of Carlson, however, shows that this is not strictly the case, there being more or less of a continuous secretion of gastric juice in the entire absence of food. The amount varies from a few c.c. up to 60 c.c. per hour, more secretion being produced when it is collected every five or ten minutes than if it is collected every thirty or sixty, thus indicating that, ordinarily, some escapes through the pylorus into the duodenum. The secretion contains both pepsin and hydrochloric acid. As to the cause of this continuous secretion, little is known. It may be an example of the periodic activities of the digestive glands described by Boldyreff, or it may in part be due to a psychic stimulation dependent upon the thought of food. That the latter is probably not the cause, is indicated by the fact that, at least in Carlson's patient, the psychic juice could not be made to flow short of giving food. The sham feeding causes stimulation of the gastric secretion through impulses transmitted to the stomach along the vagus nerves; for it has been found, in animals in which the vagus nerve has been cut, that the sham feeding no longer induces a secretion of gastric juice. The ques- tion therefore arises as to how the nerve center is stimulated. Three possible causes may be considered: (1) mechanical stimulation of the sensory nerves of the mouth; (2) chemical stimulation of the nerves; (3) the agreeable stimulation of the taste buds and olfactory endings concerned in the tasting of food. In investigating these possibilities, mechanical stimulation was readily ruled out by showing that mere taking of solid matter in the mouth did not excite any secretion, although it might cause a flow of saliva. Mere chemical stimulation could not be the cause, for no secretion was induced by placing substances such as acetic acid or mustard oil in the mouth. By exclusion, then, it would appear that the adequate stimulus must consist in the agreeable stimula- tion of the taste buds, etc. — that is to say, in the gratification of appetite. Further justification for this conclusion was readily secured by noting that foodstuffs for which the animal had no particular desire or appe- tite failed to excite the secretion. Most dogs, for example, although they may take it, are not particularly fond of bread, and when fed with 436 DIGESTION it, these animals did not produce any appetite juice. In one animal that showed considerable liking for bread, active secretion occurred when he was fed with this foodstuff. Pavlov further noted that usually it was not necessary actually to allow the animal to take the food into his mouth, but that mere teasing with savory food was sufficient to cause the secretion, and that in highly sensitive animals even the noises and other events usually asso- ciated with feeding time were sufficient to excite the secretion. In the case of a hungry animal, the mere approach of the attendant with food, or some other noise or action definitely associated with feeding time, was sufficient to excite the secretion. The appetite juice when started was found to persist for some time after the stimulus causing it had been removed. Carlson has succeeded in confirming in man most of these observa- tions. He noted, however, that the secretion produced by seeing or smelling or thinking of food is much less than would be expected from Pavlov's observations on dogs. Even when his subject was hungry, Carlson did not observe that the bringing of a tray of savory food into the room caused any secretion of gastric juice. It is, of course, to be expected that the quantity of the psychic secretion will not be the same in different individuals. It has been observed, for example, by Pavlov to vary considerably in the case of dogs, and it is very likely that it will vary still more in man, with his more highly complicated nervous system. In no case could Carlson observe any secretion of gastric juice produced by having his patient chew on indifferent substances, or by stimulating the nerve endings in the mouth by substances other than those directly related to food. In man the rate of secretion is proportional to the palatability of the food, the smallest amount, during twenty minutes' mastication of pal- atable food, being 30 c.c. and the largest 150 c.c., in a series of 156 obser- vations. A typical curve showing the amount of the secretion is given in Fig. 149. To construct this curve the gastric juice was collected dur- ing five-minute intervals while the man was chewing a meal of average composition and of his own choice. An interesting feature depicted on this curve is that the secretion rate was highest in the last five-minute period, this being the time during which the dessert was being taken, for which this man had a great relish. Quite clearly there was a direct relation between the rate of the secretion of the appetite juice and the palatability of the food. It will further be observed that it took only from fifteen to twenty minutes after discontinuing the chewing before the juice returned to its original level. The practical application of these facts in connection with the hygiene PHYSIOLOGY OF THE DIGESTIVE GLANDS 437 of diet and the feeding of patients during convalescence, is obviously very great. However perfect in other regards a diet may be, it will probably fail to be digested at the proper rate unless it is taken with relish. Frequent feeding with favorite morsels is more likely to be fol- lowed by thorough digestion and assimilation than occasional stuffing with larger amounts. We see too in these experiments an explanation of the well-established practice of starting a meal with something savory. A hors d'oeuvre is nothing more than a physiologic stimulant to appetite. It is also interesting from a practical standpoint to observe that with those who have a keen relish for sweetmeats the taking of des- sert has a real physiologic significance, for, as in Carlson's patient, it stimulates toward the end of a meal a further secretion of the gastric 25' SO' Chewing food Fig. 149. — Typical curve of secretion of gastric juice collected at 5-minute intervals on mas- tication of palatable food for 20 minutes. The rise in secretion during the last 5 minutes of mastication is due to chewing the dessert (fruit) for which the person had great relish. (From Carlson.) juice, and thus insures a more rapid digestion of the food. Good cooking, it should be remembered, is really the first stage in digestion, and it is the only stage over which we can exercise voluntary control. The Hormone Element in Gastric Secretion Although gastric digestion is initiated by the appetite juice, it is clear that this alone can not account for all the secretion that occurs during normal gastric digestion. After an ordinary meal gastric diges- tion lasts usually about four hours, whereas we have seen, particularly from Carlson's observations, that the appetite juice lasts only for some fifteen or twenty minutes after the exciting stimulus has been removed. The appetite juice, in other words, serves only to initiate the process of secretion, and the question arises, What keeps up the secretion during the rest of gastric digestion? The answer was furnished by Pavlov, who 438 DIGESTION observed animals in which not only a miniature stomach had been made, but a fistula into the main stomach as well. The behavior of the secre- tion of gastric juice as a whole could be followed by collecting that which was secreted in the miniature stomach, for it was shown, in con- trol experiments, that this secretion runs strictly parallel with that in the main stomach, being quantitatively a definite fraction of it — accord- ing to the relative size of the miniature stomach — and qualitatively identical. The miniature stomach, in other words, mirrors the events of secretion in the main stomach. It was observed that when the animal was allowed to take the food into the main stomach by the mouth and esophagus, the secretion from the miniature stomach continued to flow until the process of gastric digestion had been completed, a result which was quite different from that obtained after sham feeding. The only possible explanation for this result is that the food in the stomach sets up secretion as a result of local stimulation. To investigate the nature of this local stimulation, whether mechanical or chemical, food and other substances were placed in the main stomach through the gastric fistula without the animal's knowledge so as to avoid possible psychic stimulation, and the secretion observed from the miniature stomach. When' the mucous membrane of the main stomach was stimulated mechanically, as by placing inert objects such as a piece of sponge or sand in the stomach, no secretion occurred. Evidently, therefore, the stimulus is dependent upon some chemical quality of the food. By introducing various foods it was found that there is considerable difference in the degree to which they can excite the secretion. "Water, egg white, bread and starch, were all found to have very little if any effect. On the other hand, when protein that had been partly digested by means of pepsin and hydrochloric acid was introduced into the stomach, it immediately called forth a secretion. The conclusion is that the partly digested products, even of insipid food, are capable of directly exciting the secretion. These include proteoses and peptones, and it was, therefore, of great interest to find that a solution of commercial peptone is also an effective stimulus. This is a result of deep significance, for it indicates that the food which has been partially digested by the appetite juice will serve as a stimulus to continued secretion. The psychic juice has been aptly called the "ignition juice," because by producing partial digestion it serves to ignite the process of gastric secretion. Experimental evidence of its great importance in gastric digestion was secured by Pavlov in experiments in which he placed weighed quantities of meat attached to threads in the stomach through a gastric fistula, and after some time removed them and determined by PHYSIOLOGY OF THE DIGESTIVE GLANDS 439 the difference in weights the extent to which they had become digested. It was found that when the appetite juice was excited by sham feeding at the same time that food was placed directly in the stomach, its diges- tion was much more rapid than in cases in which it was placed in the stomach without the animal's knowing, as when he was asleep. Other foods having a direct stimulating effect on the gastric secre- tion are meat extracts and, to a certain extent, milk. This effect of meat extract is interesting in connection with the practice of taking soup as a first or early stage in dining. It not only excites the appetite juice, but also serves as a direct stimulus to the gastric secretion. As to the nature of the -mechanism by which this direct secretion takes- place, it was shown by Popielski10a that the secretion still occurs after all the nerves proceeding to the stomach are cut. Evidently, therefore, it is independent of the extrinsic nerve supply of the viscus. As a result of his experiments Popielski concluded that the secretion must depend on a local reflex mediated through the nerve structures present in the walls of the stomach itself. Another explanation of the result has, however, in recent years been given more credence by the experiments of Bayliss and Starling on the influence of hormones on the secretion of pancreatic juice (cf. page 425). Edkins10 suggested that a similar process in the stomach might account for the continued secretion of gastric juice. To test the possibility this investigator, after ligating the cardiac sphincter in anesthetized animals, inserted a tube into the pyloric end of the stomach, through which he placed in the stomach about 50 c.c. of physiologic saline. After this had, been in the stomach for an hour, he found that no water was absorbed, and that if the fluid was removed after this time, it contained neither hydrochloric acid nor pepsin. On the other hand, if during the time the saline was in the stomach a decoction of the mucous membrane of the pyloric end, made either with peptone solution or with a solution of dextrine, was injected intravenously in small quantities every few minutes, it was found that the saline contained distinct quantities of hydrochloric acid and pepsin. Furthermore, it was found that, if the peptone solution or the dextrine solution alone was injected intravenously, there was no such evidence of gastric secretion. The conclusion which Edkins drew from his experi- ments is to the effect that the half-digested products of the earlier stages of gastric digestion act on the mucous membrane of the stomach so as to produce a hormone, which is then carried by the blood to the cells of the gastric glands, upon which, like secretin, it directly develops an exciting effect. This hormone has been called gastrin. These observa- tions of Edkins have been confirmed, and they explain very simply how gastric secretion is maintained after the cessation of the secretion of the 440 DIGESTION appetite juice.10 By such a mechanism gastric juice would continue to be secreted so long as any half-digested food remains in the stomach. The action of gastrin is the first instance of a hormone control of the digestive glands. In the earlier stages of digestion, the secretion of saliva and appetite juice is mediated through the nervous system, because these juices must be produced promptly. In the later stages of gastric diges- tion, such promptitude in response on the part of the gland is no longer necessary, so that the slower, more continuous process of hormone con- trol is sufficient. Quantity of Gastric Juice Secreted According to Carlson, the total amount of gastric juice secreted in man on an average meal composed of meat, bread, vegetables, coffee or Hours 12345678123456789 10 123456 Flesh. 200 gm. Bread, 200 gm. Milk, 600 c.c. Fig. 150. — Cubic centimeters of gastric juice secreted after diets of meat, bread, and milk. (From Pavlov.) milk, and dessert, amounts to about 700 c.c., being divided into 200 c.c. in the first hour, 150 in the second, and 350 c.c. during the third, fourth and fifth hours. These figures were estimated partly on the basis of observations made on the man with the gastric fistula, and partly from the data supplied by Pavlov's observations on dogs. Carlson believes that Pavlov overestimated the relative- importance of the appetite juice in gastric digestion. He found, for example, that after division of both vagus nerves in dogs normal gastric digestion might be regained a few days after the operation, although, of course, under such circumstances no appetite juice could have been secreted. Moreover, he observed that cats when forcibly fed with unpalatable food may digest that food as rapidly as when they eat voluntarily. In support of his contention, Carlson states that he has frequently, removed all of the appetite juice from his patient's stomach before the masticated meal was put into it. without any evident interference with the digestive process. Fat has a distinct inhibiting influence on the direct secretion of gas- PHYSIOLOGY OF THE DIGESTIVE GLANDS 441 trie juice; cream takes considerably longer to be be digested than milk, and the presence of oil in the stomach delays the secretion of juice poured out on a subsequent meal of otherwise readily digestible food. By col- lecting all of the gastric juice from the miniature stomach after feeding by mouth with quantities of different protein-rich foods containing the same quantities of nitrogen, interesting observations have been recorded concerning the amount of juice secreted and its proteolytic power. The results of some of the experiments are shown in the accompanying curves (Figs. 150 and 151). It will be seen that the most abundant secretion occurs with meat, that of milk being not only smaller but also slower in starting. The digestive power is greatest in the case of bread. Hours 123456782345678923*56 10.0 e An Mm of-Protei Column o .M -•**• -0* J 0 0 0 0 < / \ i / \ \ ^ ~~~^ ^~~ / \ t * ? ^ \ / \ / \ / \ ~^- '^ j >. ^ Flesh, 200 gm. Bread, 200 gm. Milk, 600 c.c. Fig. 151. — Digestive power of the juice, as measured by the length of the protein column digested in Mett's tubes, with diets of flesh, bread, and milk. (From Pavlov.) THE INTESTINAL SECRETIONS Pancreatic Juice Eegarding the natural secretion of pancreatic juice, little need be added to what has already been said (see page 425) . The secretion begins when the chyme enters the duodenum, and attains its maximum when the outflow of this is greatest. By collecting the juice from, a permanent fistula of the pancreatic duct, it has been found that the amount varies with different foods. When quantities of food containing equivalent amounts of nitro- gen are fed, the greatest secretion is said to occur with bread and the least with milk. Such differences are probably dependent upon the amount of acid secreted in the stomach and passed on into the duodenum. It was thought at one time that, besides variation in quantity, the nature of the enzymes in the pancreatic juice might vary according to the kind of food. This, however, has been shown not to be the case. 442 DIGESTION Bile The secretion of bile runs practically parallel with that of pancreatic juice. The liver is producing bile more or less continuously, since besides being a digestive fluid it is also an excretory product. The bile produced between the periods of digestion is mainly stored in the gall bladder. When the acid chyme comes in contact with the duodenal mucous mem- brane, it excites afferent nerve endings that cause a reflex contraction of the gall bladder, and this expresses some of the bile into the duodenum. The secretin, which the acid at the same time produces, besides affecting the pancreas, acts on the liver cells, stimulating them to the increased secretion of bile. Thus, by a nervous reflex operating on the gall bladder and later by a hormone mechanism operating on the liver cell, the increased secretion of bile is insured throughout digestion. Of the bile discharged into the intestine, a certain proportion of the bile salts is reabsorbed into the portal blood. When these arrive at the liver they also excite secre- tion of bile, thus assisting secretin in maintaining the secretion through- out the process of intestinal digestion. Fig. 152. — Loop of intestine after tying off the portions, cutting the nerves running to the middle portion, and returning the loop to the abdomen for some time. (From Jackson.) Intestinal Juice The secretion of intestinal juice, or succus entericus, can obviously be studied only after isolating portions of the intestine and connecting them with fistulse of the abdominal walls. It appears here again that both a nervous and a hormone mechanism exist. Mechanical stimulation of the intestinal mucous membrane causes an immediate outflow of intestinal juice, the purpose of which under normal conditions is evidently to assist in moving forward the bowel contents. This mechanically excited juice does not contain any enterokinase and only small amounts of the other enzymes. Further evidence for nervous control of the secretion of intes- tinal juice has. been obtained by isolating three pouches of intestine be- PHYSIOLOGY OF THE DIGESTIVE GLANDS 443 tween ligatures, and then denervating the central pouch by carefully cutting all the nerves without wounding the blood vessels. On returning the pouches to the abdomen and leaving them several hours, it has been found that the middle pouch becomes distended with secretion, whereas the two end pouches remain empty (Fig. 152). If the pouches are left for several days in the abdomen, however, the secretion from the denervated portion disappears again. The explanation of the result is possibly that the nerves under ordinary conditions convey impulses to the intestinal glands, which tonically inhibit their activity. The existence of hormone control is evidenced by the fact that no enteroldnase is present in the intestinal juice unless pancreatic juice is placed in contact with the mucous membrane. Injection of pancreatic juice into the blood, however, does not cause any secretion of intestinal juice ; whereas the injection of secretin has such an effect. CHAPTER LI THE MECHANISMS OF DIGESTION MASTICATION, DEGLUTITION, VOMITING Mastication By the movements of the lower jaw on the upper, the two rows of teeth come together so as to serve for biting or crushing the food. The resulting comminution of the food forms the first step in digestion. The up and down motion of the lower jaw results in biting by the incisors, and after the mouthful has been taken, the side to side movements enable the grinding teeth to crush and break it up into fragments of the proper size for swallowing. The most suitable size of the mouthful is about 5 c.c., but this varies greatly with habit. After mastication, the mass weighs from 3.2 to 6.5 gm., about one-fourth of this weight being due to saliva. The food is now a semifluid mush containing particles which are usually less than 2 mm. in diameter. Some, however, may measure 7 or even 12 mm. Determination of the proper degree of fineness of the food is a func- tion of the tongue, gums, and cheeks, for which purpose the mucous membrane covering them is supplied with very sensitive touch nerve endings (see page 794). The sensitiveness of the tongue, etc., in this regard explains why an object which can scarcely be felt by the fingers seems to be quite large in the mouth. If some particles of food that are too large for swallowing happen to be carried backward in the mouth, the tongue returns them for further mastication. The saliva assists in mastication in several ways: (1) by dissolving some of the food constituents; (2) by partly digesting some of the starch; (3) by softening the mass of food so that it is more readily crushed; (4) by covering the bolus with mucus so as to make it more readily transferable from place to place. The secretion of saliva is therefore stimulated by the chewing movements, and its composition varies according to the nature of the food (page 431). In some animals, such as the cat and dog, mastication is unimportant, coating of the food with saliva being the only change which it undergoes in the mouth. In man the ability thus to bolt the food can readily be acquired, not, however, without some detriment to the efficiency of digestion as a whole. Soft 444 THE MECHANISMS OF DIGESTION 445 starchy food is little chewed, the length of time required for the mastica- tion of other foods depending mainly on their nature, but also to a certain degree on the appetite and on the size of the mouthful. It can not be too strongly insisted upon that the act of mastication is of far more importance than merely to break up and prepare the food for swallowing. It causes the food to be moved about in the mouth so as to develop its full effect on the taste buds; the crushing also releases odors which stimulate the olfactory epithelium. On these stimuli depend the satisfaction and pleasure of eating, which in turn initiate the process of gastric digestion (see page 435). The benefit to digestion as a whole of a large secretion of saliva, brought about by persistent chewing, has been assumed by some to be much greater than it really is, and there has existed, and indeed may still exist, a school of faddists who, by deliberately chewing far beyond the necessary time, imagine themselves to thrive better on less food than those who occupy their time with more profitable pursuits. Deglutition or Swallowing After being masticated the food is rolled up into a bolus by the action of the tongue against the palate, and after being lubricated by saliva is moved, by elevation of the front of the tongue, towards the back of the mouth. This constitutes the first stage of swallowing, and is, so far, a voluntary act. About this time a slight inspiratory contraction of the diaphragm occurs — the so-called respiration) of swallowing — and the mylohyoid quickly contracts, with the consequence that the bolus passes between the pillars of the fauces. This marks the beginning of the second stage, the first event of which is that the bolus, by stimulating sensory nerve endings, acts on nerve centers situated in the medulla oblongata so as to cause a coordinated series of movements of the muscles of the pharynx and larynx and an inhibition for a moment of the respiratory center (page 332). The movements alter the shape of the pharynx and of the various openings into it in such a manner as to compel the bolus of food to pass into the esophagus (see Fig. 153) : thus, (1) the soft palate becomes elevated and the posterior wall of the pharynx bulges forward so as to shut off the posterior nares, (2) the posterior pillars of the fauces ap- proximate so as to shut off the mouth cavity, and (3) in about a tenth of a second after the mylohyoid has contracted, the larynx is pulled up- wards and forwards under the root of the tongue, which by being drawn backwards becomes banked up over the laryngeal opening. This pulling up of the larynx brings the opening into it near to the lower half of the dorsal side of the epiglottis, but the upper half of this struc- 446 DIGESTION ture projects beyond and serves as a ledge to guide the bolus safely past this critical part of its course. (4) As a further safeguard against any entry of food into the air passages, the laryngeal opening is narrowed by approximation of the true and the false vocal cords. So far the force which propels the bolus is mainly the contraction of the mylohyoid, assisted by the movements of the root of the tongue. When it has reached the lower end of the pharynx, however, the bolus readily falls into the esophagus, which has become dilated on account of a reflex inhibition of the constrictor muscles of its upper end. This so- called second stage of swallowing is, therefore, a complex coordinated movement initiated by afferent stimuli and involving reciprocal action of various groups of muscles: inhibition of the respiratory muscles and Fig. 153. — The changes which take place in the position of the root of the tongue, the soft palate, the. epiglottis and the larynx during the second stage of swallowing. The thick dotted line indicates the position during swallowing. of those that constrict the esophagus, and stimulation of those that elevate the palate, the root of the tongue, and the larynx. It is purely an involuntary process. The third stage of deglutition consists in the passage of the swallowed food along the esophagus. The mechanism by which this is done de- pends very much on the physical consistence of the food. A solid bolus that more or less fills the esophagus excites a typical peristaltic wave, which is characterized by a dilatation of the esophagus immediately in front of and a constriction over and behind the bolus. This wave travels down' the esophagus in man at such a rate that it reaches the cardiac sphincter in about five or six seconds. On arriving here the cardiac THE MECHANISMS OF DIGESTION 447 sphincter, ordinarily contracted, relaxes for a moment so that the bolus passes into the stomach. In many animals, including man and the cat, the peristaltic wave travels much more rapidly in the upper part of the esophagus than lower down because of differences in the nature of the muscular coat, this being of the striated variety above, and of the non- striated below. The purpose of more rapid movement in the upper part is no doubt that the bolus may be hurried past the regions where, by distending the esophagus, it might interfere with the function of neigh- boring structures, such as the heart. In other animals, as the dog, the muscular fiber is striated all along the esophagus, and the bolus of food correspondingly travels at a uniform, quick rate all the way. It takes only about four seconds for the bolus to reach the stomach in the dog. The peristaltic wave of the upper part of the esophagus in the cat and presumably in man, unlike that of the intestines (see page 466), is trans- mitted by the esophageal branches of the vagus nerves. If these are severed, but the muscular coats left intact, the esophagus becomes dilated above the level of the section and contracted below, and no peristaltic wave can pass along it ; on the other hand, the muscular coat ,may be severed (by crushing, etc.) but the peristaltic wave will continue to travel, provided no damage has been done to the nerves. In the lower part of the esophagus, however, the wave of peristalsis, like that of the intestines, travels independently of extrinsic nerves. This has been observed in animals in which all of the extrinsic nerves have been cut some time previously. This difference between the upper and the lower portions is associated with the difference in the nature of the muscular fibers above noted (Meltzer).11 The propagation of the wave by the nerves in, the upper part of the esophagus indicates that the second stage and the first part of the third stage of deglutition must be rehearsed, as it were, in the medullary centers from which arise the nerve fibers to the pharynx and the upper levels of the esophagus. It is thought that the discharges from these local centers are controlled by a higher swallowing center situated in the medulla just above that of respiration, the afferent stimuli to which proceed from. the pharynx by the fifth, superior laryngeal, and vagus nerves. The exact location of the sensory areas whose stimulation is most effective in initiating the swallowing reflex varies considerably in different animals. In man it is probably at the entrance to the pharynx; in the dog it is on the posterior wall. A foreign body placed directly in the upper portion of the. esophagus of man has been observed to remain stationary until the individual made a swallowing movement. The afferent fibers in the glossopharyngeal nerve exercise a powerful inhibitory influence on the deglutition center as well as on that of respira- 448 DIGESTION tion. Thus, if swallowing movements are excited by stimulating the cen- tral end of the superior laryngeal nerve, they can be instantly inhibited by simultaneously stimulating the glossopharyngeal, and the respiratory movements stop in whatever position they may have been at the time. When the glossopharyngeal nerves are cut, the esophagus enters into a condition of tonic contraction, which may last a day or so. This shows that the inhibitory impulses are tonic in nature. This inhibition of the esophagus is indeed a most important part of the process when liquid or semiliquid food is swallowed. 'By the contrac- tion of the mylohyoid muscle, fluids are quickly shot down the distended esophagus, at the lower end of which, on account of the closure of the cardiac sphincter, they accumulate until the arrival of the peristaltic wave which has meanwhile been set up by stimulation of the pharynx. If the swallowing is immediately repeated, as is usually the case in drinking, the esophagus remains dilated because peristalsis is inhibited, and the fluid lies outside the cardiac orifice until the last mouthful has been taken. The Cardiac Sphincter The passage between the esophagus and the stomach is guarded by the cardiac sphincter or cardia. This exists in a permanently con- tracted state, or tonus, superimposed on which from time to time are rhythmic alternations of contraction and relaxation. This tonus is never very pronounced. In man it is said that a water pressure of from 2 to 7 cm. applied to the esophageal side of the sphincter will drive air or water into the stomach, this pressure being less than that of a column of fluid filling the thoracic esophagus in the erect position. During repeated deglutition the tonus becomes less and less marked, and after a number of swallows the sphincter may become completely relaxed. When this relaxation disappears, however, the sphincter becomes more contracted than usual and remains so for a longer time. The tonic condition of the sphincter is controlled by the vagus nerve, stimulation of which causes relaxation with an after-effect of strong contraction. Mechanical or chemical stimulation of the lower end of the esophagus increases the tonus of the sphincter. Forcing of the sphincter from the stomach side requires a higher pressure than from the esopha- geal. Eructation of gas, for example, does not take place until intra- gastric pressure has risen to about 25 cm. of water. In deep anesthesia, however, intragastric pressure may rise considerably higher without forcing the sphincter. In animals fed with starch paste impregnated with subnitrate of bis- muth and then examined by means of the x-rays, the variation in degree THE MECHANISMS OF DIGESTION 449 of tone of the sphincter has been observed to be responsible for occasional regurgitation of some of the gastric contents into the esophagus up to the level of the heart or even to the base of the neck. The presence of the gastric contents in the esophagus starts a peristaltic wave, which pushes the material back again into the stomach. This peristaltic wave starts in the absence of any other phases of the deglutition process, indicating that it has been excited by the presence of the material in the esophagus itself, and belongs, therefore, to the lower order of peristaltic wave, as seen in the intestines but not in the upper half of the esophagus. Regur- gitation of food into the esophagus occurs only when the intragastric pressure is fairly high. It may last for a period of from twenty to thirty minutes after the meal is taken, and disappears when the tonus of the sphincter becomes increased as a result of the presence in the gastric contents of free hydrochloric acid. Much information has been secured by listening with a stethoscope to the sounds caused by swallowing and 'by observing with the x-ray the shadows produced along the course of the esophagus when food impreg- nated with bismuth subnitrate is taken. When a solid bolus is swal- lowed only one sound is usually heard, but with liquid food there are two, one at the upper end, due to the rush of the fluid and air, and the other at the lower end (heard over the epigastrium), four or six seconds later, due to the arrival here of the peristaltic wave with the accompanying opening of the cardiac sphincter and the escape of the fluid and air into the stomach. Sometimes, when the person is in the horizontal position, this second sound may be broken up into several, indicating that, unassisted by gravity, the fluid does not so readily pass through the sphincter. The x-ray shadows yield results in conformity with the above. After swallowing milk and bismuth, for example, the shadow falls quickly to the lower end of the esophagus and then passes sloAvly into the stomach. When the passage of a solid bolus is watched by the x-ray method, its rate of descent will be found to depend on whether or not it is well lubricated Avith saliva; if not so, it may take as long as fifteen minutes to reach the stomach; if moist, but from eight to eighteen seconds. Vomiting , Vomiting is usually preceded by a feeling of sickness or nausea, and is initiated by a very active secretion of saliva. The saliva, mixed with air, accumulates to a considerable extent at the lower end of the esopha- gus, which it distends. A forced inspiration is now made, during the first stage of which the glottis is open so that the air enters the lungs, but later the glottis closes so that the inspired air is sucked into the 450 DIGESTION esophagus, Avhich, already somewhat distended by saliva, now becomes markedly so. The abdominal muscles then contract so as to compress the stomach against the diaphragm and, simultaneously, the cardiac sphincter relaxes, the head is held forward and the contents of the stomach are ejected through the previously distended esophagus. The compression of the stomach by the contracting abdominal muscles is assisted by an actual contraction of the stomach itself, as has been clearly demonstrated by the x-ray method. After the contents of the stomach itself have been evacuated, the pyloric sphincter may also relax and permit the contents (bile, etc.) of the duodenum to be vomited. The act of vomiting is controlled by a center located in the medulla, and the afferent, fibers to this center may come from many different regions of the body. Perhaps the most potent of them come from the sensory nerve endings of the fauces and pharynx. This explains the tendency to vomit when the mucosa of this region is mechanically stimu- lated. Other afferent impulses come from the mucosa of the stomach itself, and these are stimulated by emetics, important among which are strong salt solution, mustard water and zinc sulphate. Certain other emetics, particularly tartar emetic and apomorphine, act on the vomit- ing center itself, and can therefore operate when given subcutaneously. Afferent vomiting impulses also arise from the abdominal viscera, thus explaining the vomiting wrhich occurs in strangulated hernia, and in other irritative lesions, involving this region. X-ray observations have been made on the movements of the stomach of cats after the admin- istration of apomorphine (Cannon). The first change observed is an inhibition of the cardiac end of the stomach, which becomes a perfectly flaccid bag. About the midregion of the organ, deeper contractions then start up, which sweep from the pylorus, each contraction stopping as a deep ring at the beginning of the vestibule, while a slighter wave con- tinues. A very strong contraction at the incisura angularis finally develops and completely divides the gastric cavity into two parts. On the left of this constriction the stomach remains completely relaxed, but at the right of it waves continue running over the vestibule. It is while the stomach is in this condition that the sudden contraction of the dia- phragm and abdominal muscles shoots the cardiac contents into the relaxed esophagus. As these jerky contractions are continued, the gastric walls seem to reacquire their tone. CHAPTER LII THE MECHANISMS OF DIGESTION (Cont'd) THE MOVEMENTS OF THE STOMACH The Character of the Movements Even from the earliest days it has been recognized that the stomach performs two important functions: (1) receiving the swallowed food and then discharging it slowly into the intestine, and (2) initiating the chemical processes of digestion. In order to understand the mechanism by which the stomach collects and then discharges the food, it is neces- sary first of all to recall certain anatomic facts concerning the organ, and for this purpose it is most convenient to accept the description given by Cannon, which is illustrated in the accompanying figure. The organ is divided into a cardiac and a pyloric portion by a deep notch in the lesser curvature, called the incisura angularis. The cardiac portion is further subdivided into two by the cardiac orifice. The part which lies, in man, above a line drawn horizontally through the cardia is the fundus. The part lying between the fundus and the incisura angularis is known as the body of the stomach, which, when full, has a tapering shape. The pyloric portion lying on the right of the incisura angularis is further divided into two parts: the pyloric vestibule and the pyloric canal, the latter of which lies next the pyloric sphincter and in man measures about 3 cm. in length (see Fig. 154). The filled stomach of a person standing erect is so disposed that the greatest curvature forms its lowest point, which may be considerably below the umbilicus. As digestion proceeds and the stomach empties, the greater curvature becomes gradually raised, so that ultimately the pylorus comes to be the most dependent part of the stomach. From these and many other observations it is certain that the emptying of the stomach does not at all depend on the operation of the force of gravity. Indeed, that this can not be the case is perfectly clear, when we con- sider the disposition of the stomach in quadrupeds. Exact observation on the movements which the stomach performs from the time it is filled with food till it empties, have been made by the x-ray method, first introduced by Cannon.12 The method consists in feed- 451 452 DIGESTION ing the animal with food that has been impregnated with bismuth sub- nitrate, then exposing him to the x-ray and either taking instantaneous photographs of the shadoAvs or observing them by means of a fluorescent screen. The descriptions of the original observations made by Cannon Fig. 154. — Schematic outline of the stomach. At C is the cardia; F, fundus; I A, incisura an- gularis; B, body; PC, pyloric canal; P, pylorus. (From Cannon.) on the stomach of the cat have been so little modified by observations on man that we may take them as a convenient type. In the accompany- ing figure (Fig. 156) the outline of the shadow cast by the stomach is shown at intervals of an hour each during digestion. Soon after the Fig. 155. — Diagrams of outline and position of stomach as indicated by skiagrams taken on man in the erect position at intervals after swallowing food impregnated with bismuth subnitrate. A, moderately full; B, practically empty. The clear space at the upper end of the stomach is due to gas, and it will be noticed that this "stomach bladder" lies close to the heart. (From T. Win- gate Todd.) THE MECHANISMS OP DIGESTION 453 stomach has become filled, peristaltic waves are seen to take their origin about the middle of the body of the stomach, and to course towards the pylorus. Above the region at which these waves originate — that is, the cardiac half of the body of the stomach and all of the fundus — there are no waves, but as digestion proceeds the walls slowly and steadily contract on the mass of food. This so-called cardiac pouch does not, however, diminish in size so rapidly as the part of the body of the stomach over which the peristaltic waves are passing. The circular fibers of the walls of this part of the stomach — sometimes called the gastric tube — contract tonically, so that it becomes tubular in form, with the full cardiac pouch at the left and above and the pyloric por- Fig. 156. — Outlines of the shadows cast by the stomach at intervals of an hour each after feeding a cat with food impregnated with bismuth subnitrate. (From Cannon.) tion at the right. The latter portion meanwhile does not diminish much in size, although the peristaltic waves traveling over it are very pro- nounced. As will be clear from the figure, these changes in outline go on until the cardiac pouch has become practically empty and the food has been all moved along the now tubular portion of the body into the pyloric vestibule. From this description it is evident that the function of the cardiac end is to serve as a reservoir for the food, which, by a slow contraction of the walls, is gradually delivered into the gastric tube, where by peristalsis it is carried towards the pyloric vestibule. The time required for the peristaltic waves to travel from their place of origin to the pylorus is considerably longer than the interval between 454 DIGESTION the waves, so that several of these are always seen on the stomach at the same time. They sometimes become so pronounced in the pyloric region, especially in a half-empty stomach, that they appear almost to obliterate the cavity. They always stop at the pylorus, never going on to the duodenum. The rate of recurrence of the waves varies somewhat in different animals, being about six per minute in the cat and about three in man. Their initiation does not seem to depend on the presence of acid in the gastric contents, for, when food Is introduced into the stomach, they do not wait for the gastric contents to become acid in reaction (see page 482). Nevertheless1, acid does seem somewhat to stim- ulate the depth and frequency of the waves, and they recur oftener with carbohydrate than with fatty food. The pressure in the stomach contents — the intragastric pressure — is low and constant at the cardiac end and fairly high and variable in the pyloric end (in the former from 6 to 8 cm. of water, and in the latter from 20 to 30). Constancy of pressure in the cardiac end indicates that the stomach wall must adapt itself very promptly to the amount of food in the organ. The higher and more variable pressure in the pyloric end is, of course, due to the peristaltic waves, and it is interesting to note that it is sufficient to propel the gastric contents through the pylorus for several centimeters into the duodenum. The Effect of the Stomach Movements on the Food This has been studied: (1) by dividing the food into portions that are differently colored and, after some time, killing the animal, freezing the stomach and making sections of it (see Fig. 157) ; (2) by mak- ing little pellets of bismuth subnitrate with starch and observing their behavior under the x-rays; or (3) by removing samples of the stomach contents by means of a stomach tube (Rehfuss tube) inserted so that its free end lies in either the cardiac or the pyloric region. By the first of the above methods it has been found that the first mouthfuls of food lie along the greater curvature, where they form a layer over which that subsequently swallowed accumulates, with the last por- tions next the cardia. The pepsin and hydrochloric acid of the car- diac end, therefore, act soonest on the first swallowed portion of a meal, and the more recently swallowed central masses are not affected by the secretions for some time, so that opportunity is given for the saliva mixed with the food to develop its digestive action. As has been shown by removing the stomach contents with a tube at various periods after feeding with starchy food, considerable amylolysis may occur for some time. When separate samples are removed in this way from the cardiac and pyloric parts, it has been found that after THE MECHANISMS OF DIGESTION 455 half an hour the contents of both have about the same percentage of sugar, but that for some time after this interval the cardiac contents contain considerably more sugar than the pyloric. Later the percentages of sugar again become about equal, no doubt on account of diffusion. The diastatic action in the fundus is finally brought to an end when the contents become completely permeated by the hydrochloric acid. In this connection it is worthy of note that the addition of hydrochloric acid up to the point of neutrality greatly accelerates the rate of diastatic digestion. As the outer layers of food in the stomach become partly digested on account of the action of the pepsin and hydrochloric acid, the food is slowly pressed into the active right half of the stomach, where by the action of the peristaltic waves it is moved on to the pyloric vestibule. By observing the x-ray shadows cast by two pellets of bismuth subni- trate it has been noted by Cannon that, as the peristaltic wave approaches Fig. 157. — Section of the frozen stomach (rat) some time after feeding with food given in three differently colored portions. (From Howell's Physiology.) a pellet, it causes it to move forward more rapidly for a short distance, but soon overtakes it and in doing so causes the pellet to move back a little towards the fundus. This backward movement is less than the forward movement, so that after the wave has passed, the position of the pellet is a little forward of that which it would have occupied had there been no wave. The behavior of the pellet, and, therefore, of the stomach contents, is very like that of a cork floating at the edge of the sea; as each wave approaches, it hurries the cork on a little, but after its passage the cork recedes again until the second wave carries it still a little farther forAvard. As the peristaltic wave approaches the pyloric vestibule and becomes more powerful its effect on the pellets becomes more marked. They are carried rapidly along this part of the stomach, until the pylorus is reached. If this remains closed, they are shot back into the vestibule. From nine to twelve minutes may elapse before they are transferred to the pylorus from the place where they are first affected by the peristaltic wave. 456 DIGESTION These observations made on cats and other laboratory animals no doubt also apply in the case of man. Removal of the contents of the cardiac and pyloric regions separately with a stomach tube after feeding with a test meal part of which was colored with carmine or charcoal, has shown that none of the coloring material was present in the contents of the pyloric end up to twenty minutes or so after the food had been taken. It then appeared but at first only in traces. Another important distinction between the food in the two portions of the stomach relates to its consistency. In the pyloric end it is semifluid and homogeneous in character; in the cardiac end, on the other hand, it is a lumpy, rather incoherent mass. The gastric movements must greatly facilitate the digestive processes in the stomach. In the cardiac part the undisturbed condition of the food will, as we have seen, facilitate the digestive action of ptyalin, whereas in the body of the stomach the peristaltic waves, besides mov- ing the food onward, will tend to bring fresh portions of mucous mem- brane and food in contact, so that the latter becomes more thoroughly mixed with the pepsin and hydrochloric acid. In the pyloric part, where no hydrochloric acid is secreted, the contents, already sufficiently acid in reaction, become more thoroughly churned up with the local pepsin secretion, so that proteolytic action progresses very rapidly. The peristaltic waves also facilitate absorption from the stomach of such substances as glucose in concentrated solution and, probably, of hydro- lyzed protein ; water, however, is not absorbed. One effect of such absorption is the production of gastrin, which we have seen is the hor- mone concerned in maintaining the gastric secretion after the psychic flow. The fact that the mucosa of the vestibule has, relatively to the cardiac end, feAv secreting glands is in harmony with the view that absorption is an important function of this part of the stomach. THE EMPTYING OF THE STOMACH The Control of the Pyloric Sphincter When digestion has proceeded far enough in the stomach to bring the food into a homogeneous, soup] ike fluid (chyme), portions of this, as they are driven against the pyloric sphincter by -the peristaltic waves, instead of being returned as an axial stream into the stomach, are ejected into the duodenum. We must now consider the mechanism by which the pyloric sphincter opens to permit the passage of the chyme. Bombardment by the peri- staltic waves is evidently not the cause of its opening, for, as we have THE MECHANISMS OF DIGESTION 457 seen, many such waves may arrive at it without this result. Since it is evidently in order that the intestine may not suddenly become over- whelmed with large masses of food that the pylorus only occasionally opens, it might be thought that its opening depends upon the disten- tion of the upper part of the intestine. It is true that excessive disten- tion of the upper part of the intestine does hold the pyloric sphincter closed, but this can not be the physiologic stimulus, because considerable quantities of chyme are never found here. The first clue to the real nature of the mechanism was afforded by observing the behavior of the sphincter Avhen solutions are introduced into the duodenum through a fistula. Acid solutions were found to cause a complete inhibition of gastric evacuation, whereas alkaline solu- tions had no effect. This difference indicates that acids in contact with the duodenal mucous membrane reflexly excite contraction of the sphinc- ter, and that it relaxes only after the acid has become neutralized by mixing with the pancreatic juice and bile. On account of the great importance of the pyloric mechanism in insur- ing that the chyme shall enter the intestine only in such quantities that it can be properly acted upon by the intestinal digesting juices, it will be of interest' to consider briefly some of the experimental observations by -\vhich this mechanism has been studied. We may consider first the evidence that acid on the stomach side of the pylorus causes a relaxation of the sphincter: (1) When carbohydrate food is fed, it ordinarily leaves the stomach fairly rapidly, but if its acid-absorbing power is increased by mixing it with sodium bicarbonate, exit from the stomach is greatly delayed. (2) Proteins ordinarily leave the stomach more slowly than carbohydrates, but if acid proteins are fed, their exit is much more rapid. (3) If a fistula is made into the pyloric vestibule through Avhich some of the contents can be removed, it will be found that just prior to the opening of the pyloric sphincter, a distinctly acid reaction develops in the food ; and furthermore if acid solutions are injected through this fistula, they cause the pyloric sphincter to open, -\vhereas alkalies retard its opening. (4) A similar effect of acid in opening the sphincter can be demonstrated by applying it to the pyloric mucosa of an excised stomach kept alive in oxygenated Ringer's solution. The evidence that acid on the duodenal side causes closure of the sphincter is as follows: (1) When acid is placed in the duodenum through a fistula, the sphincter will not open; (2) when the pancreatic and bile ducts are ligated, the stomach empties much more slowly than normally; and (3) the discharge of protein is considerably hastened if the pylorus is sutured to the intestine below the duodenum. After such an opera- tion it was observed that the protein began to leave the stomach through 458 DIGESTION the pyloric sphincter about the same time as normally, but the subse- quent evacuation was very much accelerated, because no acid came in contact with the duodenal mucosa. Water and egg white may leave the stomach independently of any acid reflex control of the pylorus. By observations made through a duodenal fistula it has been found that, after a quantity of water has been swallowed, most if not all of it very soon enters the duodenum in a more or less continuous stream. It is no doubt on this account that drinking contaminated water is especially dangerous on an empty stomach. The nervous pathway through which these acid reflexes take place has been shown to be the myenteric plexus. Indeed, the whole mechanism is quite analogous with that which we shall see occurs in the intestine during peristalsis: the stimulus, that is, the acid, causes a contraction of the gastric tube behind it and a dilatation in front. Fig. 158. — Outlines of shadows in abdomen obtained by exposure to x-rays 2 hours after feeding with food containing bismuth subnitrate. The food in A was lean beef, and in B boiled rice. The smaller size of the stomach shadow and the much greater total area of the intestinal shadows in R than in A show that carbohydrate leaves the stomach earlier than protein. (From Cannon.) Rate of Emptying of Stomach The relationship of these facts to the rate at which different foodstuffs leave the stomach is very readily explained. The method for investigat- ing this problem, which again we owe to Cannon, consists in feeding ani- mals with- a strictly uniform amount of different foods made up, as nearly as possible, of equal consistency and containing bismuth subni- trate in the proportion of 5 gm. to each 25 c.c. By feeding such mix- tures to cats previously starved for twenty-four hours, and examining the abdomen by the x-ray at regular intervals, the shadows cast by the food after passage into the intestine can be outlined on tracing paper, and the total length* measured (Fig. 158). In taking this as an estimate of the amount of food ine the intestine, several errors are no doubt incurred *This is permissible since the shadows are practically all of the same width. THE MECHANISMS OF DIGESTION 459 on account of the crossing and foreshortening of the loops, etc., but, as their constancy testifies, there is no doubt that the results are sufficiently close for the purpose of finding out how quickly food gains access to the small intestine; and the method has a great advantage over all others in that digestion is allowed to proceed practically without interruption. The points we have to determine are: (1) when the food first leaves the stomach; (2) the rate at which different foods are discharged; (3) the time required for the passage through the small intestine. Let us consider first of all the results obtained by feeding with prac- tically pure fat or carbohydrate or protein. By plotting the length of the shadows in centimeters along the ordinates, with hours along the abscissae, curves such as those shown in Fig. 159 have been secured. When fats were fed (dash line in chart), the discharge began rather slowly, and continued at a slow rate. Even after seven hours some fat still remained in the stomach, and at no time was any large quantity z o k i 34 Hours Fig. 159. — Curves to show the average aggregate length of the food masses in the small intestine at the designated intervals after feeding. The curve for various fat foods is in the dash line, for protein foods in the heavy line, and for carbohydrate foods in the light line. (From Cannon.) present in the intestine, indicating that almost as quickly as it is dis- charged into this part of the gastrointestinal tract fat becomes digested and absorbed. The discharge of carbohydrates was quite different (light line in chart) ; it began often in ten minutes, and soon became abundant, reaching a maximum, as a rule, at the end of two hours, after which it fell off, the stomach being empty in about three hours. Protein left at a rate intermediate between that for fats and that for carbohydrates (heavy line). Little left before the first half hour; the curve then slowly rose, attaining a maximum in about four hours, and then gradu- ally declining at about the same rate as it rose. It is interesting to note that at the end of half an hour about eight times as much carbohydrate had left the stomach as protein; at the end of an hour, five times as much. These results are clearly dependent upon the rates at which the dif- ferent foodstuffs assume an acid reaction in the stomach. Carbohydrate 460 DIGESTION has no combining power for acids, so that the acid secreted with the psychic juice remains uncombined and on gaining the pyloric vestibule excites the opening of the sphincter. Protein, on the other hand, as is well known, absorbs considerable quantities of free hydrochloric acid, so that for some considerable time after it is taken, none of the acid exists in a free state. Fats owe their slow discharge partly to inhibition of gastric secretion, and partly to the longer time it takes for them to become neutralized in the duodenum, because of the fatty acid split off by the action of lipase. Interesting observations have also been made on the rate of discharge when various combinations of foodstuffs were fed. This has been done by feeding one foodstuff before the other, or by mixing the foodstuffs. When carbohydrates were fed first and then protein, the discharge be- gan much earlier than with protein alone, because the carbohydrate food first reached the pyloric vestibule (see page 454). However, at the end of two hours, when the carbohydrate curve should begin to come down, it remained high, indicating that the protein had by this time reached the pylorus and was being discharged at its own rate. When the meat was fed before the carbohydrate, the curve to start with was exactly like that for protein, becoming, however, considerably heightened later when the carbohydrate reached the pyloric vestibule. The presence of protein near the pylorus, therefore, distinctly retards the evacuation of carbohydrate from the stomach. These facts, it will be remarked, all fit in admirably with the observations which we have already detailed concerning the disposition of food in the stomach. When mixtures of equal parts of different foods were fed, the results indicated that the emptying of the stomach occurred at a rate which Avas intermediate between those of the foods taken separately. Mixing protein Avith carbohydrate, for example, accelerated the rate at Avhich protein left, and mixing fats Avith protein caused the protein to leave the stomach considerably more sloAvly than if protein alone had been fed. Influence of Pathologic Conditions on the Emptying An important surgical application of these facts concerns the behavior of food after gastroenterostomy. It has been thought that this operation would cause the food to be drained from the stomach into the intestine and thus leave the region of the stomach betAveen the fistula and the pylorus inactive. This assumption is based on the idea, Avhich AVC have seen to be erroneous, that gravity assists in the emptying of the stomach. As a matter of fact, it has been found that, if the gastroenterostomy is made Avhen there is no obstruction at the pylorus, the chyme takes its THE MECHANISMS OF DIGESTION 461 normal passage through the sphincter and, almost without exception, none leaves by the fistula. When the pylorus is partly occluded, the food sometimes passes in the usual way, and sometimes by the stomach. The cause for this predilection for the pyloric pathway depends on the pressure conditions in the gastric contents. Gastroenterostomy, there- fore, is efficient only .when gross mechanical obstruction exists at the pylorus. The operation should never be performed in the absence of demonstrable organic pyloric disease. Another objection to gastroenterostomy in the presence of a patulous pyloric sphincter rests on the fact that the food, after passing the sphinc- ter and moving along the intestine, may again enter the stomach through the fistula. This is most likely to occur when the stomach is full of food, for under these conditions the stretching of its walls separates the edges of the opening, the intestine being drawn taut between the edges, so that the opening between the stomach and the intestine assumes the form of two narrow slits, which act like valves permitting the food to enter but preventing its escape from the stomach. Only seldom under these circumstances can any food pass into the intestine beyond the stomach opening. Repeated vomiting after gastroenterostomy has been observed in experimental animals only when obstructive kinks or other demonstrable obstacles were present in the gut, the obstruction being lo- cated in that part of the intestine beyond its attachment to the stomach. When the pyloric obstruction is complete, food must, of course, leave by the fistula, digestion by the pancreatic juice and bile being still car- ried on because of the fact that for a considerable distance down the intestine, secretin, which we have seen is essential for the secretion of these fluids, is still produced by the contact of the acid chyme with the intestinal mucosa. Further provision for adequate digestion of food in such cases is secured, as some of the food after leaving the fistula passes back for some distance into the duodenum, where, however, it soon excites peristaltic waves, which again carry it forward. This insures thorough mixing with the digestive juices. From their experi- mental experience Cannon and Blake13 recommend that, when the fistula has to be made, it should be as large as possible and near the pylorus, and that the stomach afterwards should not be allowed to become filled with food. To avoid kinking of the gut, they also recom- mend that several centimeters of the intestine should be attached to the stomach distal to the anastomosis. The effect of hyperacidity of the contents on the emptying of the stomach has been studied by feeding animals with potatoes containing varying percentages of hydrochloric acid. With an acidity of 0.25 per 462 DIGESTION cent, the rate of discharge was increased, but it became slower when the acidity rose to 1 per cent. With an acidity of 0.5 per cent, the rate of discharge was about the normal. Hyperacidity, therefore, causes a retar- dation of the emptying of the stomach. The consistency of the food appears to have little influence on its rate of discharge from the stomach — at least in the case of potatoes. Dilution of protein food, however, increases the rate. Distinctly hard particles in the food retard the stomach evacuation. There is usually a considerable amount of gas in the part of the stomach above the entrance of the cardia, on account of which this part of the stomach has sometimes been called the stomach bladder. In the upright position this gas forms a bright area in the x-ray plate (Fig. 155), but when the person reclines it spreads to a new location. Its presence may influence gastric digestion by preventing the contact of the food with the mucous membrane, and by interfering with the efficiency of the peri- staltic waves in moving the food. Considerable gas therefore retards the emptying of the stomach, as has been shown experimentally by x-ray observations on animals fed with the standard amount of food followed by the introduction of air. It was noted that the air did not diminish the frequency or strength of the peristaltic waves, but that these could not efficiently act on the food. When along with gas there is also atony of the stomach walls, the retardation in the discharge will, of course, be still more pronounced. The temperature of the swallowed food does not appear to have much influence on the stomach movements or on the the rate of discharge from the organ. CHAPTER LIII THE MECHANISMS OF DIGESTION (Cont'd) THE MOVEMENTS OF THE INTESTINES The length of the small intestine and the size of the cecum of the large intestine vary considerably in different animals. In the carnivora, such as the cat, the small intestine is relatively short; in the herbivora, relatively long. Thus, it is three times the length of the body in the cat, and four to six times in the dog ; whereas in the goat and sheep, it may be nearly thirty times the length of the body. In the carnivora the cecum is either absent or rudimentary, whereas in those herbivora which do not have a divided stomach the cecum is very large and sacculated, as is also the colon. The reason for the great size in herbivora is! that practically the whole of the digestion of cellulose takes place in this part of the gut. This digestion, as we shall see later, does not depend on any secretion poured forth by the animal itself, but upon the action of bacteria and of certain enzymes (cytases) that are taken with the vegetable food. Movement of the Small Intestine The movements of the small intestine have been studied (1) by the bismuth subnitrate and x-ray method, (2) by observing them after open- ing the abdomen of an animal submerged in a bath of physiologic saline at body temperature, (3) by observing the changes in pressure produced in a thin-walled rubber balloon inserted in the lumen of the gut and connected with a recording tambour (Fig. 160), and (4) by excising portions of the intestine and keeping them alive in a bath of saline solu- tion at body temperature, through which oxygen is made to pass. THE SEGMENTING MOVEMENTS When a suitably fed animal is placed on the holder for examination by the x-ray method, no movement in the intestinal shadows is generally observed for some time. The first movement to appear is the breaking of one of the columns of food into small segments of nearly equal size. Each of these segments again quickly divides, and the neighboring halves suddenly unite to form new segments, and so on, in a manner 463 DIGESTION which will be made clear by consulting Fig. 161. This rhythmic seg- mentation, as Cannon has called it, continues without cessation for more than half an hour, and the food shadow does not meanwhile seem to change its position in the abdomen to any extent. The splitting up of the seg- ment and the rushing together of the neighboring halves proceed as a rule with great rapidity; thus, if we count the number of different seg- Fig. 160. — Apparatus for recording contractions of the intestine. (From Jackson.) ments during a definite period, we may find the rate of division in the cat to be as high as 28 or 30 a minute. In man the divisions occur at a frequency of approximately 10 per minute, which corresponds to the fre- quency with which sounds can be heard when the abdomen is auscultated. Although half an hour is the period which this process usually oc- cupies, it may last considerably longer. In certain animals, such as the rabbit, segmenting movements have not been observed, but instead THE MECHANISMS OF DIGESTION 465 of them a rhythmic to-and-fro shifting of the masses of food along the lumen of the gut, rapidly repeated for many minutes. When the intestines are floated out in a warm bath of saline solution, it is seen that the rhythmic segmentation is caused by narrow rings of. contraction. Under such conditions also it is often noted that the loops of intestine sway "from side to side. The balloon method also re- veals the presence of slight waves of contraction that pass rapidly along the gut, and follow each other at the rate of twelve to thirteen per minute. Both of the muscular coats of the intestine are involved, and it is believed that the contractions are responsible not only for the pendular move- ments but for the rhythmic segmentation observed by the x-ray method. According to this view these movements are constantly passing along the intestine, and become exaggerated by the mechanical stimulus which is offered by the masses of food to such an extent that they divide the masses into portions. The evidence for this belief rests on- the fact that 1. — Diagrammatic representation of the process of segmentation in the intestine. An ihadpw is shown in / and its segmentation in 2. The dotted lines across each mass osition of division and in 3 is shown how new masses are formed by the split portions ether. (.From Cannon.) Fig. 161. _ unbroken shado show the position 01 division ana i coming together. (.From Cannon.) when the contraction is studied by the balloon method, it becomes marked over the middle of the balloon, where the greatest tension exists. Several functions can be assigned to these movements. They cause intimate mixture of the food with the digestive juices, and by bringing ever new portions of food in contact with the mucosa, they encourage absorption. They also have an important massaging influence on the blood and lymph in the vessels of the intestinal walls. Indeed, the pas- sage of lymph from the lacteals into the mesenteric lymphatics seems to depend very largely upon these movements. THE PERISTALTIC MOVEMENTS The other movement observed in the small intestine is that known as the peristaltic wave. It occurs in two forms: (1) as a slowly advancing con- traction (1 to 2 cm. per minute), preceded by an inhibition of the walls, and proceeding only through a short distance in a coil (4 to 5 cm.); and 466 DIGESTION (2) as a swift movement called the peristaltic rush, which sweeps with- out pause for much longer distances along the canal. Further analysis of the peristaltic -wave can readily be made by the balloon method (Fig. 162). If the gut is pinched above the balloon, a marked relaxation occurs over it, and this relaxation extends for about two feet down the intestine. If, on the other hand, the gut is pinched a little below the situation of the balloon, a long-continued contraction occurs over the latter. The conclusion that we may draw from this result is that the stimulation of the gut causes contraction, above the point of the stimulus and relaxation below, this being known as ''the law of the intestine" — (Bayliss and Starling). We have seen that it applies also in the case of the cardiac and pyloric sphincters. Fig. 162. — Intestinal contractions (balloon method) after excision of the abdominal ganglia and section of both vagi. Mechanical stimulation above (/) and below (2) the balloon causes relaxa- tion and contraction respectively. (From Starling.) THE PHYSIOLOGIC NATURE OF THE RHYTHMIC AND PERISTALTIC MOVEMENTS Interesting information in this connection has been gained by obser- vation of the behavior of the movements after the application of drugs to the gut or after cutting the nerve supply. The rhythmic movements are not affected by the application of nicotine or cocaine. Since these drugs paralyze nervous structures it has been concluded that the rhythmic movements are myogenic in origin. The question is not a settled one, however, for it has been found by Magnus that, although strips of the longitudinal muscle, isolated in oxygenated saline solution, will continue to beat, they do not do so if the adherent Auerbach's plexus of nerves is stripped off from them. The nature of the peristaltic contractions is more definite; they must clearly depend upon a local nervous struc- ture, since they are paralyzed by the application to the gut of cocaine or nicotine. This local nervous system no doubt also resides in Auerbach's plexus, which must therefore be considered as complex enough to be (see THE MECHANISMS OF DIGESTION 467 page 796) endowed with the power of directing nervous impulses so as to bring about relaxation of the gut in front of the stimulus and contrac- tion over it. NERVOUS CONTROL OP MOVEMENTS The influence of the central nervous system on the intestinal movements has been studied by the usual methods of cutting and stimulating the extrinsic nerve supply. Through the splanchnic nerves tonic inhibitory impulses are conveyed to the intestine (except the ileocolic sphincter), for after these nerves' are severed the movements become more distinct. Indeed, in many animals after opening the abdomen no intestinal move- ment can be observed until these nerves have been cut. Stimulation of the peripheral end of the nerve also inhibits any movement which may mean- while be in progress. The impulses through the vagus nerve are of an Fig. 163. — The effect of excitation of both splanchnic nerves on the intestinal contractions. (From Starling.) opposite character. Section of these nerves has little effect, but stimula- tion causes contraction. (Figs. 163 and 164.) By observing the rhythmic contractions of an isolated strip of the small intestine suspended in a bath of oxygenated saline solution at body tem- perature, it can readily be shown that the presence of even a minute trace of epinephrine is sufficient to produce complete inhibition of the movement. The parallelism between the effects of splanchnic stimulation and those of epinephrine injection is very significant, for in this way the marked inhi- bition of intestinal movement which occurs during fright may possibly be explained (see page 736). The circular muscular coat of the last two or three centimenters of the ileum before it joins the cecum is definitely thicker than the rest of this coat, indicating that it has a sphincter-like action. This ileocolic sphincter, as it is called, opens when food is pressed against it from the ileum, but remains closed when food is pressed against it from the cecum. 468 DIGESTION It therefore obeys the law of the intestine. That it is physiologically distinct from the musculature of the rest of the ileum is indicated by the fact that the splanchnic and vagus nerves do not affect it in the same way; thus, stimulation of the splanchnic causes a strong contraction of the sphincter, whereas it is unaffected by stimulation of the vagus. Peristalsis is much more rapid in the duodenum than in other parts of the small intestine. During the first stages of digestion, the food ordi- narily lies mainly in the right half of the abdomen, and later in the left half. There is considerable variation in the time that elapses before it enters the colon. In the cat, carbohydrates reach this part of the gut in about four hours. Fig. 164. — The effect of stimulation of right vagus nerve on the intestinal contractions. (From Starling.) Movements of the Large Intestine On account of the great differences which we have already seen to exist in the size and relative importance of the colon as a digestive organ in different classes of animals, it is not surprising that the movements observed are very different according to the dietetic habits of the animal. Apparently the movements are much the same in the cat as in man. As the food passes through the ileocolic sphincter into the cecum and accumulates there, it gradually sets up, by its pressure, a contraction of the muscular walls of the gut somewhere about the junction between the ascending and transverse colon. This wave of contraction then .begins to travel slowly toward the cecum, without, however, being pre- ceded by any relaxation of the wall of the gut, as is the case with a true THE MECHANISMS OF DIGESTION 469 peristaltic wave. This first wave is soon followed by others, with the result that the food is forced up into the cecum, against the blind end of which it is crowded, being meanwhile prevented from passing into the ileum by the operation of the ileocolic sphincter and by the oblique manner in which the ileum opens into the cecum. As the result of the distention of the cecum set up by these so-called antiperistaltic waves, a true coordinated peristaltic wave is occasionally initiated, and passes along the ascending colon preceded by the usual wave of inhibition. These waves, however, disappear before they reach the end of the colon, so that the food is again driven back by the so- Fig. 165. — Diagram of time it takes for a capsule containing bismuth to reach the various parts of the large intestine. called antiperistaltic waves. The effect of the movements is to knead and mix the intestinal contents, and thus encourage the absorption of water from them. The resulting more solid portions then collect toward the splenic flexure, and become separated from the remaining more fluid portion by transverse waves of constriction, which develop into peri- staltic waves carrying the harder masses into the distal portions of the colon, where they collect chiefly in the sigmoid flexure. The descending colon itself is never distended with contents and merely serves as a tube for transferring the masses from the transverse colon to the sigmoid flexure. The time taken for a capsule of bismuth to reach the various parts of the large intestine is shown in Pig. 165. After a certain mass has collected in the sigmoid flexure and rectum, the increasing distention causes a reflex evacuation of this portion of the 470 DIGESTION • gut through centers located in the spinal cord. The impulses from these centers, besides contracting the rectum, etc., also coordinate the contrac- tion of the abdominal muscles and the relaxation of the sphincter ani so as to bring about the act of defecation. By the skiagraphic method it has been found that the pelvic colon gradually becomes filled with feces from below upward, and that the rectum remains empty until just before defecation. EFFECT OF CLINICAL CONDITIONS ON THE MOVEMENTS Observations of practical value have been made on the behavior of the peristaltic wave after various intestinal operations. After an end-to-end anastomosis of the gut, no evidence can be obtained by the x-ray method that any hesitation occurs in the movement of the shadows at the anas- tomosis. On the other hand, when a lateral anastomosis is established, stagnation of the food in the region of the junction may occur, this having been found, on opening the gut, to be caused by the accumu- lation of hair and undigested detritus at the opening between the op- posed loops. Another objection to lateral anastomosis is the fact that in performing the operation a considerable amount of the circular muscle is cut, which interferes with peristaltic activity. Moreover, the end of the proximal loop beyond the opening is in danger of becoming filled up with hardened material, and the end of the distal loop may become invaginated and induce obstruction in the region of the anastomosis. Observations have also been made by the x-ray method on the be- havior of the intestinal contents following intestinal obstruction. It has been observed that, as the material collects in the gut just above the obstruction, strong peristaltic waves are set up, which move the food toward the obstruction so powerfully as to cause the walls of the canal in front to become bulged, until at last the pressure causes the con- tents to be squirted back through the advancing ring of peristaltic con- traction. These waves were observed to succeed one another rapidly. When a portion of gut is reversed in position, the peristaltic waves con- tinue to travel in their old direction toward the duodenum. The effect of this is to produce a partial obstruction at the upper end of the re- ceptive gut. The type of peristalsis known as the peristaltic rush can be induced experimentally in animals by intravenous injection of ergot. It prob- ably also occurs in conditions of abnormal irritation of the gut in man, and is believed to be the characteristic activity of the gut after a • strong purge. CHAPTER LIV HUNGER AND APPETITE Hunger and appetite are distinct sensations, the former being definitely correlated with contraction of the empty stomach, and the latter being a complex of sensations operating in the nervous system along with memory impressions of the sight, taste, and smell of palatable food. Appetite is therefore a highly complex nervous integration, whereas hunger is a much simpler process. It is particularly with hunger that we shall concern ourselves at present. When a thin-walled rubber balloon of proper size is placed in the stomach and connected by a rubber tube with a water, bromoform or chloroform manometer (made of wide glass tubing 1.5 cm. in diameter and provided with a suitable float on the free limb) a tracing may be taken of the movements of the stomach. For use on man the capacity of the balloon should be from 75 to 150 cubic centimeters. The record thus obtained when the balloon is placed in the empty stomach of a normal person shows four types of wave. Two of these may be discounted, being due to the arterial pulse and the respiratory movements. The third is known as the tonus rhythm, and is caused by tonic contractions of the fundus of the stomach of varying amplitudes and occurring at a rate which varies from 18 to 22 per second. The periods of tonus in- crease during the powerful rhythmic contraction to be immediately described. While these changes in tone are occurring, no subjective sen- sation of hunger is experienced. (See Fig. 167.) The fourth and most significant type consists of powerful rhythmic contractions, alternating with periods of quiescence. These contrac- tions occupy a period of about twenty seconds, and are superimposed upon the tonus rhythm. They gradually increase in amplitude and fre- quency.; and, in the case of young and vigorous persons, may gradually pass into a condition of incomplete tetanus, after which they suddenly subside, leaving only a faint tonus rhythm. These rhythmic contrac- tions are definitely associated with the sensation of hunger, and are more marked the more intense the sensation is. When tetanus occurs the hunger sensation is continuous, but it instantly disappears when the tetanus gives place to relaxation. When the contractions are com- paratively feeble, the length of the period during which they occur is 471 472 DIGESTION about twelve minutes, being shorter than when the contractions are power- ful, in which case the periods are always initiated by weaker contrac- tions with long intervening pauses. Finally, the pauses disappear and the contractions become more and more pronounced until, as above men- tioned, a virtual tetanus, lasting from two to five minutes, may super- vene. The duration of the hunger period varies from one-half to one and a half hours, with an average of from thirty to forty-five minutes, and the number of individual contractions in a period varies from twenty to seventy. Between the hunger periods, intervals of from one-half to tAvo and one-half hours of quiescence may supervene. (See Fig. 168.) Similar contractions, often passing into incomplete tetanus, have been observed in the stomach of healthy infants, some of the observations hav- ing been made before the first nursing. The intervals of motor quies- Fig. 166. — Diagram of method for recording stomach movements. B, rubber balloon in stomach. D, kymograph. F, cork float with recording flag. M, manometer. L, manometer fluid (bromo- form, chloroform, or water). R, rubber tube connecting balloon with manometer. S, stomach. T, side tube for inflation of stomach balloon. (From Carlson.) cence between the hunger periods are shorter than in adults. In obser- vations made during sleep, it was observed that, when the contractions were very vigorous, the infant would show signs of restlessness and might awake and cry. As in the adult, the contractions are evidently associated with subjective sensations of hunger. Contractions , of the empty stomach have also been recorded on a large variety of animals, including the dog, rabbit, cat, guinea pig, bird, frog and turtle. They vary somewhat in type in different animals. . With regard to the time of onset of the tonus and hunger contractions, it has been observed that the only period during which the fundus is free of them is immediately after a large meal. After a moderate meal the tonus rhythm begins to appear in about thirty minutes. It gradually HUNGER AND APPETITE 473 increases in intensity, until by the time the stomach has nearly emptied itself the tonus has become conspicuous, and the stronger hunger con- tractions usually begin to appear. Superimposed upon those of the tonus rhythm, hunger pangs may appear in man when the stomach still contains traces of food. Fig. 167. — Tracing of the tonus rhythm of the stomach (man) three hours after a meal. (From Carlson.) By studying the shadow of the outline of the stomach produced by having a person or animal swallow two balloons, one inside the other and with a paste of bismuth subnitrate between them, it has been ob- served that the weaker type of hunger- contraction begins as a con- Fig. 168. — Tracings from the stomach during the culmination of a period of vigorous gastric hunger contractions. One-half original size. (From Carlson.) striction involving the cardiac end of the stomach, and moving toward the pyloric end as a rapid peristaltic wave. When the contractions are very vigorous, this wave spreads so rapidly over the stomach that it is difficult to determine whether it really occurs as a very rapid peristalsis or as a contraction involving the fundus as a whole. These contractions 474 DIGESTION resemble very closely the movements that have sometimes been observed after a bismuth meal, and which have been thought by clinical observers to indicate a hyperperistalsis of the stomach. The fundus is therefore not entirely passive during digestion; for, although early in this act there may be no evidence of contraction, yet the contractions of the tonus rhythm may appear and become pronounced before the stomach is en- tirely empty. In other words, the digestion contractions of the filled stomach (see page 451) pass gradually over into the hunger contractions of the empty organ. It appears that the stomach contractions produce the hunger sensa- tions by causing stimulation of afferent nerve endings in the muscle layers of the viscus. Mere pressure on the mucosa itself does not originate such a sensation; thus, sudden distention of the balloon or rubbing the mucosa with the closed end of a test tube, inserted through a gastric fistula, did not cause any sensation of hunger, unless the stimulus was so strong as to cause a contraction of the musculature of the stomach. It has been thought by some observers that, during hunger, contrac- tions similar to those of the stomach also occur in the lower end of the esophagus. It is believed by Carlson, however, that these contractions are not at all responsible for the hunger sensation, although they may give rise to a feeling that something has stuck in the esophagus. Con- tractions of the intestine have also been observed in hunger, but it is doubt- ful whether they have anything to do with the cause of the hunger sensation. REMOTE EFFECTS OF HUNGER CONTRACTIONS It is well known that during hunger certain general subjective symp- toms are likely to be experienced, such as a feeling of weakness and a sense of emptiness, with a tendency to headache and sometimes even nausea in persons who are prone to headache as a result of toxemic conditions. Headache is likely to be more pronounced or perhaps only present in the morning before there is any food in the stomach. These symptoms indicate that hunger contractions are associated with hyper- excitability of the central nervous system, and it is of considerable interest that objective signs of this association can be elicited. If the knee-jerk be recorded along with a record of the gastric contractions, it will be found that it is markedly exaggerated simultaneously with the strong hunger "contractions of the empty stomach, this augmentation being greatest at the height of the stomach contractions, when the hun- ger pangs are most intense, and falling off again to normal when these disappear (Fig. 169).. Further changes occurring during the hunger HUNGER AND APPETITE 475 period include an increase in the pulse rate and vasodilatation. By comparing plethysmographic tracings of the arm volume (see page 230) and stomach contractions, it has been found that the increase in volume occurs pari passu with the increasing tonus of the stomach, but that it begins to shrink before the stomach contraction has reached its maximum. Occasionally, however, as in acute hunger, a somewhat different rela- tionship obtains, vasoconstriction being more prominent. During each hunger contraction there is also increased salivation, the degree of which varies with different individuals. This salivation is independent of the more copious "watering of the mouth" that accompanies the thought or sight of appetizing food. j L tttttttttttttttttWtttttttttlttr' -ttttttttttttttttttmttttttttttttttfitmtttitttnitittni Fig. 169. — Showing augmentation of the knee-jerk (upper tracing) during the marked hunger con- tractions (lower tracing). (From Carlson.) HUNGER DURING STARVATION During enforced starvation for long periods of time, it is known that healthy individuals at first experience intense sensations of hunger and appetite, which last however only for a few days, then become less pronounced and finally almost disappear. It is of interest to know the relationship between these sensations and the hunger contractions in the stomach. This has been investigated by Carlson and Luckhardt, who voluntarily subjected themselves to complete starvation, except for the taking of water, for four days. During a great part of this time records of the stomach contractions were taken by the balloon method, and it was found that the tonus of the stomach and also the frequency and intensity of the hunger contractions became progressively more pronounced as starvation proceeded. Towards the end of the period it was also noted that incomplete hunger tetanus made its appearance where ordinarily, as in Carlson's case, this type of hunger contraction was infrequent. Sensations of hunger were present more or less throughout the period, being therefore probably due to the persistently increased tonus. The onset of a period of hunger contraction could usually be foretold by an 476 DIGESTION increase in the hunger sensation, and as these contractions became more marked, the hunger sensations became more intense. On the last day of starvation a burning sensation referred to the epigastrium was added to that of hunger. The appetite ran practically parallel with the sensa- tion of hunger, and both of these sensations became perceptibly dimin- ished on the fourth or last day of starvation, this diminution being, however, most marked in the sensation of appetite. Indeed, instead of an eagerness for food, there developed on the last day a distinct repugnance or indifference towards it. Accompanying these sensations of hunger and appetite a distinct mental depression and a feeling of weakness were experienced during the latter part of the starvation period. On partaking of food again the hunger and appetite sensations very rapidly disappeared, and also practically all of the mental depression and a great part of the feeling of weakness. Complete recovery from the latter, however, did not take place until the second or third day after breaking the fast. From this time on both men felt unusually well; indeed they state that their sense of well-being and clearness of mind and their sense of good health and vigor were as greatly improved as they would have been by a month's vacation in the mountains. They further point out that, since others who have starved for longer periods of time unanimously attest the fact that, after the first few days, the sensations of hunger become less pronounced and finally almost dis- appear, they must have experienced the most distressing period during their four days of starvation. Although the hunger sensation was strong enough to cause some discomfort, it could by no means be called marked pain or suffering, and was at no time of sufficient intensity to interfere seriously with work. Mere starvation can not therefore be designated as acute suffering. It is of further interest to note that dur- ing the starvation period a continuous flow of secretion of acid gastric juice was found to occur in the stomach, the presence of this acid prob- ably explaining the acid or burning sensation experienced in the epigas- trium on the last days. CONTROL OF THE HUNGER MECHANISM The control of the hunger mechanism, like that of any other mechan- ism in the animal body, may be effected through the nervous system or it may depend on the presence of chemical substances or hormones in the blood. As a matter of fact, it can readily be shown that both those methods of control are operative, and we will now consider briefly some of the facts upon which this conclusion depends. Although many facts are now known with regard to the nervous con- HUNGER AND APPETITE 477 trol of the hunger mechanism, it is difficult to piece these together in such a way as to formulate a simple theory which fits in with all the observed facts. We know that the stomach possesses in itself a local nervous mechanism by which, like the heart or intestine, it can auto- matically perform many of the movements which are exhibited in the intact animal. These local movements may, however, be considerably influenced by impulses transmitted to the stomach along the vagus and splanchnic nerves. We have therefore to seek for evidence indicating the relative importance of the local nervous mechanism in the stomach itself and of the impulses transmitted to this organ by the extrinsic nerves. We must then seek the position of the center which perceives the sensation of hunger. It will be simplest to consider first the effect of section of the extrinsic nerves in observations made on lower animals. Section of the splanchnic nerves increases gastric tonus and augments the gastric hunger contrac- tions. Section of both vagus nerves, performed of course below the level of the heart, leaves the stomach in a more or less hypotonic condition. The tonus is not entirely abolished; it varies somewhat from day to day, and may become quite pronounced even though the vagi are cut. In this hypotonic state the hunger contractions are diminished in rate and regularity. Section of both the splanchnic and vagus nerves throws the stomach into a permanent hypotonus, except in prolonged starva- tion, when hunger contractions develop that are usually of great ampli- tude and with particularly long intervals between the contractions. The general conclusion to be drawn from these experiments is that, although completely isolated from the central nervous system, the stomach still exhibits typical hunger contractions, which must therefore be essentially dependent upon an automatic mechanism in the stomach wall itself. Over this mechanism, extrinsic nerve impulses have merely a regulatory control. Variations and Inhibitions of the Hunger Contractions The afferent stimuli that may set up impulses traveling by the extrin- sic nerves to the stomach are conveyed by the nerves of sense or are of psychic origin. Stimulation of the gustatory end organs in the mouth, as by chewing palatable food, always causes an inhibition of the tonus and a diminution or disappearance of the hunger contractions. Even the chewing of indifferent substances, such as paraffin, suffices to produce distinct inhibition, unless in a case in which the contraction has passed into a tetanus. It is of interest that swallowing movements, in the ab- sence of any food substance in the mouth, are sufficient to produce a transitory inhibition of the gastric tonus — a receptive relaxation of the 478 DIGESTION stomach, as it has been aptly called. The diminution in tonus and hunger contractions in these various ways is accompanied by a diminu- tion in the hunger pains. Afferent nerve stimulation affecting the hunger contractions may also originate in the stomach mucosa itself, as has been shown by Carlson on his patient by introducing the various substances to be tested through a tube into the stomach. A glassful of cold water introduced in this way inhibits the tonus and the hunger contractions for from three to five minutes unless these are severe; this inhibition being followed by no augmentation either of the tonus or of contractions. Ice-cold water has a greater effect than water at body temperature. This result is some- what different from that which most men experience as the result of drinking a glass of cold water. Weak acids of strengths varying up to that found present in the gastric juice itself — 0.5 per cent — cause a marked inhibition of the hunger movements, but this inhibition does not persist until all the acid has escaped from the stomach or been neutralized, which explains why hunger contractions should still occur when an acid secretion is present in the stomach, as in starvation. Normal gastric juice itself produces an inhibition, which is no doubt dependent upon the acid which it con- tains, and it is probable that, at the same time that it leads to inhibition of the hunger contractions, the acid initiates peristalsis of the pyloric region (see page 453). Weak alkaline solutions have no greater effect on the hunger contractions than an equal volume of water. Weak solu- tions of local anesthetics, such as phenol or chloretone, are without effect. With regard to alcoholic beverages interesting results were obtained. Wine, beer, brandy, and diluted pure alcohol inhibit both the tonus and the contractions. The duration of this inhibition varies directly with the quantity of the beverage introduced into the stomach and with its alco- hol percentage. These observations are apparently not in harmony with the experience of most men that the taking of alcoholic beverages serves to awaken or increase the appetite, the difference being no doubt due to the fact that appetite and hunger contractions of the stomach are not dependent on each other, appetite being, as we have seen, a complex psychic affair, whereas the hunger contractions depend upon a local mechanism in the stomach wall itself. As the inhibition produced in one or other of these ways passes off, the hunger contractions are resumed at their previous intensity and not in an augmented form. From the promptness of the inhibition, it would appear that the stomach contractions are affected, not reflexly through the central nervous system or by changes in the chemical composition of the blood, but by a direct action on the neuromuscular mechanism HUNGER AND APPETITE 479 in the stomach walls, and it is important to bear, in mind that the inhibitory effects on the stomach contractions of the fundus may proceed quite independently of the changes in the pyloric region that are con- cerned with th.e mechanical processes of digestion. After one or both of the extrinsic nerves of the stomach were severed in dogs, a certain degree of inhibition could still be induced by the above methods, indicat- ing that, although section of the extrinsic nerves depresses the inhibitory reflex, it does not abolish it. Various mitigations of the hunger contractions have been discovered. Smoking has this effect, and compression of the abdomen by tightening the belt also inhibits the contractions provided they are not of marked intensity. Considerable muscular exercise, such as brisk walking or running, causes inhibition, which usually persists until after the exer- cise is discontinued. When the tonus and contractions return, in this case, they seem to be somewhat more pronounced. Application of cold to the surface of the body — as by placing an ice pack on the abdomen or taking a cold douche, procedures which are well-known to induce increased neuromuscular tonus, in general — causes an inhibition of the gastric tonus and hunger contractions, the degree of which is roughly proportional to the intensity of the stimulation. There is certainty never an increase in the gastric tonus or hunger contractions. If such stimula- tion is maintained, the inhibitory effects on the stomach gradually dimmish, even though the individual be shivering intensely. With regard to the nerve centers concerned in these phenomena, little that is definite is known. The sensory nuclei of the vagus nerve in the medulla must be considered as the primary hunger center, and through this center, not only influences affecting the stomach contractions, but also those associated with the hunger sensations, must be mediated. It would appear from observations on the hunger behavior of decerebrate animals that there can be no hunger center located on the cerebral cortex itself, for such animals exhibit practically the same hunger effects as normal animals. It is interesting to note that, at least in the case of decerebrate pigeons, this hunger behavior entirely disappears on removal of the optic thalami, where important nerve centers having to do with the bodily responses of the animal to hunger impulses would therefore appear to be located. These observations support the suggestion that has been made by several neurologists that the sense of pain is located somewhere in the thalamic region. Concerning the influence of psychic states, Carlson says that in his own case the hunger contractions became weaker and the intervals between them greater when he was suddenly awakened during his fast and saw two of his friends partaking at his bedside of a " feast of 480 DIGESTION porterhouse steak with onions, potatoes, and a tomato salad." These results are no doubt due to local inhibition dependent upon the psychic secretion of appetite gastric juice. When no such juice is produced, the sight and smell of good food does not appear to affect materially the hunger contractions of the stomach. No doubt it stimulates the appetite, but that, as we have seen, is a psychic affair. CHAPTER LV THE BIOCHEMICAL PROCESSES OF DIGESTION In a book designed primarily for clinical workers, it would be out of place to enter into details concerning the biochemical processes taking place during the digestive process. There is, however, a certain amount of fundamental knowledge which it is essential that we should consider. In the first place it should be borne in mind that in the digestion of carbohydrates and proteins, various intermediate stages are passed through before the final absorption products are formed. The highly complex molecule of which protein, for example, is composed, is first of all broken down into several smaller but still highly complex mole- cules, each of which then undergoes further disruption, until ultimately the amino acids are set free. Certain enzymes, such as trypsin, can carry this process from the beginning through the greater part of its course without the assistance of other enzymes, but in the natural proc- ess of digestion, as it occurs in the gastrointestinal tract, the different stages of the disruption are controlled by different enzymes. One enzyme prepares the food for action by the next. This interdependence of the actions of the enzymes demands that some provision should be made whereby each enzyme is secreted at the proper time; that is, when the foodstuff has already been prepared for its action by that of its prede- cessor. Thus, it would be useless after food is taken for the gastric and pancreatic juices to be secreted at the same time. Instead, the gastric juice is secreted first, and the pancreatic only after the food has been prepared for its action. This correlation in function we have already seen to be dependent largely on the action of hormones. DIGESTION IN THE STOMACH The gastric juice contains two important digestive agencies: (1) the enzyme, pepsin, and (2) hydrochloric acid. It is particularly in juices secreted in the cardiac end of the stomach that these two substances are found present; towards the pyloric end the hydrochloric acid entirely disappears, and the pepsin content becomes distinctly less. 481 482 DIGESTION The Functions of Hydrochloric Acid The functions of hydrochloric acid may be conveniently divided into physiological and biochemical. The former functions have to do with the control of the movements of the stomach, including the opening of the pyloric sphincter, and, after the chyme has entered the duodenum, with the secretion of pancreatic juice and bile. The biochemical functions are concerned: (1) in assisting the pepsin in the digestion of proteins, (2) in bringing about a certain amount of inversion of disaccharides, and (3) in having an antiseptic action on the stomach contents. Re- garding the last mentioned of these functions, it may be said that the chyme, as it is ejected from the stomach, is usually sterile, although it may contain spores and certain bacteria that are protected against the digestive agencies of the stomach. This protection is afforded by an outer covering of a chitinous nature (spores), or, as in the case of the tubercle bacillus, by a covering of waxlike material. It is believed that persons with strictly normal digestion are much less liable to infection by such bacteria, as those of typhoid and cholera, than persons with less active gastric secretion. When the acid of the gastric juice falls below the level at which it develops an antiseptic action, various bacteria and yeasts grow in the stomach contents, producing by the resulting fermen- tation irritating organic acids and gases. It is under these conditions that yeasts, sarcinae, and lactic and butyric acid bacilli find in the gastric contents a suitable nidus on which to grow. THE AMOUNT OF ACID It has long been known that considerable variations in the amount of hydrochloric acid in the gastric juice are associated with symptoms of indigestion. On this account a more or less elaborate technic has been developed for the purpose of determining the amount of hydrochloric acid in the gastric contents.* There are three things in connection with this activity that we may measure: (1) the total titrable hydrochloric acid; (2) the free hydrochloric acid; and (3) the actual hydrogen-ion concentration. The determination of the total available acids is made by titrating a measured quantity of gastric juice against a standard alkali, using phenolphthalein as an indicator. By this method about 75 c.c. of decinormal alkali solution are required to neutralize 100 c.c. of normal gastric juice. The determination of the free hydrochloric acid is made by using special indicators, such as those of Giinzberg and Topfer, which change color at a hydrogen-ion concentration of about 10'5 (see page 27). To produce this hydrogen-ion concentration, a con- *The methods can be found in any volume on clinical diagnosis. THE BIOCHEMICAL PROCESSES OF DIGESTION 483 siderable quantity — 0.05 per cent or more — of an organic acid is neces- sary, whereas it requires only a trace of hydrochloric acid. Normal human gastric juice, when titrated with one of these indicators, gives a figure which corresponds to about 0.03 N. hydrochloric acid (see page 22). For the accurate determination of the hydrogen-ion concentration, it is necessary to use the gas-chain method (see page 29). When gastric juice is collected through a fistula from an empty stomach, very little difference will be found between the free hydro- chloric acid and the total acid; that is, between th'e results obtained by the second and the first of the methods described above. This is because in such juice there is no organic matter capable of, combining with the hydrochloric acid, and there are no other acids, such as lactic or butyric, which might be produced by fermentative processes. The difference between the two titrations, however, becomes quite marked when pro- tein food is undergoing digestion in the stomach, because at its different stages of digestion protein combines with increasing quantities of the hydrochloric acid. The pathologic condition in which there is most definitely a diminution of the hydrochloric acid is cancer, either of the stomach itself or occasionally of some other part of the body. An in- crease is particularly marked in ulcer of the stomach. A considerable variation in hydrochloric acid may however be the result merely of func- tional (neurotic) conditions. THE SOURCE OF THE ACID A question that has puzzled physiologists for many years concerns the mechanism by which hydrochloric acid is secreted. The percentage of hydrochloric acid in the gastric juice is considerably above that at which any animal cells can live, and yet this acid is secreted by the lining membrane of the stomach, its source being, of course, the ' sodium chloride of the blood plasma. How then do the cells of the gastric glands bring about the separation of this powerful acid from the per- fectly neutral blood plasma ? In the first place, it is significant that the mucous membrane of the stomach contains a higher percentage of chlorine than the average of other organs and tissues, indicating that it has the power of abstracting chlorine from the blood. The excess of chlorine in the mucosa must, moreover, be but a very small fraction of that actually secreted into the the gastric juice. The chlorine content of the mucosa of the cardiac end is considerably greater than that of the pyloric. These facts indicate that chlorine is attracted by the gastric cells, but they throw no light on the question as to where the hydro- chloric acid is really formed. Is it in the cells, or only in the lumen of the gland tubes? That is to say, is it formed before or after the gastric 484 DIGESTION juice has been secreted from the cells? After intravenous injection of solutions of potassium ferrocyanide and some inert salt of iron, such as one of the scale preparations, examination of the gastric glands has shown that the prussian blue reaction, which requires the presence of free mineral acid, is most pronounced in certain of the parietal cells. A considerable amount of the precipitate is, however, also visible in the lumen of the glands and in the stomach itself. Certain observers affirm that, although some of the parietal cells may take -the stain, the vast majority of them do not do so; and, moreover, that cells incapable of forming hydrochloric. acid (e.g., of the liver) may also become stained, and that the precipitation may occur in the blood and lymph. The confusion in the results by these methods prompted A. B. Macal- lum1* and Miss M. R Fitzgerald to investigate the distribution of the chlorine in the cells by a rnicrochemical method, in which the chlorides were precipitated with silver nitrate and the silver chloride then reduced by exposing the section to light. It was found that both kinds of gas- tric-gland cell, chief and parietal, but particularly the parietal, gave the chloride reaction. Using as a stain a substance (cyaninine) which reacts blue with acid and red with alkali, Harvey and Bensley,15 however, aver that the secretion of the glands is practically neutral until the foveola is reached, where the stain becomes blue, indicating an acid reaction. This seems to show that the acid is not really secreted by the cells of the gastric gland, but is formed after secretion. According to the latter investigators, the chlorine is secreted by the cells into the fovea as some weak chloride, such as ammonium chloride, or it may be as an ester. Shortly after its secretion this weak chloride undergoes a hydrolytic or other dissociation, during which free hydro- chloric acid is liberated and ammonia or some other weak base set free. Of these two products of the reaction the weak base is reabsorbed by the gland cells, but the hydrochloric acid is left behind because the cells are impervious to it. Indirect evidence in support of this view is afforded by certain other instances in which hydrochloric acid is pro- duced by the action of cells ; thus, the mold Penicillium glaucum when it is grown in a medium containing ammonium chloride absorbs the am- monia but leaves the hydrochloric acid. The high penetrating power of the ammonia ion in practically all cells, and the fact that the mucosa of the stomach contains a higher percentage of ammonia than any other tissue in the body, must also be considered as circumstantial evidence in favor of this view. Whatever be the mechanism by which hydrochloric acid is produced, there is no doubt that the1 epithelium is impenetrable to it. When the vitality of the epithelium becomes lowered, as in anemia or after partial THE BIOCHEMICAL, PROCESSES OF DIGESTION 485 occlusion of the arteries, the acid may penetrate the cells and cause digestion of the stomach walls. Hyperacidity may on this account become dangerous, as it lowers the resistance of the cell. The digestive action of hydrochloric acid is closely linked with that of pepsin, with which it will, therefore, be considered. The Action of Pepsin It is commonly believed that before its secretion pepsin exists in the cells of the gastric glands as zymogen granules. The chief evidence for this belief appears to be that after considerable activity the amount of zymogen granules in the gland cells is found to be decidedly dimin- ished. By such an hypothesis it is easy to explain certain interesting results concerning the effect of weak alkali on the activities of extracts of the mucous membrane of the stomach. When the mucous membrane is extracted with weak acids, the extract is very active proteolytically. If this so-called pepsin solution be made faintly alkaline, or even only neutralized, and again made acid, it will be found to have lost much, if not all, of its activity. On the other hand, an aqueous extract may be rendered slightly alkaline for a short time and still display its digestive activity on subsequent acidification. The extract made with water is therefore much more resistant toward alkali than that made with weak acid, and the difference is explained on the supposition that the watery extract contains pepsinogen, whereas the acid extract contains pepsin. It is believed that there are several varieties of pepsin, because the optimum concentration of acid in which pepsin derived from the stomachs of different animals acts is not always the same. Pepsin of the dog, for example, acts best in a hydrogen-ion concentration corresponding to that of a 0.05 N. hydrochloric acid solution, whereas that of the human stomach works best at a concentration of 0.03 N. Different pepsin solutions also show a difference with regard to the optimum tempera- ture at which they act, and with regard to the nature of the protein which they most readily attack. Thus, the pepsin of a calf's stomach digests casein very rapidly, but coagulated egg white only slowly, whereas the pepsin of the pig's stomach acts on both these proteins at about the same rate. It is well known that the activity of pepsin can proceed only in the presence of acids, but this action of acids does not appear to depend on the hydrogen-ion concentration alone, for when equal quantities of the same pepsin are mixed with quantities of different acids so that the hydrogen-ion concentration of the mixtures is uniform, it is found that digestion proceeds most rapidly with hydrochloric acid and least rapidly with sulphuric acid. The S04 ion seems, therefore, to be unfavorable 486 DIGESTION for peptic activities. The acid seems to combine with the protein before the pepsin attacks the latter; for, if we first combine the protein with acid and then wash away all traces of free acid, the protein can be digested in a neutral pepsin solution without the liberation of any free acid. There is evidence to show that pepsin itself also becomes combined with the protein during the digestive process. If a piece of protein such as fibrin be immersed in a solution of pepsin, and then taken out and washed thoroughly to get rid of all adherent pepsin, it will be found, on placing it in a hydrochloric acid solution of the proper strength, that peptic digestion proceeds. Advantage may be taken of this fact to separate pepsin from a solution, but the best protein to use for this pur- pose is not fibrin but elastin. By such a method it has, for example, been shoAvn that there is some pepsin in the intestinal contents, proving thus that when the chyme passes into the intestine, the pepsin is not, as used to be thought, immediately killed by the proteolytic enzyme. With regard to the products of gastric digestion, little can be said here. The first product is a metaprotein known as acid albumin or syntonin. It is precipitated from the digestion mixture by neutraliza- tion. The next product is known as primary proteose, being precipi- tated by half saturation with ammonium sulphate. The third product is secondary proteose, produced by complete saturation with the above reagent ; and after all these bodies have been separated out, there re- mains in solution the fourth product — peptone — which among other things is characterized by the fact that with the biuret test it gives not a violet but a rose-pink color. It has often been claimed that along with these products a certain amount of free amino acids may also appear in a peptic digestive mix- ture. This, however, may be due to the action of erepsin, which is usually present in pepsin preparations. It is important to note that the term proteose is a general one, and that there are probably many varieties of this substance, differing from one another according to the protein from which they are derived. The change produced by pepsin and hydrochloric acid is of the nature of an hydrolysis, for it has been found that the amount of hydrogen and oxygen in the digestive products is greater than that in the original protein. It is by a similar process of hydrolysis that the other proteolytic enzymes, such as pancreatin and erepsin, operate, but this does not imply that the exact grouping that is split apart by the hydrolytic proc- THE BIOCHEMICAL PROCESSES OF DIGESTION 487 ess is the same for each of these enzymes. Indeed, there is considerable evidence that pepsin does not, like the other enzymes, break up the long chain of amino acids that are linked together to compose the polypep- tides, but that it only splits the big molecule of albumin or globulin into several large groups, each of which is composed of long amino-acid chains. Its action appears to be analogous with that of amylase on starch, by which, it will be remembered, the big polysaccharide mole- cule is split into smaller polysaccharide molecules, which then become attacked by the dextrinase and split into disaccharide molecules (see page 656). The evidence in support of this view is: (1) that pepsin is unable to digest polypeptides, and (2) that it is able to digest certain proteins upon which erepsin (see page 490) has no action. The hydrolytic splitting of large into smaller protein molecules, like that by which the chains of amino acids in the polypeptides are subse- quently broken up, consists in a breaking* of amino-carboxyl linkings (NHCO) (see page 598), with. the consequent liberation of a large num- ber of unattached amino groups. The number of these free amino groups can be determined quantitatively by the formaldehyde titration method of Sorensen.* By this method it can be shown that from the very start of peptic digestion the number of free amino groups increases, and pari passu the power of the digestive products to combine with free hydro- chloric acid. Indeed, when the experiments are done quantitatively and the digestion allowed to proceed for a considerable time, the increase in the formol titration is practically equal to the decrease in the free acids as determined by the Giinsberg reagent. The rate of peptic digestion is usually estimated by the law of Schiitz and Borissow, according to "which the amount of coagulated albumin that is digested in a Mett's tube is proportional to the square root of the amount of pepsin, f * The pepsin which leaves the stomach in the chyme is not all destroyed in the intestine, as was at one time believed to be the case, for, as we have seen above, some pepsin can be detected in the gastrointestinal con- tents. A part of the pepsin may be absorbed into the blood and carried back to the gastric glands to be used again. This would account for the presence of antipepsin in the blood, and also for the presence of pepsin in the urine. It is probable, however, that most of the pepsin is de- stroyed after it enters the intestine. *In this method the basic character of the amino acids is destroyed by the formaldehyde,, so that a higher degree of acidity develops in the mixture. By determining the increased acidity by titration with alkali, an estimate is oibtained of the number of amino groups. (See page 599.) tThe amount of coagulated egg albumin digested is ascertained by measuring the length digested away from the end of a column of coagulated egg white contained in a glass tube (Mett's method). (See Cobb, P. W. : Am. Jour. Physiol., l'905, xiii, 448.) 488 DIGESTION Clotting of Milk in the Stomach Besides its power of digesting protein, the gastric juice is also endowed with the property of clotting milk. This action is commonly attributed to the presence of another enzyme besides pepsin, namely, rennin; but in recent years considerable controversy has raged around the question as to whether pepsin and rennin are not the same thing. One strong argument in favor of this view is that all digestive juices that are capable of digesting protein can also clot milk. In any case, when gastric juice acts on milk, it splits the casein* of the milk into two portions, one of which, called paracasein, immediately combines with calcium to form an insoluble colloidal compound, which is precipitated and, by entangling the fat of the milk, forms the clot; the other protein remains in solution and is known as whey albumose. From studies on molecular weight it is believed that the paracasein is produced from casein by the splitting of the molecule of the latter into two, from which it would appear that the action of this enzyme is nothing more than the first stage in the hydrolysis of the casein molecule. ' The whey albumose, according to this view, is a by-product. There are many investigators, however, who believe that rennin and pepsin are not identical, since an infusion of the stomach of a calf has a powerful clotting action on milk but a very weak digestive one on egg white, whereas a similar infusion from the stomach of a pig shows exactly the reverse properties. This question is one of so controversial a na- ture that it would be out of place to go into it further here. It should be pointed out, however, that, when the gastric contents are acid in reaction, milk will become clotted by the action of the acid itself quite independently of any pepsin or rennin the juice may contain. This acid clotting of milk is probably of a different chemical nature from that produced by the enzymes. On other foodstuffs than proteins the action of the gastric juice is relatively unimportant, although polysaccharides may be considerably broken down in the cardiac end of the stomach on account of the action of swallowed saliva (see page 454), and disaccharides, as we have seen, may become split by the hydrolyzing effect of the hydrogen ion. Fat digestion also takes place in the stomach when the fat is taken in an emulsified condition, as in milk and egg yolk, but not when in masses, as in meat or butter. This action is due to the presence of a fat-splitting enzyme, or lipase, in the gastric juice. ,*In the above nomenclature casein is the same as caseinogen, and paracasein the same as casein, of the English physiologists. CHAPTER LVI THE BIOCHEMICAL PROCESSES OF DIGESTION (Cont'd) DIGESTION IN THE INTESTINES The further changes which the half-digested foodstuffs in the chyme undergo in the intestinal canal depend on the enzymes present in the secretion of the various glands and on the presence of bacteria. The most important of the digestive juices are the pancreatic juice and bile. The latter, however, does not contain any enzyme, its influence on diges- tion being entirely adjuvant. Pancreatic Digestion When we were considering the mechanism of secretion of the pan- creatic juice, we saw that the juice produced by the action of secretin on the gland cells does not contain any active proteolytic enzyme, although it contains one capable of acting on polysaccharides and another, on fat. THE ACTION OF TRTPSIN When this juice is mixed with the secretion of the duodenum or of the upper part of the small intestine, it immediately develops powerful proteolytic power. The same result may also be obtained by mixing it with an extract of the mucous membrane of the duodenum made with dilute bicarbonate solution. A very small amount of the extract is capable of increasing the digestive activity of a very considerable quan- tity of pancreatic juice, showing that the action depends on the presence of an enzyme which has been called enterokinase. This influence of the intestinal secretion is readily destroyed by heating. Large quantities of alkali are contained in the pancreatic juice and bile, so that in the upper reaches of the intestine the acidity of the chyme is practically neutralized. A little lower down, however, an acid reaction may again develop (see page 505). On account of these facts it has been concluded that the activity of trypsin is most rapid in the pres- ence of a slight excess of hydroxyl ions; i. e., in a weakly alkaline solu- tion. It is interesting to note that, as a result of the great secretion of alkali by the pancreas, extracts of this organ after death show a very high degree of acidity in comparison with extracts from other organs 489 490 DIGESTION and tissues. It has also recently been shown that the activity of trypsin does not depend on the presence of free hydroxyl ions, but that it may proceed in the presence of free acid, even up to a strength of CH = 1.5. If pepsin is present together with trypsin in a distinctly acid solution, the pepsin seems to destroy the trypsin, unless the mixture contains a considerable quantity of protein, when ,the tryptic activity may persist even for several hours. A practical conclusion that we may draw from these results is to the effect that preparations of trypsin — the so-called pancreatin, for example — if given with the food, may pass in an active condition into the duodenum, where, in the more favorable environment created by the neutralization of the excess of acid, it will develop its proteolytic power. The therapeutic administration of pancreatin is, therefore, justified (Long16). The activated trypsin acts on proteins in very much the same way as pepsin, except that the decomposition of the peptone and proteoses into polypeptides is the chief feature of the process. Thus, after tryptic digestion has proceeded for some time, only a trace of primary proteoses but considerable quantities of leucine, tyrosine and other amino acids will be found present. Some investigators believe that the thorough nature of the digestive action of activated pancreatic juice may depend on its also containing erepsin, an enzyme which we shall see to be pres- ent in considerable amount in the mucous membrane of the intestine and other tissues, and whose particular function is to split polypeptides into the amino acids. From the autolytic digestion which takes place in organs kept in a sterile condition after death, tryptic digestion differs in that it produces only small quantities of ammonia. The large quanti- ties of ammonia produced in autolytic digestion no doubt have a rela- tionship to the acids simultaneously set free during this process. In the products of tryptic digestion it is usually found that, although there has been considerable splitting of the protein into amino acids, there are still a good many amino-carboxyl (NHCO) linkages left un- broken, indicating that certain polypeptides are left intact in the mix- ture. . To split the polypeptides requires the aid of the erepsin, which is present in the mucous membrane of the intestine. Interesting inves- tigations have been made on the exact degree to which trypsin-entero- kinase can split up the various known polypeptides. This seems to depend on the structure of the polypeptide molecule and on the number of amino acids present in the chain. For example, analylglycine, but not glycylalanine is hydrolyzed, although both contain the same amino acids but linked together in a different way; and tetraglycylglycine, which contains five glycine radicles, is hydrolyzed, whereas diglycylgly- cine, which contains only three, is not. THE BIOCHEMICAL PROCESSES OF DIGESTION 491 The importance of the presence of erepsin in the mucous membrane of the intestine is that it serves as a barrier to the passage of any unsplit amino acids from the intestinal contents into the blood: It insures the breaking up of the protein molecule into its ultimate units before absorp- tion. The further fate of the absorbed amino acids will be considered under the subject of protein metabolism. THE ACTION OF LIPASE Neutral fat is decomposed into fatty acids and glycerine by the lipase present in the pancreatic juice. This enzyme may also be extracted from the glands by means of 60 per cent alcohol. Its action is remarkably accelerated by the presence of bile, and considerably depressed by inor- ganic salts. It is also very dependent on the degree of alkalinity, the optimum being a hydrogen-ion concentration of H x 10~8. The favoring action of bile is undoubtedly owing to the bile salts (see page 493), and it is probable that this action is dependent upon the influence which these have in lowering surface tension and therefore bringing about a more intimate contact between fat and water. THE ACTION OF AMYLOPSIN The action of pancreatic juice on carbohydrates depends on the amylolytic enzyme called amylopsin. In animals having no active ptyalin in the saliva, amylopsin serves as the only diastatic enzyme concerned in the digestive process. In any case, at least for the first stages of the disruption of the starch molecule — that is, its conversion into dextrines — amylopsin is a more powerful enzyme than ptyalin. It does not appear to be so efficient as ptyalin in the final stages of the hydrolysis, for it does not produce so much reducing sugar as ptyalin does. Indeed ex- tracts of pancreas will sometimes convert starch into soluble starch and dextrine with great speed, but produce scarcely any reducing sugar. On this account it is believed by many investigators that there are at least two distinct and separate enzymes in amylopsin and also perhaps in ptyalin, one a true amylase, which converts starch into dextrine, and the other a dextrinase, which converts dextrine into maltose. In the case of both ptyalin and amylopsin digestion proceeds best in a very weak acid reaction. Amylopsin, as it is secreted in the pancreatic juice, is fully activated; bile, apart from the alkali which it contains, having no influence on its digestive power. Besides amylopsin the pancreatic juice also contains maltase, and in the case of young animals or of those that take milk with their food throughout their lives, lactase also. After the suckling animal has dis- 492 DIGESTION continued taking milk, the lactase disappears from the pancreatic juice. Attempts have been made to bring it back by feeding the adult upon milk, but without success. Occasionally the pancreatic juice also con- tains invertase. The Bile Associated with the pancreatic juice in all its functions is the bile. When this fluid is prevented from entering the intestine, the digestive process becomes very imperfect, the absorption of fat being particularly interfered with (see page 691). Bile is also an excretory product, and its composition therefore is much more complex than that of the other digestive fluids. This varies very much, however, according to the method of collection. Bile from the gall bladder after death contains much more solid material, particularly bile salts and mucin, than that collected from a fistula of the bile duct or gall bladder during life. These differences will be evident from the accompanying table. Bile from Gall bladder Fistula 100 parts contain — • Water 86 97 Solids 14 3 Organic salts (bile salts) 9 0.9-1-8 Mucin and bile pigment 3 0.5 Cholesterol ' 0.2 0.06-0.16 Lecithin and fat 0.5-1.0 0.02-0.09 Inorganic salts 0.8 0.7-0.8 In general it may be said that bile obtained from a fistula in man contains only about 3 per cent of total solids, of which from one-fourth to one-half are inorganic, whereas bile from the gall bladder contains 10 to 20 per cent of total solids, of which only about one-twentieth are inorganic. The chief cause for this difference appears to be that when the bile goes to the intestine, a considerable proportion of its bile salts is reabsorbed into the portal blood and reexcreted by the liver. Some of the difference may also be caused by the fact that absorption of water takes place from the gall bladder, and that mucin and possibly cholesterol are secreted by this organ. These striking differences be- tween fistula and gall-bladder bile are observed only when the com- mon bile duct is occluded. If the bladder fistula is made with the com- mon duct left open, some of the bile gains entry to the duodenum and therefore becomes reexcreted. It is well known that a fistula of the gall bladder in man after a time closes up and the bile again takes its usual course along the bile duct into the duodenum. THE BIOCHEMICAL PROCESSES OF DIGESTION 493 Interesting observations have been collected on the amount of the secre- tion from a fistula both in man and in the lower animals. In man it is commonly stated that about 500 c.c. of bile are secreted daily, the amount varying considerably during the different hours of the day. The secretion of bile is greatly reduced by hemorrhage. It is greater on a meat diet than on one of carbohydrates. It is reduced during starva- tion, but continues to be secreted up to the moment of death. FUNCTIONS OF BILE One of the main functions of the bile salts is that they greatly assist, not only in the digestion, but also in the absorption of fats. When bile is excluded from the intestine, the feces are loaded with fatty acids which have been split off partly by the now less effective lipase and partly by the action of bacteria. The fatty acid thus liberated in the absence of bile salts is not absorbed, because the bile salts serve as the carriers of fatty acids into the epithelial cells and lacteals. They com- bine with the fatty acids, probably by forming some chemical compounds, in which they carry them into the endothelial cells where the compounds become disrupted, the fatty acid combining with glycerine to again form neutral fat and the bile salts being carried to the liver and reexcreted. The influence of bile salts in assisting the action of lipase is probably due to a lowering of the surface tension, thus bringing water and fat into closer union. This accelerating influence has also been demonstrated when synthetic bile salts have been used, showing clearly that it is really these and not any other constituent of the bile that are responsible for its accelerating influence. Bile also functionates as a regulator of intestinal putrefaction. This it does apparently because of its slight laxative properties, by which the intestinal contents are expelled before the bacteria have grown to any great extent in them. Bile itself is a favorable culture medium for certain bacteria, so that it can have no antiseptic action. Its assistance in the action of trypsin and amylopsin depends very largely upon the alkali which it contains. As an excretory vehicle bile is important, because it possesses the power of dissolving cholesterol. Toxins and metallic poisons of various kinds are also excreted in it. Although not directly concerned with the digestive function, it will be convenient to say something here concerning the chemical nature and derivation of the various biliary constituents. 494 DIGESTION THE CHEMISTRY OF BILE The Bile Salts In most animals the bile salts consist of the sodium salts of glycocholic and taurocholic acids. Each of these acids is composed of a part called cholic acid which is more or less related to cholesterol, and of glycine (CH2NH2COOH ammo-acetic acid) or taurine (C2H7NS03), a derivative of cysteine, which is a-amino-/?-thiopropionic acid (CH2HS.CHNH2. COOH). The exact form of cholic acid varies in different animals, that of the pig, for example, being different from that of man. Bile salts are an exclusive product of liver metabolism ; i. e., they are not formed in any other part of the animal body. They give a very sensitive color reaction known as Pettenkof er 's, which however is not specific of bile acids, since it is also given by oleic acid and by many aromatic substances and alcohols. It must be remembered that the part of the bile salts that is characteristic of the liver is the cholic acid, the taurine and glycine being present in other tissues and organs. When cholic acid is given to animals mixed with the food, the amount of taurocholic acid excreted with the bile is increased, indicating that there must be a store of taurine available in the organism. This store can not, however, be large, for if the feeding with cholic acid is repeated several times, it will be found that the taurocholic acid diminishes and glycocholic acid takes its place; and this increased excretion of glyco- cholic acid goes on just as long as cholic acid is fed. The reserve of taurine in the animal body appears therefore to be limited, although it is used in preference to glycine when there is an excess of cholic acid to be neutralized. On the other hand, the store of glycine seems to be inexhaust- ible. That there is no reserve of cholic acid itself in the body is indicated by the fact that no increase in taurocholic acid excretion by the bile results when cystine, the mother substance of taurine, is given with the food. If both taurine and cholic acid be fed, however, the excretion of tauro- cholic acid increases. The relative amounts of taurocholic and glycocholic acids in the bile of different animals differ considerably. Human bile contains relatively a small amount of taurocholic acid; on the other hand, the bile of the dog contains a large excess of it. Cholesterol In human bile the percentage of this important substance is not high (1.6 parts per 1000), but it is of great clinical importance because of the fact that it may separate out as a precipitate forming gallstones. The THE BIOCHEMICAL PROCESSES OF DIGESTION 495 percentage of cholesterol in these varies from 20 to 90 ; the remainder being organic material such as epithelial cells, inorganic salts, pigment, etc. The origin of cholesterol is partly endogenous and partly exoge- nous. In the former case it comes from the envelope of red blood cor- puscles and from the nervous tissues, where it is present In considerable amount. The latter source is, of course, the food. The increase in; cholesterol esters in the blood after feeding with food rich in this sub- stance has been shown, particularly in rabbits. That the bile should be the pathway through which cholesterol is excreted depends no doubt on the fact that it contains bile salts, which along with their other properties have a remarkable solvent action on cholesterol. This solvent property depends on the cholic acid part of the bile salts, which, as already remarked, is chemically very closely related to cholesterol; indeed, the relationship is so close that some have suggested that cholic acid is derived from cholesterol. This would mean that the cholesterol of blood is excreted in two ways, as cholesterol and as cholic acid. Other observers, however maintain that the cholesterol is excreted mainly by the lining membrane of the gall bladder, and that this explains why gall-bladder bile contains more of it than fis- tula bile. This evidence is, however, not very strong, for the greater excretion of cholesterol under conditions where the circulation of bile is going on may be explained as due to the presence of bile salts, which serve to carry the cholesterol out of the blood. Many problems remain to be elucidated in connection with the metabolic history of cholesterol. That some of it is absorbed when cholesterol is contained in the food might seem to indicate that its source is entirely exogenous, Against this view, however, stand two facts: (1) that the cholesterol in the feces of herbivorous animals is of the same variety as that present in those of carnivorous animals and not the phytosterol which is present in plants; and (2) that the universal presence of cholesterol in cells indicates that it must be manufactured there. The Bile Pigments The pigments of bile are bilirubin and biliverdin. The latter is pro- duced from the former by oxidation. If the oxidation be carried a stage further, a blue pigment called bilicyanin is formed. This process of oxidation can be observed in the ring test for bile pigment with fuming nitric acid. When bilirubin is reduced, urobilin, one of the pigments in urine, is formed. Bilirubin must therefore be considered as the mother substance of all these pigments, and it is of interest in connection with its derivation to know that it has the same formula 496 DIGESTION as iron-free hematin or hematoporphyrin, which is produced by treating hemoglobin with concentrated sulphuric acid. Chemical investigation has shown that bilirubin is built up from sub- stituted pyrrols, probably four such being contained in the molecule. The pyrrol group is also present in indole and tryptophane, and con- sists of four carbon atoms and an NH group linked together as a ring (see page 604). Similar pyrrol derivatives can be produced by decom- posing chlorophyl, the green coloring matter of plants. It is important to remember that bilirubin is acid in nature, and, therefore, can com- bine with alkalies to form salts. The relative amounts of bilirubin and biliverdin vary in the bile of different animals. When these pigments enter the intestine they are reduced to urobilin, part of which passes out with the feces, another part being absorbed into the blood and excreted in the urine. Part of that excreted in the urine exists, however, as a so-called chromogen named urobilinogen. The urobilinogen is converted into urobilin by the action of oxygen. The method by which urobilin is produced from blood pigment has been studied by histologic examination of the liver particularly of birds and amphibia, in which destruction of blood pigment goes on rapidly. Increased destruction of blood pigment can be induced by poisoning with certain substances such as arseniureted hydrogen. From such studies it is usually believed that the bile pigments are a peculiar product of hepatic activity, being produced from blood pigments that are de- rived from erythrocytes which have been broken down either in the liver itself or in some other viscus (e. g., the spleen). Whipple and Hooper20 have brought forward seemingly incontrovertible evidence against such a view. They have found, for example, that the bile pigments are formed just as readily in animals in which the circulation of the liver was greatly curtailed by anastomosing the portal vein with the vena cava (Eck fistula) as in normal animals. Even when the circulation was limited to the anterior end of the animal (head and thorax) bile pigment appeared in the blood when hemolyzed erythrocytes were in- jected, and it was also formed when hemoglobin was placed in the pleural and peritoneal cavities. The endothelial cells of the blood vessels and elsewhere can evidently form the pigments, at least when the liver is absent. When such a process occurs under normal conditions, it is quite probable that the liver acts merely as an excretory organ for the pig- ments in the same way as the kidney does for urea. Possessed of endo- thelial cells, the liver might itself also produce some of the pigments, but no more than other organs with a similar number of those cells. Even the derivation of bile pigments from hemoglobin is called in question, for the same workers have observed that, whereas the excre- THE BIOCHEMICAL PROCESSES OF DIGESTION 497 tion of pigment from a biliary fistula is remarkably constant in a dog fed on a fixed mixed diet, it became increased, sometimes by 100 per cent, when the diet was changed to one of carbohydrates, and depressed on a diet of meat. The question arises as to whether, after all, the bile pigments are really derived from broken-down hemoglobin. May they not be manufactured de novo out of other materials? Whipple and Hooper have also shown that bile is a most important secretion, for dogs rarely survive on an ordinary diet if bile is perma- nently prevented from entering the intestine. Intestinal symptoms soon supervene, and become progressively more severe until the death of the animal. Feeding with bile does not relieve the condition, but feeding with cooked liver seems to have a beneficial effect. After extravasation of blood in the subcutaneous tissues, as in a bruise, for example, a decomposition of hemoglobin proceeds quite like that occurring in the liver, and leads to the production of blue and brown and green pigments like those of the bile. When hemolysis is produced, as by inhalation of arseniureted hydrogen or the injection of inorganic or biological hemolysins, there is an immediate increase in the amount of bile pigment in the bile. Even the injection of hemoglobin solutions has this effect. Under these conditions of hemolysis, besides an increase in urobilin, there may be considerable quantities of hemoglobin secreted in the urine. Bile salts and pigments usually accompany each other when any- thing occurs to interfere with the free secretion of bile. For example, after ligation of the bile duct both bile pigments and bile salts accumu- late in the blood, in the serum of which they may be recognized by the ordinary chemical tests in from four to six hours after the operation. If the accumulation be allowed to proceed further, the bile pigments become deposited in the tissues, giving them the peculiar yellowish ap- pearance known as jaundice. Under these conditions the bile salts and pigments also appear in the urine. The accumulation of bile salts in the body affects certain physiologic processes; for one thing, it causes a great lengthening in the clotting time of the blood. If the blood supply to the liver is interrupted by ligation of the portal vein and hepatic artery at the same time that the bile ducts are occluded, not a trace either of bile salts or of bile pigment appears in the blood during the six to eighteen hours that the animals survive the operation. The amount of obstruction of the bile duct necessary to produce these symptoms is very slight, since bile is secreted at a very low pressure. Even a clot of mucus or a swollen condition of the mucous membrane of the duct is sufficient to produce obstruction. In the discharge of bile from the gall bladder into the duodenum it is claimed by Meltzer21 that a 498 DIGESTION reciprocal relationship exists between the contraction of the bladder musculature and the relaxation of the muscular fibers surrounding the duct in the duodenum. If this reciprocal innervation fails to operate properly, discharge of bile into the duodenum may become obstructed so that a certain amount passes back into the blood, as in cases of bile- duct obstruction. Bile also contains a certain amount of lecithin and other phospholipins. The amount varies considerably in the bile of different animals, even in animals of the same species.- It is probably derived, as already men- tioned, like the cholesterol, from the breaking-down of red blood cor- puscles that goes on in the liver. It is no doubt digested by the ferments of the intestinal tract, the liberated cholin, since it is toxic if absorbed, being further attacked by bacteria so as to become converted into cer- tain substances of a nontoxic nature. CHAPTER LVII BACTERIAL DIGESTION IN THE INTESTINE On an average diet, in twenty-four hours the feces of man weigh about 100 grams, or after drying, about 20 grams. About one-fourth of the dry matter consists of the bodies of bacteria. If plated out by the ordinary bacteriologic methods, however, it will be found that only a small proportion of these bacteria are living. The greater number have been destroyed, probably by the action of the mucin in the large intes- tine. The nitrogen content of the feces amounts to about 1.5 grams a day, of which about one-half is bacterial nitrogen. If the diet contains large quantities' of cellulose material, as in green vegetable food and fruit, the mass of feces as well as the bacterial content may be consid- erably greater.' The foregoing facts indicate that very extensive bacteriologic proc- esses must be going on all the time in the intestinal contents, and the question arises as to whether such action is beneficial or otherwise to the animal economy. To answer this question interesting observations have been made on the growth and well-being of animals excised from the uterus under strictly sterile conditions and maintained thereafter on sterile food. Such observations made on guinea pigs have shown that the animals thrive and grow perfectly for a considerable time. Experi- ments carried out on chicks have not, however, yielded similar results. Chicks hatched out from the egg under strictly sterile conditions and then fed on sterile grain, do not thrive, but do so if with the grain is mixed a certain amount of fowl excrement. These experiments, appar- ently contradictory in their results, show that for certain groups of animals bacteria are required, but not for others. The difference is probably dependent on the nature of the foods. It will be remembered that the size of the large intestine varies consider- ably according to the nature of the diet (see page 463). Animals taking great quantities of cellulose foodstuffs have very large ceca and very long large intestines; whereas those which, like the cat, live practically entirely on cellulose-free food, have a rudimentary large intestine. The size of the lower intestine is obviously dependent on the presence or absence of cellulose in the food. It will be remembered also that the forward movement of the contents of the large intestine is very slow; indeed, special provision is made, by the presence of the so-called anti- 499 500 DIGESTION peristaltic wave, to delay its movement. This suggests that an important digestive process must be proceeding in this part of the gut. In these ways conditions become established in the cecum for the active opera- tion of bacteria. They attack the cellulose, and liberate the more diges- tible foodstuffs contained in the vegetable cells, also producing out of the cellulose itself materials of nutritive value. The acids that are also produced by this process are neutralized by the carbonates secreted by the mucosa. In certain herbivorous animals — the ruminants — this process in the cecum is not relatively of such importance, because it takes place in the paunch. The animals swallow the food and it mixes in this part of the stomach with the saliva, so that bacteria and ferments contained in it, called cytases, attack the cellulose, liberating the more easily digested foodstuffs inclosed within the cell walls. As this process goes on acids accumulate in the digestive mixture. The food is then returned to the mouth, chewed over again, and swallowed again into the main stomach, where it is digested. The aid which bacteria render to digestion depends therefore on the nature of the diet. Man, being omnivorous, stands mid- way between the two groups of animals discussed above. Although the cellulose contained in his food is not itself sufficiently digested to furnish nutriment, yet it is so far acted upon as to permit the rupture of the cell, the contents of which are then digested. The cellulose is, however, of value in furnishing bulk to the intestinal contents — ''intestinal bal- last," it is sometimes called. In the small intestine in man there are bacteria capable of acting on carbohydrates and producing from them organic acids, such as lactic, acetic, etc. So long as a sufficiency of carbohydrate exists to encourage the action of these bacteria, others having an action on protein do not seem to thrive. It may be that this is to be accounted for partly by the production of acid substances by the carbohydrate fermentation, and partly by the fact that, as soon as the protein molecule is broken down by the digestive enzymes, its building-stone ammo acids are ab- sorbed. There are probably also bacteria in the small intestine capable of splitting fat into fatty acid and glycerine, but practically nothing is known of their action. In the large intestine of man, along with the cellulose-digesting bacteria already mentioned, protein-digesting bac- teria are also present. These bacteria belong to the class, Bacillus coli communis, the various members of which are known as faculative anae- robes because they can grow in the presence or absence of oxygen. If bacterial growth is excessive or there is an insufficiency of carbohy- drates in the small intestine, the bacteria attack the amino acids pro- duced by the digestive enzymes and decompose them into products that may be toxic if absorbed into the blood. BACTERIAL DIGESTION IN THE INTESTINE 501 Bacterial Digestion of Protein From a pathologic standpoint, the most important action of bacteria is that which takes place on protein. Under anaerobic conditions the intestinal bacteria have in general the power of splitting off the amino group whereas under aerobic conditions they split off the carboxyl group. This splitting off of the carboxyl group as carbon dioxide is per- formed by the so-called carboxylase bacteria, and it may take place either before or after deamidization (see page 615). If it happens after this process, the products are not highly toxic and include phenol, cresol, indole and skatole, which are partly absorbed into the blood and partly excreted with the feces. The fractions of those substances that are absorbed into the blood have their toxicity removed by conjugation mainly with sulphuric acid to form the so-called ethereal sulphates. A part is also combined with glycuronic acid (see page 632). In the case of phenol and cresol this conjugation occurs immediately after absorption, but in the case of indole and skatole it is preceded by an oxidative process, converting these substances into indoxyl and skatoxyl respectively. The detoxica- tion process occurs in the liver, as has been shown by experiments in which this organ was artificially perfused outside the body. They are then removed from the blood by the kidneys and excreted in the urine. The proportion of ethereal sulphates in this fluid is therefore an indica- tion of the extent of intestinal putrefaction of protein (see page 632). The indican, being readily detectable by the well-known color reaction of Jaffe, serves as an indicator of the extent of intestinal putrefaction. The indole and skatole which are not thus absorbed and detoxicated are excreted with the feces, to which they give the characteristic odor. The source of the phenol is tyrosine and that of the indole is trypto- phane. The chemical processes involved are shown in the following equations, in which the by-products of the reactions are in brackets. C.OH /\ HC CH COH /\ HC CH HC CH HC CH CH2 CHNH2 COOH (tyrosine) '(NH3) CH, CH., HC CH C CH, (CO2+H20) COOH (p-oxyphenyl- propionic acid) COOH (p-oxyphenyl- acetic acid) (CO,) COH COH HC CH HC CH HC M CH HC CH > / ^ V 1 1E3(C02 + H20) (paracresol) (phenol) 502 DIGESTION Putrefaction of tryptophane is probably preceded by deamidization : CH CH HC C C— CH.,.CHNH2.COOH HC ' C C— CH.,CHrCOOH I il II — > I II I! — > HC C CH (NH,) HC C CH (CO2 + H2O) v/\/ \/\/ CH NH CH NH (tryptophane) (indole-propionic acid) CH CH CH HC C C— CH...COOH HC C CH HC C C— CH, HC C CH (CO, + 11,0) HC C CH HC C C \X\X \X\X \/\/ CH NH CH NH (+CH3) CH NH (indole-acetic acid) (indole) (skatole) If, however, the carboxylase bacteria remove the carboxyl group be- fore the amino group has been removed, highly toxic substances called amines are produced. They are the so-called ptomaines. From alanine, ethylamine is formed; from tyrosine, phenolethylamine; from histidine, which it will be remembered is an important protein building-stone, imidazylethylamine, and so on. The process of formation is illustrated in the accompanying formulae: 1. CH3.CH(NH,).COOH — CO, -+- CH3.CH..(NH,) Alanine Ethylamine 2. C6H,(OH).CH:,CH(NH2).COOH = C6.! + C6H,(OH).CR!.CH2.NH2 Tyrosine Phenylethylamine 3. C3N2H3.CHJ.CH(NH2).COOH=:CO.! + C3H3N2. CH2'.CH2.NH2 Histidine. Imidazylethylamine. Similar substances are very common in the metabolic products of plants; for example, they constitute the active principle of ergot. They are also no doubt produced in the tissues of mammals, imidazylethyla- mine, commonly called histamine, being thus produced, as well as the closely related epinephrine, which is the active principle of the supra- renal gland (see page 737), and may be described as a methylated ethyla- mine derivative of tyrosine. Phenylacetic acid produced by a similar process from tyrosine may be excreted in the urine, where it forms the mother substance of homo- gentisic acid, to which the dark brown color of the urine in alkaptonuria is due. The great importance attached to these decomposition products of proteins depends on the fact that they have powerful pharmacologic actions. These actions are developed very largely upon the vascular system; histamine, for example, produces marked vasodilatation and lowers the coagulability of the blood, whereas other substances of the BACTERIAL DIGESTION IN THE INTESTINE 503 same class, like epinephrine, have the property of raising the blood pres- sure. In larger doses, serious nervous symptoms and a condition of pro- found collapse are produced. These observations have led several inves- tigators to believe that the persistent occurrence of bacterial fermen- tation and the absorption of the resulting decomposition products of protein into the blood ultimately cause arteriosclerosis and the other symp- toms that accompany senescence. It is difficult at the present time to know how much of this one ought to believe, although it can not be doubted that putrefaction has an unfavorable action on the arteries, and that an excessive degree of it causes the symptoms of ptomaine poisoning. If the ptomaines have formed in the food before it is eaten, the symp- toms develop in from one to five hours after the meal, but if the decomposi- tion occurs in the intestine on account of bacteria that are taken at the same time as the food, the ptomaines may not have developed sufficiently to cause symptoms until from twelve to forty-eight hours; sometimes, how- ever, they develop in an hour or so. Prominent among the symptoms is usually diarrhea, which develops for the purpose of getting rid of the offending bacteria and ptomaines. Actual infection of food with bacteria of the paratyphoid-enteritidis type is much more common than poisoning by substances (ptomaines} that have been generated in food before it is taken (Jordan17). Meat, milk and other protein foods are usually the carriers of the bacilli, and in most of the accurately recorded cases the meat or milk was found to be derived from animals suffering from enteritis or some other infection. Sometimes, however, perfectly good food may become infected by handling. Although the symptoms are usually acute, they may closely simulate those of typhoid fever, and the effects of the attack may linger for weeks or months. BOTULISM The commonest type of poisoning by substances actually present in the food is that known as botulism. In this the gastrointestinal symptoms are not pronounced, — indeed, paralysis of the intestinal tract with con- stipation is the rule, — but those affecting the nervous system, dizziness, diplopia and other visual disturbances, with difficulty in swallowing, are very prominent. The temperature and pulse are usually normal. In practically all of the reported cases of botulism, the source of infection has been food which after having been subjected to some preliminary treat- ment, such as smoking, pickling, or canning, had been allowed to stand for some time and then eaten without cooking. The Bacillus botulinus, which is responsible for the production of the poisons or toxins, is a 504 DIGESTION strict anaerobe and is readily destroyed by cooking, as are also the poisons. Antitoxins are formed by sublethal injections. Another but now very rare example of poisoning by products formed in food is that caused by "ergotoxin. " The treatment in such cases is to encourage diarrhea by giving pur- gatives. If the intoxication is of a more chronic character, the symptoms are vague, consisting of drowsiness, lassitude, headache, and general de- pression. The treatment here also is to clear out the intestines by a good purge. There can be little doubt that many of the unhealthy condi- tions of the skin leading to the formation of pimples, acnes, and boils, are also caused by chronic intoxication with protein decomposition prod- ucts. Again, purgation is the proper treatment. It is unnecessary in a work of this character to go further into these highly important questions. It is probable, however, that the importance of the relationship of excessive protein putrefaction in the intestine to many of the so-called minor diseases can not be overemphasized. On the other hand, we must be careful not to attribute every sort of chronic condition to this putrefaction. Toxemia is often a shibboleth of the profession. "When a chronic disease can not be diagnosed, it is put down as a toxemia. This, however, is not medical science — it is medical shirk- ing. It is certainly unsafe at the present time to conclude that the ordinary symptoms of senescence, such as hard arteries or increased blood pressure, are invariably to be attributed to this cause. It will be re- membered that Metchnikoff is largely responsible for such a view, and also that he suggested, as the surest way to ward off the chance of such intoxication, the taking of buttermilk, which would supply bacteria through whose growth in the intestine the protein-destroying bacteria would not be able to thrive. It is probable that the same result could be attained in patients showing undoubted signs of suffering from intestinal putrefaction by a change in diet in the direction of giving more carbo- hydrate, for, as we have seen, if there is a plentiful supply of this food- stuff in the small intestine, the bacteria do not tend to attack the protein. Before leaving this subject it is interesting to consider for a moment the cause of the severe symptoms that follow intestinal obstruction. This question has recently been diligently investigated by Whipple,18 who found that the nonprotein nitrogen of blood (page 606) becomes greatly increased in intestinal obstruction. The cause for this increase in non- protein nitrogen is found to be an excessive breakdown of tissue protein caused by the absorption into the blood of a proteose. When this pro- teose isolated from obstructed loops of intestine was injected into fast- ing dogs, profound symptoms of depression were produced, followed, in cases in which the dose was sublethal, by recovery in from twenty-four BACTERIAL DIGESTION IN THE INTESTINE 505 to forty-eight hours. Along with these symptoms the nitrogen elimina- tion by the urine increased by 100 per cent. A very interesting fact is that animals can be rendered immune to this proteose by progressively increasing periodic administration. When they are thus immunized, the toxic symptoms do not follow upon its injection, nor are the symp- toms produced by artificially creating an intestinal obstruction. Con- versely, when a chronic toxic condition is -kept up by a partial obstruc- tion, such as that produced by making a gastrojejunal fistula and occlud- ing the duodenum, the animals are less susceptible than normal ones to proteose injection. We have here and there incidentally referred to the reaction of various parts of the gastrointestinal contents, but we would call attention once again to this important subject, especially since many points of uncer- tainty have recently been cleared up by the accurate observations of Long and Fenger,19 who used the electrometric method for measuring the hydrogen-ion concentration. The contents of the duodenum removed by means of the Kehfuss tube in man showed a reaction varying from dis- tinctly acid to slightly acid, depending upon the proximity of the tube to the pylorus or papilla, this position being determined by x-ray exam- ination. The slight degree of alkalinity is surprising. Lower down in the duodenum the reaction was as frequently acid 'as alkaline, the de- gree of acidity, however, being so slight as to favor rather than retard the digestive powers of the pancreatic juice. To determine the reaction lower down, the observations were made on recently slaughtered animals (pigs, calves, and lambs), the small intes- tine being tied off in loops of the upper, middle, and lower thirds. The contents of the last loop were often alkaline, but might be more acid even than those of the first, which were usually faintly of this reaction. Con- siderable variations were, however, the rule. The mixed intestinal con- tents of a recently fed dog, removed immediately after death, gave PH = 6.79 ; i. e., very faintly acid. DIGESTION REFERENCES (Monographs) iPavlov, J. P. : The Working of the Digestive Glands. Trans, by Sir W. H. Thomp- son, London, Griffin, ed. 2, 1910. sStarling, E. H. : Eecent Advances in the Physiology of Digestion, W. T. Keene & Co., Chicago, 1907. 3Cannon, W. B.: The Mechanical Factors of Digestion, Internat. Med. Monographs, London, Ed. Arnold, 1911. ^Carlson, A. J.: The Control of Hunger in Health and Disease, Univ. of Chicago Press, 1917. sTodd, T. Wingate: The Clinical Anatomy of the Gastrointestinal Tract, Manches- ter, Univ. Press, 1915. 506 DIGESTION (Original Papers) ^Cannon, W. B., and Cattell, McKeen: Am. Jour. Physiol., 1916, xli, 39. caQesell, E.: Proc. Am. Physiol. Soc., Am. Jour. Physiol., 1918, xlv, 559. 7Dale, H. H., and P. P. Laidlaw: Proc. Phys. Soc., Jour. Physiol., 1912, xliv, pp. 12, 13. 7aBabkin, Bi P., Rubaschkin, W. J., and Ssawitsch, W. W.: Arch. f. mikr. Anatomic, 1909, Ixxiv, 68. sMacallum, A. B.: Ergeb. der Physiol., xi, 598-657. oMiller, F. E.: Quart. Jour. Exper. Physiol., 1913, vi, 57. icEdkins, J. S.: Jour. Physiol., 1906, xxxiv, 133-144. loaKeeton, E. W., and Koch, F. C.: Am. Jour. Physiol., 1915, xxxvii, 481; also Popielski, L.: Arch. f. d. ges. Physiol., 1901, Ixxxvi, 215. nMeltzer, S. J.: Am. Jour. Physiol., 1899, ii, 266. i2Cannon, W. B.: Am. Jour. Physiol., 1898, i, 359. isCannon, W. B., and Blake, J. B.: Am. Surg., 1905, xli, 686. Cf. No. 3. i^Macallum, A. B.: See Fitzgerald, M. P., Proc. Roy. Soc., Ixxxiii, B, 56. isHarvey, B. C. H., and Bensley, R. E.: Biol. Bull., Wood's Hole, 1912, xxiii, 225. i^Long, J. H., et al.: Jour. Am. Chem. Soc., 1917, xxxix, 162 and 1493; also ibid., 1916, xxxviii, 38. 17 Jordan, E. V.: Food Poisoning, Univ. of Chicago Press, 1917. isWhipple, G. H., Cooke, J. V., and Stearns, T.: Jour. Exper. Med., 1917, xxv, 479. Also Whipple, G. H., Stone and Bernheim: Ibid., 1913, xvii, 286 and 307. isLong, J. H., and Fenger, F.: Jour. Am. Chem. Soc., 1917, xxxix, 1278. soWhipple, C. H., and Hooper, C. W. : Am. Jour. Physiol., 1916, xl, 332 and 349 ; ibid., 1917, xlii, 257 and 264; Hoope: Ibid., p. 280. ziMeltzer, S. J.: Am. Jour. Med. Sc., 1917, cliii, 469. CHAPTER LVIII THE EXCRETION OF URINE BY R. G. PEARCE, B.A., M.D. It will be advisable to introduce the subject by a brief review of the essential structural features of the kidney, in so far as they apply to the excretory function of the organ. STRUCTURE OF THE KIDNEY The kidney is mainly derived from the surface of the celom, and is a mesodermal structure. In this respect it differs from ordinary secreting glands, which are endodermal in origin. Just as it is more or less unique in its development as a gland, it is also unique in its method of functioning. The physiologic theories of the mechanism of urinary secretion are closely related to the highly characteristic structure of the kidney. For this reason a brief survey of the structure of the different parts of the uriniferous tubules and the epithelial cells with which these are lined, is advisable. The uriniferous tubule, which is the secreting unit of the kidney, takes its origin in the capsule of Bowman, which may be likened to a hollow sphere of very delicate epithelium, one side of which is invaginated by a very much convoluted capillary mass, the glomerulus. The capsule opens up by a narrow twisted neck into a tubule, which is rather tortuous in the cortex (the proximal convoluted tubule), but soon takes a sharp descending course in the medulla towards the pelvis of the kidney, and doubles back (loop of Henle) in a straight course again to the cortex, where it again makes a twisted course (the distal convoluted tubule), and terminates in a collecting tubule, which, uniting with other tubules, collects the urine and conducts it to the pelvis of the kidney. The capsule is lined with very thin epithelial cells, especially over the capillaries comprising the glomerulus. The proximal and distal tubules 507 508 THE EXCRETION OF URINE contain epithelium showing a prominent striation. These striations are rows of granules, which run towards the lumen of the cell, becoming less distinct as they approach it and apparently standing in close rela- tionship to the rather prominent internal (lumen) striated border of the cell. Some histologists believe that the striations at the border are Fig. 170. — Diagram of the uriniferous tubules (C) the arteries (A), and the veins (B) of the kidney. really cilia, which are described as being immobile. The cilia are shown in Fig. 171. The descending limb of Henle's loop is lined with a thin pavement epithelium with large bulging nuclei. The distal convoluted tubule is lined with cells not unlike those found in the proximal tubules, except that the inner border is not striated. The diameter of the lumen THE EXCRETION OF URINE 509 of the capsule varies with the activity of the kidney, as is shown in the following figures given by Brodie and Mackenzie.1 RESTING KIDNEY MM. KIDNEY DURING DIURESIS MM. Mean 'diameter of capsule " " " glomerulus ' ' " space of capsule Lumen of proximal convoluted tubule " " distal « " 93.4 90.4 3.0 0.0 7.2 123.8 100.0 23.8 17.6 20.6 The urinary tubule has a remarkable blood supply. The renal arteries arise directly from the abdominal aorta and are very short. They run through the medulla to the cortex, and join with neighboring arteries to B. Fig. 171. — Cross sections of convoluted tubules from kidney of rat. A, during slight secretion; B, during maximal secretion. (From Sauer.) form arches from which proceed branches, that radiate into the cortex and give off smaller branches each of which very shortly breaks up into a small capillary tuft, — the glomerulus, — which lies in the invaginated sphere of Bowman's capsule. The capillaries collect into an efferent vessel, which appears to be smaller than the afferent artery, and this vessel in emerging from the capsule again breaks up to form a capillary network about the con- voluted tubules, forming their sole blood supply. These capillaries coalesce to form the renal vein. The blood of the kidney must, accord- ingly, pass through two sets of capillaries. The kidney is richly supplied with nerves, which are for the most part derived from the celiac ganglion and are in connection with the splanch- 510 THE EXCRETION OP URINE nic and the vagus. Other branches from plexuses in the region of the suprarenal body and the aorta join with those coming from the celiac ganglion to form what is known as the renal plexus, which is arranged in a network along the blood vessels and on the walls of the pelvis of the kidney. These fibers are distributed to the very smallest blood ves- sels, and nerve fibers have been observed among the cells of the tubules. THE MECHANISM OF THE EXCRETION OF THE URINE The great number as well as the variety of substances which are pres- ent in both the blood and the urine makes it appear improbable that urine excretion is dependent upon chemical combinations within the renal cells, and leads us to seek a physicochemical mechanism to explain the phenomenon. Can we discover the processes by which the kidney fabricates a highly concentrated solution of salts from a very dilute solution of the same salts in the blood plasma ? The problem is compli- cated by the fact that the ratios existing between the concentration of each urinary salt in the urine and the concentration of the same salt in the blood are different. In other words, the urine is not merely concentrated blood plasma freed from protein. The passage of Avater and salts through the capillary wall and through the basement membrane surrounding the renal cell probably takes place by simple diffusion. If it were otherwise, an expenditure of energy would be required, and it is difficult to understand how a basement membrane could bring about energy changes. Any substance to which the cell membrane is permeable will diffuse into the cell until an equi- librium is established between its concentration within the cell and that of the lymph or blood plasma. A nondiffusible substance will not' enter the cell because it can not pass through the cell membrane, and if it exerts an osmotic pressure, it will also tend to keep the water in which it is dissolved from entering. If water does pass into the cell under these conditions, it is due to the expenditure of energy opposed to and greater than that which is offered by the osmotic pressure of the nondiffusible substances. Possible sources for such energy are the pres- sure of the blood in the renal capillaries, which would exert a force op- posite to that of its osmotic pressure, and the presence within the cell of a concentration of salts greater than is present in the blood, and able to exercise a sufficient osmotic force to draw fluid into the cell against the osmotic force of the nondiffusible salts. The passage of the urinary constituents through the cell might also be due to- simple diffusion, the substances passing through the cell to be extruded on the other side in THE EXCRETION OF URINE 511 the same concentration as in the blood. In this case, the renal cells would act merely as a filter, the urine having the same concentration of each urinary salt as is present in the blood. A comparison of the concentrations of the urinary salts in the urine and the blood shows, however, that the urine is not merely a deprotein- ized blood plasma, so that other factors must be sought to explain the excretion. Since the concentration of the urine requires the expenditure of much more energy than is provided by the known physical factors, it is generally accepted that the renal cell in some manner supplies this energy by its metabolic activity. It is impossible at present even to surmise the nature of the process. Two possibilities may be considered. One is that the urine is a filtrate of the blood which has passed through a portion of the renal epithelium into the tubules as a very dilute fluid, resembling the blood plasma minus its colloidal substances, and that this dilute fluid is concentrated by the reabsorption of fluid and of salts by other cells of the kidney, and again replaced in the blood stream. The other is that the salts and fluid are each actively and individually ex- creted by the kidney. Whichever condition is the true one, the fact remains that the change in the concentration entails the expenditure of a great amount of energy on the part of the renal cells. The energy which the kidney must use in the actual work of concen- trating the urine from the fluid of the blood plasma can not be com puted from a comparison of the concentration of the urinary salts as a whole in both the blood and the urine. Each constituent must be con- sidered apart. We can not, for example, determine the molecular con- centration of the blood plasma and the urine (by measuring A) (page 10) and estimate the work which is expended in producing the con- centration from the observed difference. On the basis of such comparisons, however, it is said that the excretion of 100 c.c. of urine requires at the minimum 500 kilogrammeters of work (Cushny2). Even this conserva- tive estimate may be wrong, for- it does not take into consideration the possibility that the excretion of water by the kidney requires energy expenditure on the part of the renal cells. „ „ f. V .,,». a . fr , V rt * Theories of Renal Function For many years two rival hypotheses have dominated the teaching of the mechanism of renal function. Bowman and Heidenhain postulated that the constituents of the urine are secreted by the vital activity of the epithelium of the capsule and the tubules. The glomerular capsule secretes the water and the easily diffusible salts in a dilute solution, and the uriniferous tubules add to this fluid the various organic and inor- ganic salts to bring the urine to the necessary concentration. This 512 THE EXCRETION OF URINE theory has been termed the vital theory. Ludwig, on the other hand, advanced what is termed the physical theory, which holds that the glomerulus and capsule act simply as a filter, which allows the fluid of the blood plasma to pass through in a very dilute solution and in large amounts. This fluid is concentrated by physicochemical processes on its passage along the urinary tubules to the pelvis of the kidney. Both of these theories are inadequate and fail to explain the phenom- ena which research has shown to occur in the kidney, but they have served to develop what Cushny terms a modern theory of urinary excretion. The Modern Theory of Urine Formation. — This theory accepts the general scheme of filtration and reabsorption of Ludwig, but pays due respect to the fact that the known physical forces are not adequate to explain the reabsorption which must occur in the tubules. It therefore supplements Ludwig 's theory by assuming a vital activity on the part of the epithelium of the tubules in reabsorbing fluids and salts from the dilute filtrate coming from the glomerulus and capsule. A large amount of plasma fluid is filtered through the walls of the glomerular vessels. This fluid has the same concentration of the salts to which the capsule is permeable as does the blood plasma, but it is free of the col- loidal substances normally present in the plasma. The blood leaving the glomerulus is therefore a somewhat concentrated solution of plasma col- loids, and must have returned to it the proper amount of water and salts to make it an optimum fluid for the body cells. This is accomplished by active absorption from the glomerular filtrate. The salts that are of no use to the body are not reabsorbed and therefore appear 'in highly concentrated form in the urine. These salts are termed nonthreshold sub- stances, and since their presence in the plasma is unnecessary, they con- tinue to be excreted as long as they are present in any concentration in the blood. The salts that are necessary for the plasma are termed threshold substances, and are reabsorbed until they are again present in the plasma in optimal strength. For example, urea continues to be ex- creted as long as any is present in the blood, while glucose is completely reabsorbed so long as its concentration remains under a more or less fixed level. It is impossible to give a summary of the arguments which have been advanced in support of any of the theories. However, since the modern theory appears to offer a better explanation of the established facts, it may be wise to recount some of the best experimental evidence in support of it. First, we must inquire as to the amount of deproteinized blood plasma which the capsule must filter off from the blood in order to furnish the THE EXCRETION OF URTNE 513 amount of the various salts excreted each day and the amount of water absorbed by the epithelium of the tubules to account for the concentra- tion in which the salts are found in the urine. In order to produce 20 grams of urea in 1200 c.c. of urine, 60 liters of blood-plasma fluid con- taining 0.03 per cent of urea would have to be filtered through the cap- | 20 sule [ — =6000), and 5.9 liters of water returned to the blood from I 0.03 the uriniferous tubules. Since the bloodflow through the kidneys is very great, at least 500 liters per day, only about 13 per cent of the fluid con- tained in the blood passing" through the glomerulus would pass by filtration through the capsule of Bowman. The fact that such a large amount of fluid would have to be reab- sorbed from the uriniferous tubules (59 liters) is a possible a priori criticism of the theory, but Cushny points out that the amount each tubule would have to absorb per hour would be very small (in his ex- periment on a cat amounting to less than 0.014 c.c. per hour). The filtration of the protein-free blood fluid through the renal capsule, like that through any other membrane, depends on several factors. (1) There must be a difference in the pressure between the blood and the urinary filtrate. , In the laboratory the pressure used in filtering is usually supplied by gravity, but in the case of the filtration of the urine through the capsule the force is furnished by the pressure of blood in the glomerular vessels. (2) The character of the filter determines what substances shall pass. The renal capsule is a membrane normally im- pervious to the proteins of the blood, but pervious to the other constitu- ents. Under certain conditions it loses this character. (3) The char- acter of the fluid determines how readily it will filter through the mem- brane. If the fluid contains a substance which can not pass through the filter and which exerts an osmotic pressure in opposition to the filtering force, the rate of filtration as well as the amount filtered, will be reduced. If the capsule acts as a filter it should be possible to alter the rate of urine excretion by varying any of these factors, and experimentally this is true. The factors can be varied in several ways. If the blood pressure is raised by tying off several of the branches of the aorta, the urine is appreciably increased, or if the blood pressure is decreased, as can be done by compressing the renal artery by means of a screw clamp, the amount of urine is decreased. In the artificially perfused kidney, the fluid exuding from the ureter increases as the pressure of the perfusion fluid is increased, arid decreases as the pressure is decreased. Whether changes in the pressure in the blood are directly responsible for variations in the rate of urine excretion, or whether they act indirectly by varying the rate of the bloodflow in the kidneys, has been the subject of much 514 THE EXCRETION OF URINE debate. Probably both factors are involved, as is shown by the follow- ing observations. If the blood pressure is increased by vasoconstriction in the splanchnic area produced by stimulation of the splanchnic nerves, the flow of blood through the kidney is decreased and the excretion of urine falls. Apparently, secretion can continue only as long as the col- loids of the plasma are not notably increased, for, as the osmotic pressure due to the indiffusible colloids rises, the pressure in the capillaries is no longer able to oppose it. The same point has been beautifully shown by Starling and his pupils, who found that the secretion of urine ceases when the capillary pressure in the glomerulus fell below that exerted by the osmotic pressure of the blood proteins, the critical pressure being from 30 to 40 mm. Hg. They also found that dilution of the blood with saline solution by reducing the osmotic pressure of the proteins in the plasma, was accompanied by an increase in the rate of excretion; excre- tion in such cases being maintained at a blood pressure below the normal critical pressure. If the dilution of the blood was made with saline con- taining gelatin or gum arabic, on the other hand, the diuretic effect was greatly decreased, and any fall in the blood pressure was followed by a suppression in the urine (Knowlton9). These experiments evidently indicate that the saline produces its diuresis by diluting the plasma proteins and loAvering their osmotic pressure, since when the osmotic pressure of the blood is maintained by the addition of colloids in which this is present, no diuresis occurs. The significance of these facts, in connection with the raising of lowered blood pressure after hemorrhage, has already been alluded to (page 139). This view is confirmed by the experiments of Barcroft and Straub,10 who showed that the oxygen consumption is often not appreciably raised during the diuresis produced by the injection of saline. If the diuresis produced by this means was due to an actual increase in the work of the kidney, the oxygen consumption would have been increased. In the frog, the glomerulus and the tubules are supplied with blood by the renal artery, as is the case in the mammal, but the tubules cu- riously enough are also supplied with some of the blood coming from the lower extremities and the trunk through a vessel which has no counter- part in the mammal — the renal portal vein. The blood, therefore, which is supplied to the tubule is a mixture from the glomerulus and the renal portal system. By ligating the renal vessels it is possible to cut off the blood supply of the glomerulus while leaving the tubules supplied by the renal portal vein. Normally the pressure in the renal portal system is not sufficient to force blood back through the glomerular vessels. Liga- ture of the renal vessels at once results in a suppression of the urine. If the glomerular vessels are perfused with Ringer's solution at a THE EXCRETION OF URINE 515 pressure equal to that found in the aorta, a considerable flow of fluid may be secured from the ureters, but no fluid is obtained when the renal portal vein is perfused at a pressure equal to that normally present in this vein. Rowntree and Geraghty11 found that phenolsulphonephthalein added to the perfusion fluid passed through the renal portal vein, did not cause secretion, but when urea was added to the perfusate, fluid con- taining the dye was obtained from the ureter. Unfortunately the pres- sure employed in these experiments may have allowed some fluid to be forced backward into the glomeruhis, so that the results may be due to filtration through the capsule. Renal artery jMalpighian corpuscle Renal-portal vein Fig. 172. — Diagram of blood supply of Malpighian corpuscle and of convoluted tubules in amphibian kidney. (Redrawn from Cushny.) It is generally accepted that the proof that the capsule acts as a filter is fairly complete. Unfortunately such decisive experimental facts can not be offered to prove the assumption that the epithelium of the tubules reabsorbs the excess of water and salts which are filtered off through the capsule. If the modern theory of urine excretion is correct, the cells of the tubules must not only absorb large amounts of water, but they must also allow for the reentrance into the blood, either completely or partially, of certain salts, while they must reject others entirely. "We have called attention above to the fact that the glomerular filtrate is very different from the urine that is finally passed. The urine contains a very high percentage of small molecules, and the proportion in which they 516 THE EXCRETION OF URINE are present is entirely -different from that in the blood plasma or in the glomerular filtrate. This is shown in the following figures, which give an average normal value for the urea, uric acid, chlorine, and glucose in 100 c.c. of protein-free blood plasma and 100 c.c. of urine. In the third col- umn is given the change in concentration which has occurred in the kidney. 100 C.C. PROTEIN- FREE BLOOD PLASMA CONTAINS 100 C.C. URINE CONTAINS CHANGE IN CONCENTRATION IN TlIE KIDNEY Urea Uric Acid Chlorine Glucose .033 .0022 .41 .1 2. .05 .6 60 22.7 1.5 Here the blood plasma fluid contained but 0.033 per cent of urea, and the urine 2 per cent. Accordingly, 6 liters of glomerular filtrate would f2 be required to furnish 100 c.c. of urine, \- - = 6000). Six liters of I 0.33 glomerular filtrate would contain 6.6 grams of sugar, 0.132 grams of uric acid, and 24.6 grams of chlorine. But 100 c.c. of urine contains no glucose, 0.05 grams of uric acid and 0.6 grams of chlorine. According to the modern theory, these figures indicate that during the passage of the urine through the tubules 5900 c.c. of water, 6.6 grams of sugar, 24 grams of chlorine and 0.067 grams of uric acid would have to be absorbed by the renal epithelium in the production of 100 c.c. of urine containing the concentration given above. Among the most convincing experiments that can be offered in sup- port of the absorption of fluid and salts by the tubules, are those in which the pressure of the urine in the tubules is slightly increased by partial closure of the ureter (Cushny). In these experiments the ureter of one kidney is partly closed with a clamp and the excretion obtained from this kidney is compared with that of the opposite normal kidney. In general, obstruction of the ureter results in a decrease in the amounts of water, chloride and urea excreted. But, curiously, the urea content is decreased relatively less than is the chloride and water content. These results can be explained on the basis that any pressure acting to oppose the head of pressure producing filtration in the glomerulus will reduce the amount of the glomerular filtration, and accordingly the time allowed for 'the passage of this filtrate along the tubules is increased and absorp- tion becomes more complete. Since urea is probably not absorbed at all and chloride is, the discrepancy in the effects on the excretion of urea and chlorine in the partially obstructed kidney can be explained. When very large amounts of water are taken by mouth, it often hap- THE EXCRETION OF URINE 517 pens that the urine excreted has a concentration of salts less than that present in the fluid of the blood. Some investigators believe that such a condition is possible only on the assumption that water is actively ex- creted, but a more plausible explanation based on the modern theory is that the water that is absorbed from the alimentary tract reaches the kidney as a dilute saline solution, and is rapidly filtered off in a form somewhat more dilute than the optimal solution which blood plasma must have for the well-being of the tissues. The tubules reabsorb the amounts of water and of those salts, such as chlorides, uric acid, and sugar, nec- essary to restore the plasma to the optimal concentration, but do not absorb the nonthreshold substances, such as urea. It is impossible to analyze the forces that are responsible for such a degree of absorption by the epithelium of the tubules. For the present Ave must classify them, for want of a better term, as vital forces. The questions that await immediate investigation are whether absorption actually takes place, and, if it does so, what factors cause it to vary. Many attempts have been made, by destroying the capsules or the tubules by means of poisons or by operation, to determine directly or indirectly the question of the function of the tubules. In such experiments, however, the number of factors involved con- fuse the issue and make the results practically valueless so far as de- termining the normal function of the tubules. Other experimenters have attempted to show absorption in the tubules by injecting diffusible substances, such as chemicals and dyes, into the ureter under what they deemed sufficient pressure to force the solution into the tubules, and by an examination of the blood or the tissues to determine whether or not the injected substances had been absorbed. The results obtained by this method are not convincing, probably chiefly because of the difficulty in reaching the tubules. Indeed, it is very questionable Avhether it is possible to inject a substance into the tubules from the ureter. Years ago Heidenhain, the exponent of the vital theory of excretion, believed that he had demonstrated the ability of the renal cells to ex- crete dye substances injected intravenously. Since he failed to find evidence of dye excretion in the capsule, but found masses of dye in the tubules and stained granules in the cells of the tubules, he concluded that the cells of the tubules had the power to excrete the dye, and from analogy he believed that the tubules must likewise excrete the water and the various urinary salts. Subsequent work, however, has failed to confirm his belief that the capsule is not concerned in the excretion of the dye, and it is as reasonable to explain the results of the experi- ments with the dyes by assuming that the masses of dye substances found in the tubules and in the cells are due to the reabsorption of 518 THE EXCRETION OF URINE water and perhaps of some of the dye from the dilute glomerular filtrate, as to accept Heidenhain's hypothesis. In the following table taken from Cushny the movements of the con- stituents of the plasma may be followed through the kidney. The ulti- mate destination of each is indicated in the enclosures. 67 LITERS PLASMA CONTAIN 62 LITERS FILTRATE 61 LITERS REABSORBED FLUID CONTAIN 1 LITER URINE CONTAINS PER CENT TOTAL IN ALL PER CENT TOTAL PER CENT TOTAL Water 92 62 1. 62 1. 61 1. 95 950 c.c. Colloids | 8 5360 gm.| — — — — Dextrose 0.1 67 gm. 67 gm. 0.11 67 gm. — — Uric acid Sodium Potassium Chloride 0.002 1.3 0.3 200 0.02 13.3 0.37 248 1.3 200 13.3 248 0.0013 0.8 " 0.32 196.5 " 0.019 11.8 " 0.40 242 ' < 0.05 0.05 gm. 0.35 3.5 ' 0.15 1.5 ' 0.6 6.0 ' Urea Sulphate 0.03 20 0.003 1.8 20 1.8 2.0 2.0 ' 0.18 1.8 ' (From Cushny.^) It will be noted that the dextrose alone is completely absorbed, and that the urea and the sulphate are not absorbed at all from the glom- erular filtrate. The other salts are partly absorbed. As already mentioned, Barcroft and Straub have shown that the diuresis which results from the injection of saline into the blood is not accompanied by any increase in the oxygen consumption of the kidney. This observation, coupled with the fact that the total amount of chloride, urea, and sulphate which is excreted during saline diuresis, is greater than under normal conditions indicates that the excretion of these salts is not due to any vital secretory power of the kidney, but rather to factors that are extrarenal in origin. The diuresis produced by adding urea or sodium sulphate to the blood, on the other hand, is accompanied by an increase in the oxygen con- sumption of the kidney. This increase can not be due to active elimina- tion of these salts by the tubules, the work of which requires oxygen, for no increase in oxygen consumption accompanies the increased ex- cretion of the same salts under saline diuresis. Sulphate and urea are nonthreshold substa'nces, and are not absorbed by the tubules. The explanation of the oxygen consumption is probably that the osmotic pressure which these bodies in the glomerular filtrate exert makes it necessary for the epithelium to oppose a greater absorbing force to con- centrate the urine, and hence a greater expenditure of energy is requird. Diuretics. — The action of the xanthine compounds — caffeine, theo- bromine and theophylliiie — in the production of diuresis is unexplained. THE EXCRETION OP URINE 519 It may be due in part to vascular changes and in part to reduction in the resistance to nitration brought about by alteration in the permea- bility of the capsule. According to the modern theory the polyuria in diabetes is produced by the excessive amount of water taken and by the inability of the kidney to concentrate the urine against the osmotic pressure offered by the concentrated sugar solution in the tubules. The presence of the hy- perglycemia in an amount higher than is present in the optimal blood plasma in this disease makes sugar a nonthreshold substance, so to speak, and none is absorbed. The diuresis following the injection of sugar is therefore of the same type as that produced by sulphate and urea. The diuretic action of the digitalis group is dependent upon its influence on the circulatory system. If the circulation is already sufficient, digitalis does not cause diuresis. The cause of the diuresis produced by pituitary extract is not known. It may be owing in part to its action on the cir- culation and in part to a direct action on the kidney. Albuminuria. — The plasma proteins ordinarily do not obtain entrance into the tubules of the kidney. In disease such as acute nephritis and cardiac failure, the plasma colloids are filtered off through the capsule, probably because of some change that has occurred in the permeability of its membrane due to inflammation or asphyxia. In these cases the urine is usually reduced in amount. Probably there is no purely glom- erular or tubular type of nephritis, both structures sharing in the dis- ability. While it can not be said that any of the so-called renal tests that have been advanced in recent years are free from criticism, they nevertheless have contributed very useful information.- The fact that the kidney of the chronic nephritic excretes a urine of more or less fixed low specific gravity would suggest that here there is an impairment of the resorbing mechanism, and the failure of a kidney to excrete the proper amount of dye, as in the phenolsulphonephthalein test, suggests an impairment in the filtering apparatus. Hard and fast rules can not be applied, however, and probably the tests must at present be inter- preted for the kidney as a whole. The Influence of the Nervous System on the Secretion of Urine. — In spite of numerous and repeated attempts to demonstrate that a nervous mechanism governs the excretion of urine, no proofs \yhich are above criticism have been forthcoming. Stimulation of the splanchnic nerves results in a diminution in the excretion of urine, probably because of a diminution in the blood supply of the renal vessels owing to the vasocon- striction. Stimulation of the vagus nerves below the level of the cardiac branches has been said to result in the augmentation of the rate of urine excretion (Asher and Pearce12). The results are doubtful, however, since 520 THE EXCRETION OF URINE there is no increase in the oxygen absorption under the above conditions (Pearce and Carter13). In the light of the modern theory this vagal diure- sis would be interpreted as due to an inhibition of the absorption in the tubules rather than an augmentation in the actual excretion of urine. There is no doubt that the renal nerves profoundly affect the excretion of urine, but that they do so directly is very improbable, since perfectly Si SZ Fig. 173. — Nerve supply of the kidney. K, kidney; S\, S», major and minor splanchnic nerves; V, vagus; C.G., Celiac ganglion; A, aorta. (From Cushny.) adequate renal function c.an be maintained in animals that have had the kidnej's entirely removed and then replaced. There are numerous re- flexes that affect the rate of urine excretion by constriction of the renal vessels. Injury to the bladder or ureter, abdominal injuries to the kid- ney, or even cold applied to the skin, may result in incomplete suppres- sion of the urine. CHAPTER LIX THE AMOUNT, COMPOSITION, AND CHARACTER OF URINE BY R. G. PEARCE, B.A., M.D. In the chapters on digestion and metabolism, we have followed the course which food takes with especial reference to the nutrition of the body. The excretion of these elements of nutrition is taken up under a number of the subdivisions of physiology, viz., respiration, digestion, kidney function and the skin. In the chapters on digestion attention was called to the fact that the feces, besides containing the indigestible resi- due of the aliment, contain several excretory products which at one time or another have actually been within the body proper. These in- clude normally the pigments of the body and many of the heavier mineral salts, such as iron, magnesium, lime and phosphates; and under abnormal conditions, as when the metals are given as medicine, bismuth and mer- cury. The respiratory system excretes most of the oxygen and carbon. In this chapter we shall take up the manner in which the body rids itself of the nitrogenous and some of the mineral waste materials. Even at the risk of repetition, it will be advantageous to recapitulate certain facts concerning the essential chemical structure of the urinary constituents, so that we may be in a position to appreciate the kidney function in health and disease. We now know that the kidney does not form any of the specific con- stituents of its secretion (except hippuric acid). These substances are formed in the various tissues of the body, and are brought to the kidneys by the blood, where they are eliminated. But while the constituents are .unchanged in chemical composition in the urine from that in which they are found in the blood, they do occur in greatly changed proportions. It is this variation in the concentration of the urinary constituents in the blood and the urine which presents the most important and at the same time the most difficult question in the physiology of the kidney. In the following table the percentage composition of the blood plasma is compared with that of an average sample of human urine. The third column gives the change in concentration which each constituent under- goes in passing through the renal filter. 521 522 THE EXCRETION OF URINE BLOOD PLASMA PER CENT URINE PER CENT CHANGE IN CONCENTRATION Water 90-93 95 — Proteins, fats and other colloids Dextrose 7-9 0.1 z z Urea 0.03 2 60 Uric acid 0.002 0.05 25 Creatinine Ammonia 0.001 0.04 40 Sodium 0.32 0.35 1 Potassium 0.02 0.115 7 Calcium 0.008 0.015 2 Magnesium Chlorine 0.0025 0.009 0.006 0.27 2 30 Phosphates (PO4) Sulphates (SO4) Amino acids 0.003 0.18 60 The Amount of Urine The amount of urine passed in twenty-four hours varies with the amount of fluid ingested and the proportion of fluid retained by the body or excreted by other channels. Under ordinary conditions a twenty-four- hour sample amounts to from 1000 to 1800 c.c. of urine. On a constant water intake the volume of urine is extremely variable for any single day or part of the day (Addis and Watanabe3). The average volume of urine excreted by twenty individuals on the third, fourth and fifth days of a constant diet in which the fluid intake was 2,070 c.c., varied from 1,013 to 1,712 c.c. for a twenty-four-hour period, from 684 to 1,195 c.c. for the first twelve hours of the day, and from 501 to 788 c.c. for the first eight hours of the day. In normal subjects the amount of urine excreted during the night is usually less than that during the day. This is such a constant finding that in cases where more than 50 per cent of the urine is excreted in the twelve hours of the night, suspicions of renal disease should be aroused. The Specific Gravity of Urine In urine collected at different times of the day the specific gravity may show a variation of ten points. Indeed, the specific gravity of the urine has been taken as a functional test by clinicians. With a constant food and water intake the variations found in the specific gravity of samples, of urine taken at two-hour periods in normal and pathologic conditions are very useful as criteria of the functional state of the kidney. Fixa- tion of the specific gravity at either a low or a high figure is not the usual normal finding. The following figures will illustrate: AMOUNT, COMPOSITION, AND CHARACTER OF URINE 523 DAY NIGHT 8-10 A.M. 10.12 A.M. 12-2 P.M. 2-4 P.M. 4-6 P.M. 6-8 P.M. 8-8 P.M.-A.M. Normal person In Hypertensive Nephritis In Myocardial Decompensation 1.016 1.010 1.018 1.019 1.009 1.020 1.012 1.010 1.019 1.014 1.009 1.018 1.020 1.019 0.020 1.010 1.010 1.021 1.020 1.009 1.022 (Compiled from Mosenthal's figures.) The proportion of water to total solids is often very similar in plasma and urine, but when water is taken in large quantities the urine shows much greater changes than does the blood, and the solids may sink to a very low concentration. On the other hand, when little fluid is taken or when the skin and bowel eliminate a large amount of fluid, the urine may become very concentrated without any change in the blood plasma. The total solids in urine can be determined with approximate accuracy by multiplying the last two figures of the specific gravity by the con- stant coefficient 0.233 (Haeser). The Depression of Freezing Point While the solids of the blood consist, for the most part, of proteins and colloids, those of the urine are made up of inorganic salts and small organic molecules. The molecular concentration — that is, the total number of molecules in a given quantity of fluid — is under ordinary conditions much greater in the urine than in the blood. The molecular concentra- tion may be determined by the depression of the freezing point of a fluid below that of distilled water (see page 10). Blood freezes almost con- stantly at -0.56° C., while urine may freeze at variations of temperature between -1° C. and -2.5° C. ; if very concentrated it may freeze at a temperature as low as -5° C., or if dilute the freezing point may be as high as -0.075° C. The variability of the freezing point and the specific gravity of the urine lead us to a consideration of the relationship of the urinary volume to its concentration. In the first place, the volume of water ingested is more frequently than otherwise in excess of the minimum absolutely re- quired by the body, and is subject to greater variation than the sub- stances excreted in the urine. The kidney is able to eliminate one con- stituent of the plasma which may be present in excess without involving any changes in others. For example, when salt is added to the food and excreted in the urine, the total chlorides are increased, but the amount of urine and the other constituents may remain unchanged; or, again, as may happen, excess of salt leads to an increase in the volume of the urine, but the salt concentration remains constant while that of the other urinary bodies is decreased. Similarly, although the rate of urea 524 THE EXCRETION OF URINE excretion is not demonstrably augmented by an increase in the volume of the urine, an increase in the rate of urea excretion induced by the ingestion of urea is accompanied by a larger volume of urine. That these two factors may not stand in a causal relationship to each other is sug- gested by recent work of Addis and Watanabe,3 who find no quantitative relationship between the rate of increase in urea excretion and the increase in urine volume, and who believe that the apparent relationship is due to a common cause, such as alteration in the rate of circulation or change in the activity of the kidney cells. Nevertheless, there appears to be a limit set to the power of the kidney to take the urinary salts or water from the plasma and to place them in the urine in quite different propor- tions. The definite amount of water required to hold the urinary salts has been termed the "volume obligative" (Ambardr>). These limits of concentration may be fixed by the energy which the kidney can bring to act against the osmotic resistance. The inconstancy in the behavior of the kidney toward ingested salts is probably due to the fact that the salts reach the kidney in the concen- tration in which they are held by the blood plasma, and not as they were ingested. If salt is absorbed rapidly enough to disturb the salt equilib- rium of the tissues and plasma, then water will be abstracted from the tissues, and the plasma on reaching the kidney will eliminate the salt and water together. The difference in the reaction arises from the varied activity in the tissues in general rather than in the kidney itself. The Reaction of Urine In man and the carnivora this reaction is generally acid to litmus or phenolphthaleih. The cause is found in the fact that the end products of protein metabolism give rise to sulphuric and phosphoric acids the acidity of which gives the urine an acid reaction. In the herbivorous animals the alkaline reaction is due to the fact that vegetables and fruits contain salts of dibasic or polybasic acids, such as acid potassiuiv malate, citrate, acetate, and tartrate. Oxidation of these in the bod" gives rise to carbonates. Some of the carbonic acid is excreted through the lungs, and hence the associated base, generally sodium or potassium, is combined so as to form a weak basic salt. The measurement of the acidity of the urine in terms of gram anions or cations, like the same measurement in blood, requires the use of the rather difficult electrical or indicator method, the principle of which has been described in Chapter V. Expressed in terms of CH, the acidity varies between 4.7 x 10-7 and 100 xlO"7. The total potential acidity — that is, the number of. II ions which will be formed in the face of a con- AMOUNT, COMPOSITION, AND CHARACTER OF URINE 525 tinual neutralization of those in solution — may be obtained fairly accu- rately by titrating the urine with %0 normal alkali in the presence of neutral potassium oxalate, using phenolphthalein as an indicator (Folin). The results may be expressed in acidity per cent in terms of c.c. N/10 NaOII required to neutralize 100 c.c. of urine. If the ammonia excretion is added to the titration results, the total potential acidity is very closely measured. The urine is more alkaline shortly after meals than at other times, since acid is being excreted by the gastric glands. It is more acid on a meat than on a vegetable diet, and is acid during starvation because protein is then the chief metabolite. In disease there is no characteristic variation, save that the urine is more generally acid, which may be ex- plained by the fact that in serious illness the diet is restricted. When the acidity is increased, the excretion of ammonia is usually greater, since ammonium carbonate, the forerunner of urea, acts as an alkali and neutralizes the acid radicles. This rise in ammonia, however, is not always proportional to the acid radicles present, since the fixed alkali derived from fruits and vegetables may be sufficient to neutralize the acid formed. THE SOLID CONSTITUENTS For practical reasons we shall divide the constituents of the urine into normal and abnormal. The former are present in the average urine in amounts sufficient to be detected by ordinary means; the latter only rarely appear in detectable quantities. In a person eating an ordinary diet the most important organic and inorganic constituents of the urine are as follows: TOTAL SOLIDS (40 TO 60 GRAMS) IN ONE LITER OF NORMAL URINE ORGANIC CONSTITUENTS, "25-40 GM. INORGANIC CONSTITUENTS, 15-25 GM. Urea, 20-35 gm. Sodium chloride (NaCl), 8-15 gm. Creatinine, 1.011.5 gm. Phosphoric acid (P2OS), 2.5-3.5 gm. Uric acid, 0.5-1.25 gm. - Sulphuric acid, (SO3), 2-2.5 gm. Hippuric acid, 0.1-1.7 gm. Potassium (K,O), 2-3 gm. Other constituents (ethereal sulphates, Sodium (Na2O), 4-6 gm. oxalic acid, urinary pigments, etc.), Calcium (CaO), 0.1-0.3 gm. 1.5-2.3 gm. Magnesium (MgO), 0.2-0.5. Ammonia (NH3), 0.3-1.2 gm. Iron (in pigme"nt), 0.001-0.010. (Compiled from Mosenthal's* figures.) These urinary salts are present in the blood, and are excreted only by the kidney. An investigation of the mechanism of renal secretion must therefore include a study of the relationship existing between the con- centration of the urinary salts in the blood and in the urine. 526 THE EXCRETION OF UEINE The Normal Organic Salts of the Urine Nitrogenous Constituents.— The greater number of the organic salts of the urine are made up of bodies which contain nitrogen, and which are derived from the protein element of nutrition. The proteins, which form the chief building material of the body, are broken up into their con- stituent amino acids in the intestinal tract and absorbed as such by the blood. Portions of these acids are taken up by the tissues to repair and to replace those proteins which have been discarded, and the remaining protein, in excess of the body need for amino acids, is deamidized, the major portion of the carbon, oxygen and hydrogen being oxidized to form C02 and water, and the lesser portion of these elements being com- bined with the nitrogen to form urea, ammonia, uric acid, etc. A similar fate later awaits the nitrogen moiety which found a place in the tissues, and which is replaced in turn by new nitrogenous bodies.* Since all the ingested nitrogen, except a small and rather constant amount which is lost by the feces and the sweat, is excreted in the urine, the total nitrogen of the urine has been taken as a measure of the nitro- gen or protein metabolism of the body. In normal conditions the protein metabolism is adjusted in such a manner that the nitrogen intake is equal to the nitrogen output, a condition known as nitrogenous equilib- rium. If the nitrogen intake is reduced below the actual body needs, the excretion of nitrogen is greater than the intake which indicates that the body protein is replacing the protein usually furnished by the food. The minimum amount of protein that the body must have to maintain equilibrium varies in individuals, but is on the average between 5 and 6 grams of nitrogen a day, which corresponds to about 40 grams of pro- tein. With the ordinary diet it is usually between 12 and 20 grams a day, or represents from 75 to 125 grams of protein. Since protein is not stored by the body except in periods of growth or after periods of undsr- nutrition, an increase in the protein food is accompanied by an increase in the nitrogen excreted in the urine. For this reason, unless the amount of nitrogen ingested is known, the study of the total nitrogen of the urine gives no information concerning the nature of the nitrogen metabolism of the body. The total output of nitrogen per day usually amounts to 10 to 15 grams — from 1 to 2 per cent of the urine by weight. All the nitrogenous bodies of the urine are normally nonprotein, and arise from similar bodies in the blood, where they exist in concentra- tions of from 20 to 30 mg. per 100 c.c. In excreting the nitrogen of the urine the kidney therefore takes it from a solution in which it is found in a concentration of 0.03 per cent on the average and delivers it to a *For further details see page 610. AMOUNT, COMPOSITION, AND CHARACTER OF URINE 527 solution containing an average of 1.00, or concentrates it at least 30 times. Urea. — The chief of the nitrogenous bodies of the urine is urea, the origin of which has been fully described in the chapters on metabolism. No constituent of the urine is subject to greater variation both in abso- lute and in relative amounts. On an average diet containing 120 grams of protein per day, the absolute urea excretion may amount to about 30 grams ; on a low protein diet it may be only a few grams. When the pro- tein intake is high, the nitrogen eliminated as urea may be 90 per cent of the total nitrogen; but when the protein intake is low, this proportion may fall to 60 per cent. The difference is because on a low protein diet the greater percentage of nitrogen eliminated is endogenous in origin, and urea, which is the chief constituent of the exogenous nitrogen moiety of the urine, is accordingly decreased on low diets. In recent years the importance of the relationship between the con- centration of the urinary constituents in the blood and the urine has been much insisted upon, and since the estimation of the amount of urea in the blood and the urine is relatively simple, most of the work has been done by using these values. Ambard and Weil5 believe that a quantitative relationship exists between .the rate of urine excretion and the concentration of urea in the blood and the urine, since the urea in the blood acts as a stimulus to the renal cells. By comparing the rate of urea excretion and the concentration of urea iri the blood and urine in a mathematical formula, they have obtained a value which they be- lieve is more or less fixed for the normal kidney. This expression is known as Ambard' s coefficient and formula* and has been used as a means of evaluating the functional capacity of the kidney. Whatever the value of the formula may be in expressing the relationship existing between the rate of urea excretion and the concentration of this salt in the blood, it is certain that, in diseased conditions where impair- ment of the kidney is certain, the concentration of urea in the blood re- mains permanently at an abnormally high average level, although the *Ambard and Weil's formula is: Ur K = , in which: 70 vT" D x — x — — P V2S K = coefficient of urea excretion (Constant of Ambard). Ur = grams of urea per liter of blood. D — output of urea in grams per 24 hours. P == weight of the patient. C = grams of- urea per liter of urine. 70 = standard weight. 25 = standard concentration of the urine. The average value for this constant in normal individuals is said to lie between .06 and .09. - Critical reviews of the work have been published recently by Maclean6 and by Addis and Watanabe.* 528 THE EXCRETION OF URINE amount of urea excreted during twenty-four hours may be exactly the same as under normal conditions. Probably the increased concentration of urea in the blood under these conditions is a compensatory measure to provide sufficient pressure to cause its excretion through a damaged outlet. It is this increase in urea of the blood which is indicated by the term iirea retention in nephritis. It must not be lost sight of, however, that the approximate constancy of the combined formula is due in large part to the mathematical con- struction, and also to the fact that any increase in the concentration of urea in the blood is usually accompanied by an increased rate of urea excretion. The factors which are most variable occur as the square or the square roots of their values, and thus the disturbing effect they pro- duce on the constancy of the resultant of the formula is greatly re- duced, while the most constant factor, the concentration of urea in the blood, is used with modification. In such a complex mechanism as the renal function it is very probable that other factors are of great im- portance in controlling the rate of urinary excretion. . Many of these factors can not admit of mathematical expression. The writer seriously doubts the advisability of adopting an empirical formula as a means of expressing .unknown physiologic laws. Such measures are apt to give a sense of knowledge altogether false, and thus hinder research progress. The upper limit of blood urea-nitrogen is about 20 mg. per 100 c.c.. which would correspond .to about 0.45 gm. of urea per liter of blood. The average figure is half of this amount. The maximum concentration of urea in the urine is seldom over 8 per cent. On this basis the kidney can raise the concentration of the urea in the urine, at a conservative estimate, from 100 to 200 times. Normally the daily output of urea nitrogen may range from 8 to 12 gm., and the nitrogen which it contains is roughly 80 per cent of the total excretion for the day. Ammonia. — The chief source of ammonia in the body is from the ni- trogenous portion of the deamidized amino acids. The ammonia found in excess in the portal blood is derived from ingested ammonium salts and from ammonia resulting from bacterial action on proteins in the intestinal tract. The ammonia of the body is present chiefly in the form of ammonium carbonate, and it is this salt that is the precursor of urea. Because ammonium carbonate is so readily converted into urea by the tissues of the body, little ammonia is normally present in the systemic blood. The greater portion of the ammonia that finds its way into the urine serves as a base to transfer acid radicles either ingested or formed within the body. The amount of ammonia in the urine, therefore, is an indirect measure of the extent of urea formation and of the acid bodies AMOUNT, COMPOSITION, AND CHARACTER OF URINE 529 of the blood. For the latter reason the determination of the ammonia excretion in urine is of some clinical importance. The ingestion of mineral acids increases the ammonia excretion, while alkalies tend to reduce it. During fasting and in diseases such as diabetes, where there is an abnormal metabolism, the amount of ammonia in the urine is in- creased. Ordinarily the daily output of ammonia nitrogen does not exceed 0.5-0.6 gm., constituting 3-5 per cent of the total amount of nitrogen. Creatinine. — On a meat-free diet the daily excretion of creatinine is remarkably constant, amounting to from 7 to 11 mg. per kilogram of body weight. For this reason its determination is accepted as an in- dispensable feature in metabolism investigations involving urine an- alysis. Any gross variation from the normal amount indicates the certain failure of the attendants to collect all of the twenty-four-hour specimen of urine. Normally the blood contains from 1 to 2 mg. per 100 c.c. The creatinine is one of the last of the urinary constituents to accumu- late in the blood during renal insufficiency, and for this reason affords a reliable prognostic indication concerning the patients' condition. A rise in the creatinine concentration of the blood is evidence of serious renal disease, patients with concentrations of 5 mg. never recovering (Chase and Meyers)7 The concentration of creatinine in the urine is about 100 times greater than in the blood. In adult man creatine does not appear in the urine save during starva- tion or wasting diseases. In woman it is absent save after postpartum resolution of the uterus. Children commonly excrete creatine along with creatinine until the middle years of childhood. The Purine Bodies and Uric Acid. — The most important purine in human urine is uric acid. Xanthine is the next in importance, and small amounts of hypoxanthine, guanine, and adenine are found. Among the most interesting of the salts of the urine to the clinician are the urates, because an accumulation of uric acid in the body was believed to be responsible for many obscure clinical conditions. It is quite true that the salts of uric acid are found in higher than normal amount in some diseases, especially gout, leukemia, and chronic nephritis, but the many vague theories associated with uric acid and disease have long ago been exploded. The human body has the almost unique distinction among mammals of not being able to destroy any of the uric acid it produces, and hence all the uric acid formed during metabolism must be excreted in the urine. Unfortunately the kidney appears to be less competent to rid the body of this waste than it is of the other urinary metabolites, and one of the 530 THE EXCRETION OF URINE earliest signs of renal insufficiency is now held to be a failure of the kidney to prevent the uric acid of the blood from increasing. Perhaps the reason for the inability of the kidney to excrete uric acid readily lies in the fact that its salts are among the least soluble of those in the urine. It is on this account that when the urine cools, a red sediment of urates containing certain pigments often separates out. The uric acid of the urine is possibly derived entirely from the purine metabolism of the body, in which the nucleins either of the body cells or of the exogenous food take part. It is decreased during starvation and increased by eating food rich in nucleins, such as liver and sweet- breads. Under ordinary conditions the excretion of uric acid amounts to from 0.3 to 1.2 gm. per day (0.02 to 0.10 per cent), the variation being de- pendent upon the state of health, diet, or personal idiosyncrasy. The blood of a normal individual contains on the average 1.8 mg. of uric acid per 100 c.c. The kidneys are therefore able to concentrate the uric acid in the urine from 30 to 60 times over its concentration in the blood plasma. The purines found in coffee and tea (caffeine, etc.) are excreted in the urine as salts not of uric acid but of methylated xanthines. Hippuric Acid. — This is a constant constituent of the urine of her- bivorous animals, and is usually present in small amounts in human urine. The amount rarely exceeds 0.7 gm. a day, but on a diet rich in fruits and vegetables it may exceed 2 gm. It is interesting, since it is the only urinary constituent that is synthesized by the renal cells. Amino acids are always present in small amounts in the urine, con- stituting, according to D. D. Van Slyke, about 1.5 per cent of the total nitrogen. The estimation of the amino-acid nitrogen of the urine has not been found to be of any clinical significance.8 The aromatic oxyacids are normally present in the urine in varying amounts. These include phenol, indoxyl, skatoxyl, and phenylacetic, paraoxyphenyl, propionic, oxymandelic and homogentisic acids. These bodies are derived from phenylamino acids, such as tyrosine, tryptophane, and phenylalanine. It is believed that the putrefactive decomposition of proteins in the large intestine results in the production of these toxic bodies. The body protects itself by oxidizing them and uniting them to sulphuric acid to form the ethereal or conjugated sulphates, which are found in the urine in the form of sodium or potassium salts. The determination of the amounts of these bodies in the urine has therefore been taken as an index of the putrefaction going on within the bowel. The chief of these bodies is urinary indican, which is found usually as a potassium salt. The test for indican in the urine consists in oxidiz- AMOUNT, COMPOSITION, AND CHARACTER OF URINE 531 ing the indoxyl in an acid solution by means of ferric chloride to indigo blue, and shaking out the indigo blue with chloroform. The depth of the color of the chloroform affords a rough means of determining the amount of indican present. The fact that the indican test is nega- tive must not be taken to mean that the intestinal processes are normal, for if the intestine fails to contain phenylated amino acids, or the proper bacteria are not present, no indican will be found. On the other hand, the putrefactive process of the large bowel may not be very extensive, yet the amount of indican in the urine be increased, because of greater absorption due to constipation. Skatole, a fecal-smelling substance, is formed by certain kinds of bac- teria. The greater proportion of this substance is excreted by the bowel, but if the person is constipated, some of it may find its way into the blood to impart a fecal odor to the breath and urine. Its presence therefore has some diagnostic importance. A very interesting body which is sometimes found in the urine is homogentisic acid. It is thought to be an intermediate step in the metab- olism of tyrosine, and is found in the urine of people suffering from alkaptonuria. The disease is remarkable in that it appears to run in families and produces no ill effects. Homogentisic acid is a strong reducing agent, and for this reason may be confused with sugar in Fehling's test. The inorganic constituents of the urine include the acids: chlorides, sulphates and phosphates; and the bases: sodium, potassium, magnesium, and calcium. The Acids of the Urine. — The chlorides compose the bulk of the acid radicles in the urine. Although they appear to be necessary constituents of the living cell, they do not, so far as known, enter into combinations with the organic constituents. The tissues appear to require a rather definite concentration of sodium chloride in order to carry on their work, for reduction in the sodium-chloride intake of the body results in a reduction in the chloride excretion by the urine. In salt starvation the chlorides may disappear entirely from the urine, the amount of chloride excreted appearing to be closely related to the amount of salt ingested. When the intake is constant, the rate of excretion is likewise more or less constant, but a sudden reduction in the salt of the diet may be accompanied by a slight decrease in the salt content of the blood, with an attendant loss of water. On the other hand, when the salt is again taken, there is a retention of salt and of water, with a consequent increase in body weight, until equilibrium is re-established on the old level. While the above is the usual reaction, a considerable retention of salt without an increase in the \vater content of the body may occur in 532 THE EXCRETION OF URINE some apparently normal cases. This is due probably to the deposition of salt in the tissues. Careful studies fail to confirm the idea that there is a fixed relation- ship between the salt and the water of the body. As with the nitroge- nous constituents, however, there appears to be a relationship between the rate of excretion of chlorides and the amount of chloride in the blood. Ambard believes that this relationship, like that of the excretion of urea to the blood urea, is capable of being expressed mathematically (see page 527), if allowance is made for the fact -that NaCl is not excreted after it falls beloAV a certain concentration in the blood equal to about 5.62 gm. per 1000 c.c. This level is more or less constant for normal individuals, but is considerably increased in disease of the kidney. This is known as the threshold of chloride excretion. The amount of sodium chloride excreted in the urine in twenty-four hours varies between 8 and 20 gm. a day, according to the intake. It is therefore apparent that the kidney is able to concentrate the salts of the plasma from ten to twenty times. The Sulphates. — Since the inorganic sulphates do not form an im- portant constituent of the food, the greater portion of the sulphates of the urine are derived from the sulphur found in the protein, molecule. For this reason the sulphates of the urine, like the nitrogen, are a meas- ure of protein metabolism. An increase in the nitrogen excretion is accompanied by an increase in the sulphur excretion, the ratio being about 5 to 1. The daily output of sulphur is between 1 and 3 gm. The greatest output is in the form of the alkaline sulphates, about 10 pe cent in combination with aromatic bodies, and a small amount in com- bination with amino acids and neutral organic salts. The phosphates of the urine are derived from the food and from the oxidation of phosphorus-containing bodies in the tissues such as nuclein, lecithin, etc. The daily excretion varies between 1 and 5 gm., calculated as P203. When calcium or magnesium is present in the food, they are excreted by the bowel as phosphate, and proportionately less is found in the urine. The amount usually excreted in the feces equals about 30 per cent of the total. Since phosphates in the urine exist as a mixture of the mono- and di- sodium hydrogen phosphates, - they have an important bearing on the reaction of the urine, the amount of each varying with the degree of the acidity of the urine. On a heavy protein diet the urine is acid on account of the sulphuric and other acids formed from the meat, and in this case there is a greater amount of phosphoric acid and the mono-sodium hydrogen phosphate. When the urine is alkaline or less acid, as it is on a vegetable diet, there AMOUNT, COMPOSITION, AND CHARACTER OF URINE 533 is a large amount of the disodium hydrogen phosphate. Since calcium and magnesium phosphates are more soluble than the diphosphates of the same metals, deposits of the earthy phosphates are often found in neutral or alkaline urines. When the urine is heated, the diphosphate of calcium breaks up into the mono-calcium and a tri-calcium phos- phate, Avhich accounts for the fine turbidity often taken for albumin in the flame test. Addition of acid will cause this to disappear. The crys- tals of triple phosphates which occur in alkaline urine are ammonium magnesium phosphate, NH4MgP04. KIDNEY REFERENCES (Monographs) Beddard, A. P. : Recent Advances in Physiology, Longmans, Green & Co., London, 1906. Cushuy, A. R.: Secretion of Urine, Longmans, Green & Co., London, 1917. (Original Papers) JBrodie, T. G., and Mackenzie, J. J.: Proc. Roy. Soc., 1914, Ixxxvii, B, 593. sCuslmy, A. R.: Secretion of Urine, 1917, p. 48. 3Addis and Watanabe: Jour. Biol. Chem., 1916, xxiv, 203. JMosenthal, II. O.: Arch. Int. Mod., 1915, xvi, 733. 5Ambard .and Weil: Physiologic normale et pathologique des reins, Paris, 1914, J. B. Bailliere et fils. eMaclean, F. C. : Jour. Exper. Med., 1915, xxii, 212. "Chase and Meyers: Jour. Am. Med. Assn., 1916, Ixvii, 931. s\ran Slyke, D. D., and Meyer, G. M.: Jour. Biol. Chem., 1912, xii, 399; and 1913, xvi, 197, 213 and 231. oKnowlton, F. P.: Jour. Physiol., 1911, xliii, 219. loBarerof t, J., and Straub, H. : Jour. Physiol., 1910, xli, 145. uRowntree and Geraghty: Jour. Pharm. and Exper. Therap., 1910, i, 579. i2Asher and Pearce, R, G.: Zeitschr. f. Biol., 1913, Ixiii, 83. i-Pearce, R. G., and Carter, E. P.: Am. Jour. Physiol., 1915, xxxviii, 350. PART VII METABOLISM CHAPTER LX METABOLISM Introductory. — The object of digestion, as we have seen, is to render the food capable of absorption into the circulatory fluids — -the blood and lymph. The absorbed food products are then transported to the various organs and tissues of the body, where they may be either used at once or stored away against future requirements. After being used, certain substances are produced from the foods as waste products, and these pass back into the blood to be carried to the organs of excretion, by which they are expelled from the body. By comparison of the amount of these ex- cretory products with that of the constituents of food, we can tell how much of the latter has been retained in the body, or lost from it. This constitutes the subject of general metabolism. On the other hand, we may direct our attention, not to the balance between intake and output, but to the chemical changes through which each of the foodstuffs must pass be- tween absorption and excretion. This is the subject of special metabolism. In the. one case we content ourselves with a comparison of the raw. ma- terial acquired and the finished product produced by the animal factory; in the other we seek to learn something of the particular changes to which each crude product is subjected before it can be used for the purpose of driving the machinery of life or of repairing the worn-out parts of the body. In drawing up a balance sheet of general metabolism, we must select for comparison substances that are common to both intake and output. In general the intake comprises, besides oxygen, the proteins, fats and car- bohydrates; and the output, carbon dioxide, water and the various nitrog- enous constituents of urine. This dissimilarity in chemical structure be- tween the substances ingested and those excreted limits us, in balancing the one against the other, to a comparison of the smallest fragments into which each can be broken by chemical agencies. These are the elements, and of them carbon and nitrogen are the only ones which it is possible to measure 534 METABOLISM 535 with accuracy in both intake and output. From balance sheets of intake and output of carbon and nitrogen and from information obtained by ob- serving the ratio between the amounts of oxygen consumed by the animal and of carbonic acid excreted; we can draw far-reaching conclusions re- garding the relative amounts of protein, fat and carbohydrate that have been involved in the metabolism. As has already been stated, the essential nature of the metabolic proc- ess in animals is one of oxidation — that is, one by which large unstable molecules are broken down to those that are simple and stable. Dur- ing this process of catabolism, as it is called, the potential energy locked away in the large molecules becomes liberated as actual or kinetic energy — that is, as movement and heat. It therefore becomes of importance to compare the actual energy which an animal expends in a given time with the energy which has meanwhile been rendered available by metabolism. "We shall first of all consider this so-called energy balance and then pro- ceed to examine somewhat more in detail the material balance of the body. ENERGY BALANCE The unit of energy is the large calorie (written C.), which is the amount of heat required to raise the temperature of one kilogram of water through one degree (Centigrade) of temperature.* We can determine the calorie value by allowing a measured quantity of a substance to burn in com- pressed oxygen in a steel bomb placed in a known volume of water at a certain temperature. Whenever combustion is completed, we find out through how many degrees the temperature of the water has become raised and multiply this by the volume of water in liters. Measured in such a calorimeter, as this apparatus is called, it has been found that the number of calories liberated by burning one gram of each of the proxi- mate principles of food is as follows : Carbohydrates ....................... | Sugar ....................... 4.0 Protein .................................... 5-0 Fat ........................................ 9.3 The same number of calories will be liberated at whatever rate the com- bustion proceeds, provided it results in the same end products. When a substance, such as sugar or fat, is burned in the presence of oxygen, it yields carbon dioxide and water, which are also the end products of the metabolism of these foodstuffs in the animal body ; therefore, when a gram of sugar or fat is quickly burned in a calorimeter, it releases the same *The distinction between a calorie and a degree of temperature must be clearly understood. The former expresses quantity of actual heat energy; the latter merely tells us the intensity at which the heat energy is being given out. 536 METABOLISM amount of energy as when it is slowly oxidized in the animal body. But the case is different for proteins, because these yield less completely oxi- dized end products in the animal body than they yield when burned in oxygen; so that, to ascertain the physiologic energy value of protein, we must deduct from its physical heat value the physical heat value of the incompletely oxidized end products of its metabolism. It is obvious that we can compute the total available energy of our diet by multiplying the quantity of each foodstuff by its calorie value. Methods. — In order to measure the energy that is actually liberated Fig. 174. — Respiration calorimeter of the Russell Sage Institute of Pathology, Bellevue Hospital, New York. At the right is seen the table with the absorption tubes; and in the middle, at the back, the electric control table for regulating the temperature of the double walls of the calorimeter. At the extreme left is the oxygen cylinder. (Lusk's Science of Nutrition.) in the animal body, we must also use a calorimeter, but of somewhat dif- ferent construction from that used by the chemist, for we have to provide for long-continued 'observations and for an uninterrupted supply of oxy- gen to the animal. Animal calorimeters are also usually provided with means for the measurement of the amounts of carbon dioxide (and water) discharged and of oxygen absorbed by the animal during the observation. Such respiration calorimeters have been made for air sorts of animals, the most perfect for use on man having been constructed in America (see Fig. 174). As illustrating the extreme accuracy of even the largest of these, METABOLISM 537 it is interesting to note that the actual heat given out when a definite amount of alcohol or ether is burned in one of them exactly corresponds to the amount as measured by the smaller bomb-calorimeter. All of the energy liberated in the body does not, however, take the form of heat. A variable amount appears as mechanical work, so that to measure in calories all of the energy that an animal expends, cue must add to the actual cal- ories given out, the calorie equivalent of the muscular work which has been performed by the animal during the period of observation. This can be measured by means of an ergometer, a calorie corresponding to 425 kilogram* meters of work. That it has been possible to strike an accurate balance between the intake and the output of energy of the animal body, is one of the achievements of modern experimental biology. It can be done in the case of the human animal ; thus, a man doing work on a bicycle ergometer in the Benedict calorimeter gave out as actual heat 4,833 C., and did work equalling 602 C., giving a total of 5,435 C. By drawing up a balance sheet of his intake and output of food material during this period, it was found that the man had consumed an amount capable of yielding 5,459 C., which may be considered as exactly balancing the actual output, It would be out of place to give a full description of the respiration calorimeter here. The general construction will be seen from the accom- panying figure of the form of apparatus in use for patients in the Russell Sage Institute, New York. One of the most interesting details of its con- struction concerns the means taken to prevent any loss of heat from the calorimeter to the surrounding air. This is accomplished in the following way: The innermost layer of the wall is of copper; then, separated from this by an air space, is another wall of copper, outside of which are two wooden walls separated from each other and from the outer copper walls by air spaces. The two copper walls arc connected through thermoelectric couples, so that an electric current is set up whenever there is any differ- ence in their temperatures. The current is observed by means of a gal- vanometer placed outside the calorimeter, and from its movements the ob- server either heats up or cools down the outer copper walls so as to cor- rect the difference of temperature causing the current. This is done by an electric heating device or by cold water tubes placed between the outer- most copper and the innermost wooden walls. Since the temperature of the two copper walls is the same, there can be no exchange of heat between them, and consequently none of the heat that is absorbed by thfe inner cop- per walls is allowed to be carried away. All the heat given out by the animal is absorbed by the stream of cold water flowing through the coils *A kilogram meter is the product of the load in kilograms multiplied by the distance in meters through which it is lifted. 538 METABOLISM of pipe in the chamber. The heat used to vaporize the moisture from skin and lungs must of course also be measured. This is done by collect- ing the water vapor in a sulphuric-acid bottle placed in the ventilat- ing current. By multiplying the grams of water by the factor for the latent heat of vaporization, we obtain the calories of heat so eliminated. "The calorimeter contains a comfortable bed and is provided with two windows, a shelf, a telephone, a fan, a light, and a Bowles stethoscope for counting the pulse. The ordinary experiment takes about as long as a trip from New York to New London. Patients, as a rule, doze from time to time or else try to work out some scheme by which they can amuse them- selves without moving. After three or four hours they are rather bored by the quiet, and the observations are not prolonged beyond this time. They are allowed to turn over m bed once or twice an hour, but reading and telephoning are discouraged, since these increase the metabolism. The air in the box is fresh and pure, the patient suffers no discomfort, and objections to the procedure are very infrequent. Most of the patients are only too glad of the extra attention, and they insist that the calor- imeter has a marked therapeutic value." (Du Bois.) Normal Values. — Having thus satisfied ourselves as to the extreme accuracy of the method for measuring energy output, we shall now con- sider gome of the conditions that control it. To study these we must first of all determine the basal ~h.eat production — that is, the smallest energy output that is compatible with health. This is ascertained by allowing a man to sleep in the calorimeter and then measuring his calorie output while he is still resting in bed in the morning, fifteen hours after the last meal. When the results thus obtained on a number of individuals are calculated so as to represent the calorie output per kilogram of body weight in each case, it will be found that 1 C. per kilo per hour is discharged — that is to say, the total energy expenditure in 24 hours in a man of 70 kilos, which is a good average weight, will be 70X24 = 1,680 C. When food is taken the heat production rises, the increase over the basal heat production amounting for an ordinary diet to about 10 per cent. Besides being the ultimate source of all the body heat, food is there- fore a direct stimulant of heat production. This specific dynamic action, as it is called, is not, however, the same for all groups of foodstuffs, being greatest for proteins and least for carbohydrates. Thus, if a starving animal kept at 33° C. is given protein with a calorie value which is equal to the calorie output during starvation, the calorie output will increase by 30 per cent, whereas with carbohydrates it will increase by only 6 per cent. Evidently, then, protein liberates much free heat during its as- similation in the animal body; it burns with a hotter flame than fats or carbohydrates, although before it is completely burned it may not yield METABOLISM 539 so much energy as is the case, for example, when fats are burned. This peculiar property of proteins accounts for their well-known, heating qual- ities. It explains why protein composes so large a proportion of the diet of peoples living in cold regions, and why it is cut down in the diet of those who dwell near the tropics. Individuals maintained on a low protein diet may suffer intensely from cold. If we add to the basal heat production of 1,680 C. another 168 C. (or 10 per cent) on account of food, the total 1,848 C. nevertheless falls far short of that which we know must be liberated when we calculate the available energy of the diet, which we may take as 2,500 C. What be- comes of the extra fuel? The answer is that it is used for muscular work. Thus it has been found that if the observed person, instead of lying down in the calorimeter, is made to sit in a chair, the heat production is raised by 8 per cent, or if he performs such movements as would be necessary for ordinary work (writing at a desk) it may rise 29 per cent — that is to say, to 90 C. per hour. There is, however, practically no difference in the en- ergy output of a person lying flat or lying in a semi-reclining posi- tion, as in a steamer chair. Allowing eight hours for sleep and sixteen hours for work, we can account for about 2,168 C., the remaining 300 odd C. that are required to bring the total to that which we know, from statis- tical tables of the diets of such workers, to be the actual daily expenditure, being due to the exercise of walking. If the exercise is more strenuous, still more calories will be expended ; thus, to ascend a hill of 1,650 feet at the rate of 2.7 miles an hour requires 407 extra calories. Field workers may expend, in 24 hours, almost twice as many calories as those engaged in sedentary occupations. Standard for Comparison When the energy output per kilo body weight is determined in animals of varying size, the values are greater the lighter the animal. This is evident from the following results obtained on dogs: Weight of dog Heat production in calories per Tctio per day (1) 31.2 35.68 (2) 18.2 46.2 (3) 9.6 65.16 (4) 0.5 66.07 (5) 3.19 88.07 (Rubner) When, on the other hand, instead of body weight, the area of the sur- face of the body is taken as the basis of calculation, results that are almost constant are obtained. Following are the results in the above animals on this basis : 540 METABOLISM Surface in. square cm. (1) 10,750 (2) 7,662 (3) 5,286 (4) 3,724 (5) 2,423 Ileat production in calories per square meter of sur- face per day 1036 1097 1183 1153 1212 (Rubner) Such results have prompted observers to conclude that the determining factor in the calorie output of warm-blooded animals is the relative sur- face of the animal. This is greater the smaller the animal, with the con- sequence that heat is more rapidly lost to the surrounding air from the surface, thus requiring more active combustion. Until quite recently it has been generally believed that such a relationship between body surface and heat production did actually exist, but, thanks to the work of F. G. Bene- dict7 and E. F. and D. Du Bois6, it is now known that the calculations were 2 200 190 180 CO £l70 LJ SI60 §150 o £l40 §130 ^\20 no inn 0 30 40 50 60 70 80 90 100 I 0 200 190 180 170 160 150 140 130 120 110 inn 16 \ \ \ X \ NO x \2l X s22 X 53 N n 5 \ \ \ \ \ N8 NB \ \ \ \ \ ^\ ^^ — K U \ \ \ \ \ \ \ \l S X X X X \ X X ^ -\ \ \ \ \ \ \ \ \ \ N k X ^ X > ^ £ \ \ \ \ \ \ \ \ X x \ \ \. 19 kU \ \ \ \ \! \ \ \ \ X s^ \ > TB, -!<* } \ \ \ \ \ \ ss \ X > > \ \ \ \ k X s, X i 4 15 ft* a 5 \ V. X N x i.; \ \ \ \ ^ 1.1 20 30 40 50 60 70 80 WEIGHT-KILOGRAMS 90 100 110 Fig. 175. — Chart for determining surface area of man in square meters from weight in kilo- grams (Wt.) and height in centimeters (Ht.) according to the formula: Area (Sq. Cm.) := Wt. 0.425 XHt. 0.725 X71.84. (From Dubois and Dubois, Arch. Int. Med., 1917, vol. 17.) based upon incorrect computations of the body surface. In the older re- searches the calculation was made by using a formula known as Meeh 's, in which weight was multiplied by a certain factor (viz., 12.312 x T^Aveight). Du Bois, however, has shown that an average error of 16 per cent is in- curred in using this formula. For accurate measurement the body was covered with thin underwear, which was then impregnated with melted paraffin and reinforced with paper strips to prevent it from changing in area when removed. This model of the surface was afterwards cut up into flat pieces and photographed on paper of uniform thickness, the pat- METABOLISM 541 terns being then cut out, and weighed. From the results it was easy to calculate the actual surface area. Where the height and weight are known, a fairly accurate computation of the surface can be secured by using the following formulas : A="W°-425 XH°-725X71.84; A being the surface area in square centimeters; H the height in centimeters; and W, the weight in kilograms. Based on this formula, a chart has been plotted from which the surface area may be de- termined at a glance (Fig. 175). Another method recently employed by Benedict is based on measurements made from photographs of the subject in various poses. By the use of these more accurate measurements of body surface, it is now known that, although the surface-area law gives us constant results for the energy output of different individuals of similar build, and offers us a much more accurate basis for comparing those of different laboratory animals, yet it breaks down when applied to men in widely differing states of body nutrition. Thus, in the case of a man who starved for a month, the calorie output per square meter of surface decreased towards the end of the fast by 28 per cent. Obviously, therefore, it would be incorrect to draw conclusions regarding possible changes in energy output of a series of emaciated or corpulent individuals by comparison of their calorie output per square meter of surface with that of normal individuals. The determining factor of energy output is undoubtedly the general condition of bodily nutrition — the active mass of protoplasm of the body (Benedict). That there is a relationship between the body surface and metabolism is undoubted, but the relationship is not -a causal one. At present, therefore, the only safe method to employ in comparing the metabolism of normal and diseased individuals is that called by Benedict "the group method," in which the metabolism of groups of persons of like height and weight is compared, it being assumed that such individuals have the same general growth relations. For the application of this group method, however, more extensive data will be required than exist at pres- ent, and although some of the conclusions drawn from results computed on the surface-area basis may have to be revised, it is probable that they are in general correct. Influence of Age and Sex The energy output is low in the newly born ; it increases rapidly during \ the first year, reaching a maximum at about three to six years of age, and then rapidly declining to about twenty, after which' it declines much more slowly. The decline in the earlier years does not proceed steadily, how- ever, for at the period just preceding the onset of puberty a decided in- crease becomes evident, indicating that at this period the metabolism of 542 METABOLISM the growing organism is being stimulated. Females have a lower energy output than males, and the stimulating influence of puberty is less marked in them. In round numbers, 40 C. per square meter of surface per hour is the energy output of normal men, a 1.5 per cent deviation being considered as decidedly abnormal. The average metabolism of fat and thin subjects is the same, but that of women is 6.8 per cent lower than that of men. The basal metabolism of a group of men and women between the ages of forty and fifty was 4.3 per cent below the average for the larger group between the ages of twenty and fifty; and that of a group between fifty and sixty years was 11.3 per cent lower. Influence of Diseases The measurements have been made by the direct method which has just been described, but since the much simpler indirect method (page 554) yields comparable results, it is being adopted for clinical purposes. These results were obtained by making parallel determinations of energy out- put by both methods, in disease as well as in health. Some of the ob- servations that have been made on the energy output in various diseases are as follows: In very severe cases of exophthalmic goiter, heat produc- tion may be increased by 75 per cent over the normal ; in severe cases, by 50 per cent. The warmth of the skin and the sweating, which are promi- nent symptoms of this disease, are therefore accounted for by the in- creased elimination of heat, and it is considered possible that the other symptoms would be produced in any normal individual were his metabo- lism maintained for months or years at the high level which it occupies in goiter. In the opposite condition of myxedema, the energy output is markedly reduced, but rises slowly during treatment with thyroid extract, or much more rapidly with the very active thyroid hormone recently iso- lated by Kendall. In diabetes it has often been thought that the rapid emaciation and loss of strength were dependent upon an excited state of metabolism, or a useless burning up of the energy material. The most recent work, however, clearly shows that this is not the case, the basal metabolism as calculated per unit of body surface being within, the limits indicated above. During the starvation treatment the energv output may be much below the normal. In uncompensated cases of cardiorenal dis- ease, there is increased energy output. In pernicious anemia the metabo- lism is normal, although in severe cases there may be an increased demand for oxygen. Even at the risk of repetition, it is important to point out that in all these diseases the energy output is the same whether measured directly or by the indirect method about to be described. METABOLISM 543 THE MATERIAL BALANCE OF THE BODY We must distinguish between the balances of the organic and the in- organic foodstuffs. From a study of the former we shall gain information regarding the sources of the energy production whose behavior under various conditions we have just studied. From a study of the inorganic balance, although we shall learn nothing regarding energy exchange — for such substances can yield no energy — we shall become acquainted with several facts of extreme importance in the maintenance of nutrition and growth. To draw up a 'balance sheet of organic intake and output requires an accurate chemical analysis of the food and of the excreta (urine and ex- pired air). Methods for Measuring Output The principle by which the output is measured will be understood by referring to Fig. 176, from which it will be seen that the calorimeter is connected with a closed system of tubes provided with an air-tight ro- Water to absorb ha at Hll ~ta- Chamber for antmol window Fig. 176. — Diagram of Atwater-Benedict respiration calorimeter. As the animal uses up the O2, the total volume of air shrinks. This shrinkage is indicated by the meter, and a corresponding amount of O2 is delivered from the weighed O2-cylinder. The increase in weight of bottles II and III gives the CO2; that of" I, the water vapor. tary blower or pump to maintain a constant current of air, as indicated by the arrows. Following the air stream as it leaves the chamber, we note a side tube connecting with a meter to indicate changes in volume of the 544 METABOLISM air in the system. Beyond this and the pump is a specially constructed bottle containing concentrated H2S04, then one containing soda lime, and lastly another H2S04 bottle. The first H2S04 bottle absorbs all the water vapor contained in the air coming from the chamber; the soda lime bottle absorbs the C02, and the second H2S04 bottle absorbs water that is pro- duced in the chemical reaction involved in the absorption of the C02 by the soda lime (2NaOH+C02=H20+Na2C08). By weighing these ab- sorption bottles before and after an animal has been for some time in the chamber, the weight of H20 and of C02 given out can be determined. An- other side tube leads to an oxygen cylinder, the valve of which is manip- ulated so as to cause oxygen to be discharged into the system at such a rate as to compensate exactly for that used up by the animal, as indicated by the behavior of the meter. The amount of oxygen required is de- termined either by weighing the oxygen cylinder before and after the ob- servation or by measuring the volume of oxygen used by passing it through a carefully calibrated and very sensitive water meter inserted on the side tube that connects the 02 cylinder with the main tubing of the system. Since muscular activity causes pronounced changes in the rate of me- tabolism, means are usually taken to secure graphic records of any move- ments made during the observation. The growing importance in clinical investigations of measurements of the respiratory exchange and the necessity for having methods that are as simple as is consistent with accuracy, have led to the introduction of several other forms of apparatus, of which those of F. G. Benedict and of Tissot* are the most important. In the former a tightly fitting mask, applied over the nose and mouth is connected, by a short T-piece, with the same tubing as that used in the respiration calorimeter. The patient thus breathes in and out of the air stream that is passing along the tubing without any of the obstruction experienced when the breathing has to be performed through valves, as in the older (Zuntz) forms of portable respiratory apparatus. It is particularly for studies on man that this apparatus has been devised. The Tissot and Douglas methods are shown in Figs. 179 and 180.* To complete the investigation, it is necessary that the urine and feces be collected and the nitrogen excretion measured. "When the respiratory excreta are measured over a considerable period of time, as in the large calorimeter, the urine is collected for the same period, but when shorter respiratory measurements are made, the urine of the twenty-four hours is usually taken. Principles Involved in Calculating1 the Results. — Provided with the an- alyses furnished by the above methods, we proceed to ascertain the total "The Tissot method will be found described in full elsewhere (page 554). METABOLISM 545 amounts of nitrogen and carbon excreted and to calculate from the known composition of protein how much protein must have undergone metabolism. We then compute how much carbon this quantity of pro- tein would account for, and we deduct this from the total carbon excre- tion. The remainder of carbon must have come from the metabolism of fats and carbohydrates, and although we can not tell exactly 'which, yet we can arrive at a close approximation by observing the respiratory quo- tient (R. Q.), which is the ratio of the volume of carbon dioxide exhaled CO to that of oxygen retained by the body in a given time, i. e., n 2 . By ob- == «W«|_=o.82 547 548 METABOLISM 4. Conversion of fat into carbohydrate: 2G,H6(018HJS01), + 6402 = 16C6H12O6 + 18CO2 + 8H2O (Olein.) 5. Conversion of carbohydrate into a mixed fat: 13C6H12O6 = C55H104O6 + 23CO2 + 26H2O. ( Oleostearopalmit in. ) Taking carbohydrates first, the general formula may be written CH20, from which it is plain that, to oxidize the molecule, oxygen will be re- quired to combine with the carbon alone, according to the equation, CH20 + 02 = C02 + H20. In other words, the volume of carbon dioxide pro- duced by the combustion will be exactly equal to the volume of oxygen used in this process, in obedience to the well-known gas law that equi- molecular quantities of different gases occupy the same volume. The respiratory quotient is therefore unity (Equation 1). With fats and pro- teins, however, the general formula must be written CH2-f-0, indicating therefore that for its complete oxidation the molecule must be supplied with oxygen in sufficient amount to combine not only with all of the car- bon, but also with some of the hydrogen, forming water; so that the vol- ume of C02 produced will be less than the volume of oxygen retained, and the respiratory quotient will be less than unity. As a matter of fact, as the above equations show (2 and 3), the respiratory quotient for fats and proteins lies somewhere between 0.7 and 0.8, being usually nearer 0.7 in the case of fats, and nearer to 0.8 in the case of proteins. That the conditions hypothecated in the equations exist in the animal body during the combustion of the foodstuffs can easily be shown by ob- serving the respiratory quotient of animals on different diets. An her- bivorous animal, such as a rabbit, when it is well fed gives invariably a respiratory quotient of about 1, whereas a strictly carnivorous animal, such as the cat, gives a respiratory quotient of about 0.7. Even more striking perhaps is the comparison of the respiratory quotients in an herbivorous animal while it is well fed and after it has been starved for a day or two. In the latter case the respiratory quotient will fall to a low level because, by starvation, the animal has been compelled to change its combustion material from the carbohydrate of its food to the protein and fat of its own tissues. As already explained (page 545), it is from the respiratory quotient that we are enabled to tell what proportions of fat and carbohydrate, respectively, are undergoing metabolism. A useful table showing the percentage of calories produced by each of these foodstuffs, after allow- ing for protein, is given by Graham Lusk (see page 565). THE CARBON BALANCE 549 Influence of Metabolism. — Apart from diet, the respiratory quotient may often be altered by changes in the metabolic habits of the animal. These are most conspicuously exhibited in the case of hibernating animals. In the autumn months, when the animal is eating voraciously of all kinds of carbohydrate food and depositing large quantities of adipose tissue in his body, the respiratory quotient may be considerably greater than unity, indicating therefore either that relatively more carbon dioxide is being discharged or less oxygen retained. As a matter of fact, it can easily be shown that it is the former of the causes that is responsible for the higher quotient, the explanation for the increased production of C02 being that, as the carbohydrate changes into fat, the relative excess of carbon in the former is got rid of as C02, as indicated in Equation 5. On the other hand, if the animal is examined while in his winter sleep, it will be found that the respiratory quotient is now extremely low, often not more than 0.3 to 0.4, which may be interpreted as indicating either an excessive absorption of oxygen or a markedly decreased excretion of carbon dioxide. As a matter of fact, there is a great diminution in both the excretion of carbon dioxide and the intake of 02, because the whole metabolic activity of the animal is extremely depressed, but this diminution affects the oxygen to a much less degree, indicating therefore a relative increase in the oxygen retention. The explanation is that the oxygen is being used in the chemical process in- volved in the conversion of the- fat back into carbohydrate. Whatever may be the relationship between fat and carbohydrate in the nonhibernating animal, there is no doubt that during hiberna- tion, before the fat stores are burned, fat is converted into something closely related to carbohydrates, the equation for the process being rep- resented as given above (No. 4). In man and the higher mammalia, the only condition, apart from diet which can affect the nature of the combustion process is disease; thus in total diabetes (page 678) the organism loses, the power of burning carbohydrate, so that whatever the diet may be, the respiratory quotient is very low, never higher than that representing combustion of fat and protein. It has been claimed by certain investigators that in diabetes the respiratory quotient may fall considerably below 0.7, indicating, as in hibernating animals, that fat is being converted into carbohydrate. The most recent and carefully controlled observations, however, deny this claim, and for the present we must assume that in the body of man fat is not converted into carbohydrate (see page 664). In numerous other diseases investigated by.Du Bois and others6 no qualitative change in the combustion processes in man has been brought to light. 550 METABOLISM THE MAGNITUDE OF THE RESPIRATORY EXCHANGE It is evident that the amount of carbon dioxide expired and of oxy- gen retained will be proportional to the energy liberation in the animal body. Even at the risk of repetition it should be noted that the energy exchange can be very accurately calculated from the result of the material balance sheet — indirect calorimetry, as it is called (page 562). On account of the comparative simplicity of measuring the carbon dioxide output and oxygen intake, it is natural that many of the obser- vations that have been made on energy production in the animal body depend on the use of this method, justification for which is found in the complete agreement between the results of direct and indirect calorim- etry in a great variety of diseases and conditions in man (Du Boisr>).* In the first place, it is interesting to compare the respiratory ex- changes of different animals computed per kilo body weight. This is shown in the following table. ANIMAL WEIGHT GM. OXYGEN AB- SORBED PER KILO AND HOUR GM. CARBON DIOXIDE DISCHARGED PER KILO AND HOUR GM. VOL. C02 TEMPERA- TURE OF AIR VOL. 02 Insecta Field cricket 0.25 2.305 — — Amphibia Edible frog 0.0(5:? 0.060 0.69 15°-19° (44.2 c.c.) (30.76 c.c.) 0.105 0.1134 0.78 — (73.4 c.c.) (57.7 c.c.) Aves Common hen 1280 1.058 1.327 0.91 19° . 900 ocn (740 c.c.) (675 c.c.) q oq<* A igeon Sparrow ^O^-OoU 22 9.595 O.43O 10.492 0.79 18° (6710 c.c.) (5334.5 c.c.) "Mammalia Ox 6,18,000 0.389-0.485 — — 660,000 Sheep 66,000 0.490 0.671 0.99 16° (343 c.c.) (341 c.c.) Dog 6213 1.303 1.325 0.74 15° (911 c.c.) (674 c.c.) Cat 2464 1.356 1.397 0.75 -3 2° 3047 (947 c.c.) (710 c.c.) > > 0.645 0.766 0.86 29.6° (450 c.c.) (389 c.c.) Rabbit 1433 1.012 1.354 0.97 18°-20° Guinea pig 444.9 1.478 1.758 0.86 22° Rat (white) 80.5 3.518 7° (1789 c.c.) ATousp ^ ^ «?PC "8.4 17° Man CHJ 66,70 0.292 0.327 — (Modified from Pembrey.)17 *For the convenience of those who may desire to know more about the methods of analysis that are suitable in the clinic, a chapter on the subject will be found beginning on page 554. THE CARBON BALANCE 551 Several factors operate to explain these differences, and of these the following are of importance: 1. The Body Temperature. — Increase in body temperature entails in- creased combustion. This explains why the metabolism of a bird is greater than that of a mammal of the same size, for, as is well known, the temperature of a bird is two or three degrees centigrade above that of other animals. Rise in body temperature also explains, in part at least, the increased metabolism observed in fever. 2. The Temperature of the Environment. — In considering this we must distinguish between the effect produced on warm-blooded and on cold- blooded animals. Since the body temperature of a cold-blooded animal is only one or two degrees Centigrade above that of its environment, it follows that the metabolic activity will be directly proportional to the temperature of the latter. In a warm-blooded animal, on the other hand, the body temperature remains constant Avhatever changes may occur in that of the environment, this constancy of body temperature being dependent on the fact that the intensity of the combustion processes is inversely proportional to the cooling effect of the atmosphere. Thus, suppose the external temperature should fall, then the loss of heat from the body will tend to become greater, and to maintain the body tempera- ture at a constant level, the body furnaces must burn more briskly, with the result that an increased excretion of carbon dioxide and intake of oxygen will occur. This influence of the surrounding atmosphere on the metabolic activ- ity of w-arm-blooded animals has, as already pointed out, been used by several investigators to explain the greater combustion per kilo body weight of small as compared with large animals. The argument is that, since the surface of small animals relatively to their mass is much greater than in large animals, the cooling of the small animals will be proportion- ately greater. The relationship between surface and mass is shown by tak- ing two cubes and putting them together; the mass of the two cubes is equal to double that of either cube, whereas the surface is less than double, since two aspects of the cubes have been brought together. To prove the contention, the respiratory exchange has been computed per square meter of surface instead of per kilo body weight, with the result that a very close correspondence in the metabolism of different animals has been observed ; but this question has already been discussed, and we now know that the law of cooling can not be the only one that determines extent of the respiratory exchange (see page 541). 3. Muscular Exercise. — This has a most important influence on the ex- change and it is particularly in connection with it that studies in carbon- dioxide output and oxygen intake have been of great practical value, par- 552 METABOLISM ticularly when the investigations are undertaken on men doing ordinary types of muscular exercise, such as walking or climbing. It is true that the influence of muscular exercise on the energy metabolism may also be studied by having a person in the calorimeter do exercises on an ergometer, but the results thus obtained are in many ways not nearly so valuable as those which can be secured by observing the respiratory exchange of persons doing ordinary types of muscular exercise in the open. The folloAving table of observations on horses is of interest in this connection. CONDITION AIR EXPIRED CARBON DIOXIDE OXYGEN ABSORBED C02 IN LITERS DISCHARGED IN IN LITERS PER 02 PER MINUTE LITERS PER MINUTE MINUTE Best 44 1.478 1.601 0.92 Walk 177 4.342 4.766 0.90 Trot 333 7.516 8.093 0.93 It will be observed that the metabolism increases extraordinarily for even a moderate degree of work, but that at the same time the respiratory quotient remains constant. From observations on the respiratory ex- change of working men and animals, extremely important facts concern- ing the efficiency of muscular work have been secured. The form of respiratory apparatus (Zuntz or Douglas) employed for this purpose must be capable of being strapped on the man's back without causing any embarrassment to his bodily movements. By a comparison of the respiratory exchange with the amount of work done, the efficiency of the work can readily be determined. It has been found, for example, that the efficiency is much greater after the man or animal has got into the swing of the work, his energy expenditure per unit of work being much greater during the first half hour's work in the morning than it is later on. This indicates that after a little practice the muscles can ex- ecute a given movement and perform a given amount of work much more smoothly than when they are not in training. Another interesting outcome of the investigations has been to show that work done under ab- normal conditions that tend to produce any kind of muscular strain is done inefficiently. It has been found in marching soldiers, for example, that the slightest abrasion of the foot greatly increases the energy expenditure, for the man, in trying to avoid the pain produced by the abrasion, brings into operation muscular groups that are really not required for the efficient performance of the movement, but are used instead to avoid pressure on the sore. Fatigue also causes inefficient performance of work; that is to say, the fatigued person, on attempting THE CARBON BALANCE 553 the same amount of work as he performed before becoming fatigued, will do so at a much greater expenditure of energy. There is a diurnal variation in the respiratory exchange, which is in general parallel with the body temperature ; it rises during the day, the time of activity and work, and falls during the night, the time of rest and sleep. Food also affects respiratory exchange, but it will be unnec- essary to go into this further after what has been said on page 547. CHAPTER LX1I* A CLINICAL METHOD FOR DETERMINING THE RESPIRATORY EXCHANGE IN MAN BY R. G. PEARCE, B.A., M.D. Principle. — Since the determination of the respiratory exchange in man is of some importance in the study of certain diseases of the respira- tion, circulation and metabolism, and also because directions for carry- ing out the necessary procedures are not generally available, we have thought it might be of assistance to include here brief directions for the Tissot and the Douglas methods. These methods have been found to compare favorably in accuracy with others in use at present, t and be- cause of their adaptability and simplicity they are specially suited for clinical work. By these methods the energy metabolism of the body is calculated from oxygen consumption or carbon dioxide excretion per minute (indirect calorimetry) (page 546), the figures for which are determined from the volume and percentile gaseous composition of the expired air. The subject breathes through valves which automatically partition the inspired and expired air. The expirations from a number of respirations are collected in a spirometer or bag, and the volume of the respirations per minute is determined. The gaseous composition of the expired air is determined by gas analysis, and the oxygen consumption and energy output of the body are calculated from the data obtained. Description and Use of Parts of the Apparatus: 1. THE MOUTHPIECE AND VALVES. — The mouthpiece is made of soft pure gum rubber, and con- sists of an elliptical rubber flange having a hole in the center 2 cm. in diameter, to which on one side a short rubber tube is attached. On the opposite side of the hole, at right angles to the rubber flange, are at- tached two rubber lugs. The rubber flange is placed between the lips, and the lugs are held by the teeth. The rubber tube of the mouthpiece is connected to the tube carrying the valves. The nose must be tightly closed if mouth breathing is used. This is accomplished by a nose clip, which consists of a V-shaped metal spring, the ends of which are pro- vided with felt pads. A toothed rachet is attached to the ends of the *This chapter is added for the convenience of workers in this subject. tCarpenter: Carnegie Institution of Washington Reports, No. 216, 1915. 554 METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 555 spring, and serves to hold the spring tightly clamped on the nostrils in the proper position (see Fig. 177). Some individuals experience great distress when made to breathe through the mouth. For these it is best to use a face mask. Unfortu- nately at the present time no mask is entirely satisfactory. Perhaps the best is one sold by Siebe, Gorman & Co.,* which is pictured in the cut. Fig. 177. — A, Nose clip; B, Face mask; C, Mouth piece. After being placed in position the face mask should be tested for leaks, which can be done by putting soap around the edges. 2. THE VALVES. — The valves of Tissot are probably the best for the purpose, but they are expensive and difficult to obtain. We have made perfectly satisfactory valves from the prepared casings used in the manufacture of bologna sausage. These can be obtained preserved in salt, and they will keep indefinitely on ice. When needed a short piece *This mask has been used extensively by Carpenter. The agent in this country is II. X. Klmer, 1140 Monadnock Bldg., Chicago. 556 METABOLISM is taken, washed free from salt by allowing water from the tap to run through it, and softened in a weak glycerine solution. The gut becomes very soft and pliable, and does not dry quickly. A piece of the casing about 10 cm. long is threaded through a glass tube of about 15 mm. bore and 4 to 6 cm. long. One end of the casing is brought around the outside of the tubing and secured by means of a thread. The lower end of the membrane is pinched off and the casing is then cut a little more than half way across its middle, so that the opening will lie just within the free end of the tube when the casing is drawn back through it. The loose end of the casing is slightly twisted — an essential procedure — and is then secured by a thread on the outer side of the tube. If properly made, the valve will work freely without vibration, and the opening be sufficiently large to allow a good current of air to pass. It should col- lapse instantly and be air-tight when the current of air is reversed. The back lash, or lag of closure, of these valves is extremely small, and they will open or close with a pressure of air not exceeding the pressure Fig. 178. — Diagram of respiratory valves. changes in normal respiration. When not in use, the valves should be kept in glycerine water on ice. Valves prepared in this way have been in use a month without loss of efficiency. They are, however, made with so great ease that new valves are provided for each subject, and they are therefore especially adapted to ward work (Fig. 178). The valves are inserted in reverse order into a supporting metal T-piece, and the joints made air-tight by tape. The stem of the T is connected with the mouthpiece. Through a rubber tube of about 3/4 inch bore, the expired air is collected in the spirometer, or Douglas Bag. 3. THE TISSOT SPIROMETER is pictured in Fig. 179. We have found the 100-liter size to be very serviceable in the clinic. This instrument is mounted on a platform having rubber wheels, and can be moved about the wards with ease. The bell of the spirometer is made of aluminum and is suspended in a water-bath between the double walls of a hollow cylinder made of galvanized iron. The height of the bell is 72 cm. and the diameter 42 cm. An opening at the bottom of the cylinder connects through a three-way stopcock with the rubber tube leading from the expiratory valve of the mouthpiece (see Fig. 177). METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 557 The bell is counterpoised by means of a weight. In the original Tissot spirometer an automatic adjustment permitted water in amount equal to the w,ater displaced by the bell to flow from the spirometer cylinder into a counterpoise cylinder as the bell ascended out of the water. Fig. 179. — The Tissot spirometer. In actual experiment, subject is reclining or lying down and the valves and mouthpiece are held with a clamp. The bell, being heavier out of water than when it is immersed, is accord- ingly counterpoised in any position, although Carpenter has shown that this refinement is unnecessary. An opening in the top of the spirometer permits the insertion of a rubber stopper, through which are passed a thermometer, a water manometer, and a stopcock with tube for drawing 558 METABOLISM the sample of air. A scale on the side of the instrument gives the vol- ume of the air. During an observation the subject sits in a reclining position or lies upon a couch. When the bell of the spirometer is placed at zero, the mouthpiece adjusted in the mouth, and the nose clamped, respiration is started, the expirations being passed through the stopcock, which is so turned as to allow them to pass to the outside air. After a few minutes the stopcock is turned so that the expirations are passed into Fig. 180. — The Douglas bag method for determining the respiratory exchange. The arrange- ment of mouthpiece, valves, and connecting tubes shown here has been found to be more con venient than that recommended by Douglas. the spirometer for a definite length of time. At the end of the period the cock is again turned, and after the barometric pressure, temperature, and volume of the air have been noted, the composition of the air is determined in the Haldane gas analysis apparatus. 4. THE DOUGLAS BAG. — The Douglas bag is made of rubber-lined cloth, and is capable of holding from 50 to 100 liters. It is especially useful for investigations during exercise, since it is fitted with straps so that the bag can be fastened to the shoulders (Fig. 180). It is then connected with the A7alves, the mouthpiece of which is placed between the lips. METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 559 Respirations are commenced with the three-way valve turned so as to allow the expirations to pass directly outside. After respiratory equi- librium is established, the three-way valve is turned during an inspira- tory period so that the succeeding expirations may pass into the bag. The time required to fill the bag comfortably is determined with a stop- watch. The air which has been collected in the bag during the period is thoroughly mixed and passed through a meter, the temperature and barometric pressure are noted, and a sample analyzed in the Haldane A. B. Fig. 181. — Haldane gas apparatus {A) and Pearce sampling tube (B). gas-apparatus. The bag should be emptied completely by rolling it up when nearly empty. 5. The Haldane Gas-analysis Apparatus. PRINCIPLE. — The Haldane method of analysis of expired air is simple and easily learned. The ap- paratus (Fig. 181) consists of a gas burette, a control burette of the same size (both surrounded with a water jacket), and bulbs containing dilute caustic potash or soda solution for the absorption of the carbon di- oxide and an alkaline pyrogallate solution for the absorption of the 560 METABOLISM oxygen. The gas burette is connected with the bulbs by a two-way stopcock, which allows a sample of gas to pass into either bulb. A con- trol tube (10) is put into connection with the burette through a manometer tube, which is connected with the alkali bulb, and can be made to com- pensate for any changes in temperature that may occur during the course of the analysis. For an analysis the gas is transferred to the burette from the sampling tube, saturated with water vapor over mercury, and then measured, after which it is transferred into the caustic solution to free it from C02, and returned to the burette to determine the loss of volume due to C02 absorption. It is then transferred into the alkaline pyrogallate solution, which frees it from oxygen, after which it is again brought back to the burette to determine the loss in volume due to the absorption of the oxygen. THE APPARATUS. — The detail of the Haldane apparatus is shown in the accompanying cut. The measuring burette (1) holds 21 c.c. The bulb is of 15 c.c. capacity, and the graduated stem, which is about 4 mm. in bore and 60 cm. in length, is graduated to 0.01 c.c. from 15 c.c. to 21 c.c. The stopccck at the top of the burette is double-bored, so that in one position air can be drawn in from a gas sampler (2) and in another sent into the absorption bulbs (3). The lower part of the burette ex- tends through the rubber cork at the bottom of the water jacket (4). A piece of rubber tubing is attached to the bottom of the burette and is passed through a metal tube, furnished- on its inside with a metal disc which presses against the rubber tubing, the pressure being controlled by means of a fine adjusting screw (6). Below this a glass stopcock (7) con- nects with rubber tubing to the mercury leveling bulb (5). The absorption bulb for C02, containing 20 per cent NaOH or KOH (9), is put in con- nection with the burette by suitably turning stopcocks (3 and 8).* The control burette (10) is also in connection with this bulb through the manometer tube (11). t Any variation in temperature which may occur during the analysis will cause the level of the alkaline solution in the manometer to change. When final readings of the shrinkage of volume are made, the level of the caustic solution is returned to the level of that in the manometer. By so doing any error due to temperature changes is avoided, since change in temperature must be equal in the two burettes. The absorption bulb for oxygen (12) is filled with a solution made by dissolving 10 grams of pyogallic acid in 100 c.c. of a nearly saturated KOH solution. The specific gravity of the KOH should be 1.55, which is obtained approximately by dissolving the sticks (pure by alcohol) in an *The. stopcock (8) is double-bored, so that the tube leading from the burette can be brought into connection with either 9 or 12. tThis tube also has a three-way stopcock (19), so that it may be opened to the outside. METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 561 equal weight of water. The mark (13) on the stem of the bulb indi- cates the level at which the solutions should stand. Enough pyrogallate solution is introduced through tube 15 to fill bulbs 12 and 14 two-thirds full. Then pyrogallate solution is poured into tube 16 until the* differ- ence in level of the fluids is sufficient to produce enough pressure to raise the level of the pyrogallate solution in 12 to the level 13 on the stem. Stopcock 8 must be open during this procedure. It may be neces- sary to add or take away a little pyrogallate solution through 15 to at- tain the above level. Care must be taken to allow for complete absorption of oxygen from the air that is entrapped between 14 and 16 before an analysis is made ; otherwise changes will be produced in the level of the pyrogallate solu- tion. The air in the capillary tubing connecting the burettes with the absorption bulbs must also be freed of C02 and 02. This can be accom- plished by making a dummy analysis of atmospheric air before the real analysis. Great care must be taken to have atmospheric pressure in all the tubes at the start of the analysis. This is accomplished by opening the stopcock in the burette first to atmospheric air and then to the ab- sorption bulbs, until -no further change in the level of the fluids in the stems of the absorption bulbs occurs. This level is then marked and used as the standard. A small amount of water in the burette over the mercury assures saturation of the air with water vapor. Time for drain- age must be allowed before making readings. A very serviceable sampling tube for the transfer of air can be made from a 30 c.c. ground-glass syringe, to which is attached a two-way stopcock. A cut of this is shown in Fig. 181. The dead space in these syringes is washed out by working the piston back and forth several times. A thin coating of vaseline prevents leakage of the gas. We have found that these sampling tubes will retain a sample of expired air with- out change up to eight hours. MANIPULATION OF APPARATUS. — The sampling syringe (20) is attached to opening 2 of the burette, and its stopcock ( 17) opened to atmospheric air. The level of the mercury is raised to the level of the stopcock of the syringe and is then turned so that syringe and burette are in communication. The bulb of mercury is lowered so that the mercury falls in the burette. This draws the piston of the syringe with it, and fills the burette with air from the syringe. It is advisable to put a little positive pressure on the piston of the syringe in the maneuver to prevent possible leakage. When all of the air is in the burette a slight positive pressure is produced in the burette by gently pressing on the piston, and immediately there- after the stopcock on the syringe (17) is again turned to the original position. This allows the pressure of air in the burette to come to that 562 METABOLISM of the atmosphere. The height of the mercury is now adjusted to a con- venient height in the burette by closing cock 7 and turning the milled screw 6. The cock 18 is now made to communicate with the absorption bulbs. -If the air in the burette is at atmospheric pressure, no change will occur in the level of the .fluids. The reading is then taken on the burette. The next step in the analysis consists in turning stopcock 8 to com- municate with the caustic soda solution in bulb 9, and the leveling tube (5) is raised, forcing mercury into the burette and the air into bulb 9. The gas is passed back and forth several times until absorption is com- plete, as can be determined by the fact that the level of the mercury in the burette remains constant when the fluid in the bulb is returned to its original level (13} on the stem. In this adjustment it is convenient to make the gross leveling by the mercury bulb and the fine leveling by closing 7 and turning 6 until the fluid in 9 is at the original height. The reading on the burette indicates the loss in volume due to the C02 absorbed. The oxygen is removed by a similar procedure, the gas being passed into the alkaline pyrogallate solution by turning cock 8 to communicate with bulb 12. The absorption of oxygen is slower than for C02, and more care must be taken to get complete absorption. The air in the tubing between the fluid in 9 and stopcock 8 must be washed out sev- eral times in order to get the oxygen which is left in it after the absorp- tion of the C02. When this is complete, the final reading on the burette is made and the loss in volume from the second reading represents the oxygen. THE CALCULATIONS The calculation of the percentile composition of the air and of the re- spiratory quotient is represented in the following example of an actual analysis: (The temperature and barometric pressure as taken at the time of the experiment were 20° C. and 747 mm. Hg.) CO, analysis — 1st reading of burette 20.00 2nd reading of burette after absorption of CO, 19.20 CO2 absorbed 0.80 0.80 -4- 20 = 4.0 per cent CO2 in expired air. 02 analysis — 2nd reading of burette 19.20 3rd reading of burette after absorption of O2 15.90 O2 absorbed 3.30 3.30 -=- 20 = 16.50 per cent of 02 in expired air. METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 563 Determination of E.Q. — O2 in atmospheric air — 20.94% O2 - CO2 in expired air (1(5.50 + 4) = 20.50% 100-20.94 = 79.06%," N in atmospheric air. 100-20.50 = 79.50%, N in expired air. Since the nitrogen is not changed in volume, the last figure shows that more oxygen must have been taken in during inspiration than 02 + C02 has been given back in expiration. This obviously must 'be taken into account in the calculations. The amount of 02 actually inspired for each 100 c.c. of air expired is found as follows: 20.94 (% O2 in atmospheric air) 79.06 (% N2 in atmospheric air) X 79'50 <% N' in exPired air) > or °'265 (con stant factor X 79.5 (% N found for this observation) =21.07, the volume of O2 which would have been present in expired air to account for N present. t 21.07-16.50 = 4.57% O2 actually absorbed. 4.00 - 0.03 (CO2 in inspired air) = 3.97% CO2 excreted. 3.97 ' — 0.87, the respiratory quotient, or ratio of CO, excreted to O., absorbed. Total Gas Exchange. — The volume of air expired in 15 minutes into the Tissot spirometer was found to be 100 liters measured at 20° C. and 747 mm. Hg (brass-scale barometer). This volume of gas must be cor- rected so as to give the volume of dry air at 0° and 760 mm. Hg. To do this two things must be taken into account. (1) Since the expired air is saturated with water, the pressure due to water vapor must be subtracted from the observed barometric pressure to obtain the true pressure. The vapor tension of water for various temperatures is given in Table II on page 564.. (2) The barometer tube lengthens, or contracts with heat or cold, and therefore the barometric readings must be corrected. The corrections for ordinary barometric readings are found in Table III, page 565. The figure corresponding to the temperatures is subtracted from the barometric reading in order to obtain correct barometric pres- sure. In the above experiment, the correction for the barometer is 2.41 mm. (see Table III, page 565), and that for vapor tension at 20° C. is 17.4 (see Table II, page 564). Actual Barometric Pressure.— 747 - (17.5 + 2.39) = 727.21 mm. The coefficient of expansion of gases is taken as 0.003665) or 1/273 ; therefore the volume of 0° equals the volume at 1° divided by 1-0.003665 t; and hence "This is the constant O percentage :n air. tThis calculation can be simplified by using an abbreviated table (page 564) giving the O» figure corresponding to the various percentages of N in the expired air. 564 METABOLISM Vx273 V V° = 273+T=l + 0.003665 t ' When V° = VP The volume of gas being inversely as the pressure, Vo — _ _ , where V = volume at 760 P pressure; or working both corrections together, VPx273 VP Vo_ 760 x (273 + t)~760 (1 + 0.003665 t) This formula applied to the present problem reads: Vo = 100 x 727.2 _ _ = 89.2 liters. 760 (1 + 0.003665x20) The latter calculation can be considerably simplified by using standard tables which give constants for corrections of gas volumes. These are easily obtainable and are given in part in Table IV. According to these tables for 20° C. and 727.21 mm. Hg B.P., the factor is 0.89124; therefore: 0.89124 x 100 — 89.124 liters, 0°C. and 760 mm. Hg. 0.89124x4.57 = 40.7 liters ^of O2 in 15 min., or 16.28 L. per hour. The Caloric Value Calculated from the Gas Exchange. — By reference to Table V giving the heat value of 1 liter of 02 at various respiratory quotients, it is found that at a R.Q. of 0.87, 4.888 calories are expended ; 16.28 liters of 02 is therefore equivalent to 18.4 x 4.888 = 79 calories. The results must be calculated for surface area as well as body weight. Suppose the subject weighed 85 kg. and was 170 cm. in height; by refer- ence to the chart for determining the surface area of man (page 540), this would be found to be 1.96 square meters. The caloric expenditure 79 per square meter in the above case is therefore ~~ = ^ TABLE I THE PERCENTAGE OF OXYGEN WHICH is EQUIVALENT TO THE NITROGEN FOUND IN THE EXPIRED AIR To obtain the nitrogen in the expired air, add the percentage of CO2 and O2 found and subtract the sum from 100. The table gives the percentage for O2 corresponding to this figure: 78.7 78.8 78.9 79.0 79.1 79.2 79.3 79.4 79.5 79.6 79.7 79.8 %O, 20.86 20.88 20.90 20.93 20.96 20.98 21.01 21.04 21.07 21.10 21.12 21.14 79.9 80.0 80.1 80.2 S0.3 80.4 80.5 80.6 21.16 21.19 21.22 21.25 21.28 21.31 21.35 21.38 _ TABLE II TENSION OF AQUEOUS VAPOR IN MILLIMETERS OF MERCURY To obtain the dry barometer pressure, subtract the mm. Hg. corresponding to the temperature of the air from the barometer pressure at the time of the experiment: Temp. 15° 16° 17° 18° 19° 20° 21° 22° 23° 24° 25° Mm. 12.7 13.5 14.4 15.4 16.3 17.4 18.5 19.7 20.9 22.2 23.5 METHOD FOR DETERMINING RESPIRATORY EXCHANGE IN MAN 565 TABLE III TEMPERATURE CORRECTIONS TO EEDUCE EEADINGS OF A MERCURIAL BAROMETER WITH A BRASS SCALE TO 0°C. Subtract the appropriate quantity as found in table from the height of the barometer. The table is for a barometer with a brass scale, and the values are a little lower (about .2 mm.) than for the glass scale. The corrections for intermediate temperatures can be approximated. Temp. 700 mm. 710 mm. 720 mm. 730 mm. 740 mm. 750 mm. 760 mm. 770 mm. 15° 20° 25° 1.69 2.26 2.83 1.72 2.22 2.87 1.74 2.32 2.91 1.77 2.36 2.95 1.79 2.39 2.99 1.81 2.42 3.03 1.84 . 2.45 3.07 1.86 2.48 3.11 TABLE IV TABLE FOR REDUCING GASEOUS VOLUMES TO NORMAL TEMPERATURE AND PRESSURE The observed volume, when multiplied by the factor corresponding to the temperature and pressure, will give the volume of the expired air reduced to 0° and 760 mm. Mm. 15° 16° 17° 18° 19° 20° 21° 22° 23° 24° 25° 720 730 740 750 760 770 .898 .910 .922 .935 .947 .960 .894 .907 .919 .932 .944 .957 .891 .904 .916 .928 .941 .953 .888 .901 .913 .925 .938 .950 .885 .897 .910 .922 .934 .948 .882 .894 .907 .919 .931 .945 .880 .891 .904 .916 .928 .940 .877 .888 .901 .913 .925 .936 .873 .885 .897 .910 .922 .933 .870 .882 .894 .907 .919 .930 .867 .879 .891 .904 .916 .927 R. Q. CALORIES FOR 1 LITER O2 KELATIVE CALORIES CONSUMED AS Number Carbohydrate Fat per cent per cent 0.707 4.686 0 100 0.71 ' 4.690 1.4 98.6 0.72 4.702 4.8 95.2 0.73 4.714 8.2 91.8 0.74 4.727 11.6 88.4 0.75 4.739 15.0 85.0 0.76 4.752 18.4 81.6 0.77 4.764 21.8 78.2 0.78 4.776 25.2 74.8 0.79 4.789 28.6 71.4 0.80 4.801 32.0 68.0 0.81 4.813 35.4 64.6 O.S2 4.825 38.8 61.2 0.83 4.838 42.2 57.8 0.84 4.850 45.6 54.4 0.85 4.863 49.0 51.0 0.86 4.875 52.4 47.6 0.87 4.887 55.8 44.2 0.88 4.900 59.2 40.8 0.89 4.912 62.6 37.4 0.90 4.924 66.0 34.0 0.91 4.936 69.4 30.6 0.92 4.948 72.8 27.2 0.93 4.960 76.2 23.8 0.94 4.973 79.6 20.4 0.95 4.985 83.0 17.0 0.96 4.997 86.4 13.6 0.97 5.010 89.8 10.2 0.98 5.022 93.2 6.8 0.99 5.034 96.6 3.4 . 1.00 5.047 100.0 0.0 (From Lusk.) CHAPTER LXIII STARVATION In order to furnish us with a standard with which we may compare other conditions, we shall first of all study the metabolism during starva- tion. A valuable chart compiled from observations made in the Carne- gie Institution of Washington on a man who fasted for thirty-one days is reproduced in Fig. 182. The Excretion of Nitrogen. — When an animal is starved, it has to live on its own tissues, but in doing so it saves its protein, so that the excretion of nitrogen falls after a few days to a low level, the energy requirements being meanwhile supplied, so far as possible, from stored carbohydrate and fat. Although always small in comparison with fat, the stores of carbohydrate vary considerably in different animals. They are much larger in man and the herbivora than in the carnivora. Dur- ing the first few days of starvation it is common, in the herbivora, to find that the excretion of nitrogen is actually greater than it was before starvation, because the custom has become established in the metabolism of these animals of using carbohydrates as the main fuel material, so that when carbohydrates are withheld, as in starvation, proteins are used more than before and the nitrogen excretion becomes greater. We may say that the herbivorous animal has become carnivorous. The same thing may occur in man when the previous diet was largely carbohy- drate; thus, almost invariably in man the nitrogen output is larger on the third and fourth days of starvation than on the first and second. Another factor influencing the nitrogen excretion during the early days of the fast is the amount of previous intake of nitrogen; the greater this has been, the greater the excretion. By the seventh day, however, a Uniform output of nitrogen will usually be reached irrespective of the individual's protein intake. During the greater part of starvation, most of the energy required to maintain life is derived from fat, as little as possible being derived from protein. This type of metabolism lasts until all the available resources of fat have become exhausted, when a more extensive metabolism of protein sets in, with the consequence that the nitrogen excretion rises. This is really the harbinger of death — it is often called the premortal rise in nitrogen excretion. It indicates that all the ordinary fuel of the animal economy has been used up, and that it has 566 STARVATION 567 [NUTRITION LABORATORY OF THE CARNEGIE INSTITUTION OF WASHINGTON, BOSTON, MASSACHUSETTS] METABOLISM CHART OF A MAN FASTING 31 DAYS APRIL 14 -MAY 15.1913 23456789 10 II 12 13 14 15 16 17 IB 192021 22232425262725299031 OXYGEN AND CARBON f 6» V 4.00 3.90 3.80 3.70 ALVEOLAR CO, TENSION, mm.—* 3'60 38.0 3.50 36.0 3.40 34.0 3.30 32.0 3.20 30.0 3.10 81000 PRESSURE, mm. ^l'° 3'00 132 2.90 128 2.80 124 2.70 120 2.60 1 16 2.50 HEAT PER 24 HRS.CAL5. ' ' 2 Z?° 108 1700 104 1650 100 1600 BODY TEMPERATURE, *C x6 I55° 36.8 1500 36.4 1450 ' I.T2 1350 1.08 1300 1.04 1250 1.00 1200 RESPIRATORY QUOTIENT — 1 .92 0.80 RESPII1ATION RATE ^5L -75 \ 7 \ * \ / S \ \ s ^ / \ \ \ / \ M / EN X, P£« S Kl s O.F €R •1IN TE Sy \ s ^, \ / \ ? \ / \ s \ / * ^ \ \ \ / --. LV :o AR _ct i / "•• y ^ / X. — i s x^ ~[_ s - ~^ / X ^--i \ \ S s^ & \ s f >• \ ^ / \ ^ X ( \ \ s- .£* 5 ND OU E P ER LO PER Ml LIT S . __ s s \ / X / *• X ^ ' x • , XB LOC D F SYS Rt TO SU&E .1C. LY NG x \ X s ^ \ ( -~, [X ? \ \ / s FT S \ / \ / ^ is, * — ' , — - V > ~^ X, | X ... ^ _^ *f r* ' k^BOO T EM ERATURE s s 5 ^ H' \ / •--^ X H^ ^^. \ / \ | | ~~ X \ — • 1 — •^ MP IT [S D) [R 14 IRS 47 4^ \ / tAMl, X i ' ^ ' \ 9 > s* X EAT PE RK LO. / PEf X HF ^~ / S — s x RE 5PI *AT ORV 0 JOT EN f ^ ^^ ^ \ / X ^ S ^ X *. — • ' — PULSE RATE ». *-^° 70 15.0 12.5 CHLORINE (Cl). CMS. 65_H>.0 60 2.25 2.00 TOTAL NITROGEN, GMS. 15 l-75 1 2.0 1.50 1 1.0 1.25 10.0 1.00 9.0 .75 8.0 .50 PHOSPHORUS (P,0j), GMS. — — =jj5 CARBON IN URINE. GMS. 1 2.00 8-OXYBUTYHICACIO.eMS.J J7o |50 10.0 1.00 8.0 .50 6.0 .00 \ " x RE SP RA1 I0^ Fi> TE ,' 1, -— X / "~ i — X t _- v — ^^ ,' X ^^ ^ \ — - S s ** X f \ S v ^ , * "-^. / \ i X ^- s, s ~xJJILSE RATE.MORNINJ^- ^ ^ ? ' ^ ^ / \ > v TTT&EL x x 1 ^s, ' s : >UL >E W N E. K Gh ^. T (WIN Ml Ml ** ^ i s \ roi AL NIT RO ;E^ XI \ / N -— • \ ^. \ / \ \ ( ^• ce LO HI g EJ CO \ / X N, i / \ ., X / -'S, / x | .... •-• ^'* / s y 3 •IOSPHORU s( P20 X s) 1 X s* X ^^: AR 301 IN ±=; UR NE ™»i ^1 ^ = / X . — - X / -— S / / •— ^ s x yl \ / / \ ^ ^ B- X M' /I BU ^ TYI 1C A AC n ^ f \, . 7_ 5 *^ s, / s / \ ^ ,^ — ^ URIC ACIO-N, GM. '° * TOTAL SULPHUR (S), CM. ^»° '[" .60 .075 .50 .050 / \ / \ / / \ i C_, Cll -h s — ^ fs — \ / v / ^^ ^r' \ / — ^^ — ^ \ / 1 T N , — , N / X 14 15 16 17 18 19 20 21 Zl 23 24 25 26 2.7 26 Z? 30 3.1 Fig. 182. — Curve constructed from data obtained from a man who fasted for thirty-one days. The days of the fast are given along the abscissae, and the various measurements along the or- dinates. (From F. G. Benedict.) 568 METABOLISM become necessary to burn the very tissues themselves in order to obtain sufficient energy to maintain life. Working capital being all exhausted, an attempt is made to keep things going for a little longer time by liq- uidation of permanent assets. But these assets, as represented by pro- tein, are of little real value in yielding the desired energy because, as we have seen, only 4.1 calories are available against 9.3, obtainable from fats. These facts explain why during starvation a fat man excretes daily less nitrogen than a lean man, and why the fat man can stand the starva- tion for a longer time. The premortal rise is, however, not prevented by feeding oil, which would seem to indicate that death may be due not so much to the absence of fuel as to serious nutritional disturbance of es- sential organs; e. g., there may be no available material to supply the glands of internal secretion with the building stones they must have (see page 580). Not only is there this general saving of protein during starvation, but there is also a discriminate utilization of what has to be used by the different organs, according to their relative activities. This is very clearly shown by comparison of the loss of weight which each organ un- dergoes during starvation. The heart and brain, which must be active if life is to be maintained, lose only about 3 per cent of their original weight, whereas the voluntary muscles, the liver and the spleen lose 31, 54 and 67 per cent, respectively. No doubt some of this loss is to be accounted for as due to the disappearance of fat, but a sufficient remainder represents protein to make it plain that there must have been a mobilization of this substance from tissues where it was not absolutely necessary, such as the liver and voluntary muscles, to organs, such as the heart, in which energy transformation is sine qua non of life. The vital organs live at the expense of those whose functions are accessory. The energy output per square meter of body surface steadily declines. In the man examined by Benedict, it was 958 C. per square meter of surface at the end of the first twenty-four hours, but only 737 on the thirty-first day of the starvation period. The oxygen intake and carbon- dioxide output correspondingly diminish. The behavior of the nitrogenous metabolites in the urine is of par- ticular interest, the following facts being of significance: Urea nitrogen relatively falls and NH3 - N rises. For example, on the last day of feeding the percentage output of NH3 - N in relation to total nitrogen was 3.16 ; on the eighth day of the fast it was 14.88 (Cathcart).2 Acidosis is the cause. The total amount of c'reatinine and creatine shows only a slight fall, but creatinine relatively decreases and creatine increases (Cathcart). Since creatine is a substance peculiar to muscle tissue, it is possible by STARVATION 569 comparing the creatine and creatinine output with that of nitrogen to determine whether all of the nitrogen liberated by the breakdown of muscle has been excreted, or whether some has been retained either for resynthesis in the muscle itself or for use elsewhere. If the muscle breakdown as calculated from the creatine-creatinine output is greater than that calculated from the nitrogen, synthesis of the noncreatine remainder must be occurring ; whereas if the breakdown calculated from nitrogen is greater than that calculated from creatine, etc., other tis- sues than muscle must be contributory. Stored nitrogen or free nitro- gen in transit from tissue to tissue for utilization is the most likely source of such excess nitrogen. That transference of nitrogenous substances from place to place in the body in starvation is proved (1) by the constant presence of amino ni- trogen in the blood and tissues (Van Slyke) ; and (2) by the effect of copious water drinking. The latter causes a decided increase in the out- put of nitrogen, but it does not appear that the extra nitrogen is due to increased protein breakdown. It is probable, however, that in such cases there would also be an increase in endogenous protein metabolism, since the washed-out free nitrogen would have to be replaced. Excretion of Purines. — Although at first they fall somewhat, the total amount increases as the fast progresses. Perhaps the first decline is due to general using up of hypoxanthine of muscle and the later rise to the breakdown of nuclei (page 638). Excretion of Sulphur.— It is important to compare the excretion of sulphur and nitrogen. In the early days of starvation a ratio of 17 N : 1 S has been found, but later one of 14.5: 1, which is practically the same as that in muscle (i.e., 14; 1), indicating that late in fasting the main source of protein supply is muscle. Several of the changes observed during starvation can be attributed to the condition of acidosis which supervenes. The acids are derived from incomplete combustion of fat (see page 683), and are represented by ^-oxybutyric, the amount being sometimes considerable (10-15 grams a day), especially in obese individuals. The large ammonia excretion (sometimes 2 grams a day) is evidently for the purpose of neutralizing the excess of acid. Another consequence of the acidosis is the decline in the alveolar tension of C02 (page 354), and it is possible that some of the circulatory changes shown in the chart may also be dependent on it. The method of repeated fasting used for reducing obesity is quite safe if the acidosis is carefully watched. Many secondary changes also occur in the starving organism. Thus, the mobilization of fat is often responsible for a pronounced increase in the fat content of the blood (see page 698), and that of protein explains 570 METABOLISM the presence of an amount of ammo nitrogen not much below that of normal animals (viz., 4 mg. per 100 c.c. of blood). Similarly with carbohydrates, early in the condition the blood sugar becomes much lower than normal, but then remains steady. This is significant when we remember that after two or three days of starvation all of the avail- able glycogen has been used up. It indicates that carbohydrate must be essential for life, and that it is produced in starvation from proteins (see page 667). Starvation ends in death in an adult man in somewhat over four weeks but much sooner in children, because of their more active metab- olism. At the time of death the body weight may be reduced by 50 per cent. The body temperature does not change until within a few days of death, when it begins to fall, and it is undoubtedly true that if means are taken to prevent cooling of the animal at this stage, life will be prolonged. Death from starvation must be due either to a general failure of all the cells or to injury of certain organs that are essential for life. Since the loss of protein from the body as a whole may vary between 20 and 50 per cent at the time of death by starvation, it is unlikely that general failure can be its cause. If it were so, death would always occur when some fixed loss of protein had occurred. Certain organs evidently cease to perform their function, either because they are deprived of raw mate- rial for the elaboration of some substance (hormone) necessary for life, or because the organs themselves wear out from want of nourishment. NORMAL METABOLISM Apart from the practical importance of knowing something about the behavior of an animal during starvation, such knowledge is of great value in furnishing a standard with which to compare the metabolism of animals under normal conditions. Taking again the nitrogen balance as indicating the extent of protein wear and tear in the body, let us consider first of all the conditions''ilnder which equilibrium may be re- gained. It would be quite natural 'to suppose that, if an amount of pro- tein containing the same amount of nitrogen as is excreted during starvation were given to a starving animal, the intake and output of nitrogen would balance. We are led to make this assumption because we know that any business balance sheet showing an excess of expenditure over income could be met by such an adjustment. But it is a very differ- ent matter with the nitrogen balance sheet of the body; for, if we give the starving animal just enough protein to cover the nitrogen loss; we shall cause the excretion to rise to a total which is practically equal to STARVATION 571 the starvation amount plus all that we have given as food; and although by daily giving this amount of protein there may be a slight decline in the excretion, it will never come near to being the same as that of the intake. The only effect of such feeding will be to prolong life for a few days. Nitrogenous Equilibrium. — To attain equilibrium we must give an amount of protein whose nitrogen content is at least two and one-half times that of the starvation level. For a few days following the estab- lishment of this pure protein diet, the nitrogen excretion will be far in excess of the intake, but it will gradually decline until the two practically correspond. Having once gained an equilibrium, AVC may raise its level by gradually increasing the protein intake. During this progres- sive raising of the ingested protein, it will be found, at least in the car- nivora (cat and dog), that a certain amount of nitrogen is retained by the body for a day or so immediately following each increase in pro- tein intake. The excretion of nitrogen, in other words, does not immedi- ately folloAv the dietetic increase. The amount of nitrogen thus retained is too great to be accounted as a retention of disintegration products of protein ;" it must therefore be due to an actual building up of new pro- tein tissue — that is, growth of muscles. Nitrogenous equilibrium on a protein diet alone is readily attainable in the cat, and less readily in the dog. But in man and the herbivorous animals, it is impossible to give a sufficiency of protein alone to maintain equilibrium; there will always be an excess of excretion over intake. Indeed it scarcely requires any experiment to prove this, for it is self- evident when we consider that there are less than 1000 C in a pound of uncooked lean meat, and that there are few who could eat over three pounds a day, an amount, however, which would scarcely furnish all of the required calories. A person fed exclusively on flesh is therefore being partly starved, even although he may think that he is eating abundantly and be quite comfortable and active. This fact has a prac- tical application in the so-called Banting cure for obesity. Protein Sparers. — Very different results are obtained when carbohy- drates or fats are freely given with the protein to the starving animal. Nitrogen equilibrium can then be regained on very much less protein, so that we speak of fats and carbohydrates as being "protein sparers." Carbohydrates are much better protein sparers than fats; indeed they are so efficient in this regard that it is now commonly believed that car- bohydrates are essential for life, and that when the food contains no trace of carbohydrates, a part of the carbon of protein has to be con- verted into carbohydrate. This important truth is supported by evi- dence derived from other fields of investigation (e. g., the behavior of 572 METABOLISM diabetic patients, in whom the power to use carbohydrates is greatly depressed). The marked protein-sparing action of carbohydrates is il- lustrated in another way — namely, by the fact that we can greatly diminish the protein breakdown during starvation by giving carbo- hydrates. In this way we can indeed reduce the daily nitrogen excre- tion to about one-third its amount in complete starvation. Carbohy- drate starvation is said to entail a failure of the muscles to use again in their metabolism certain of the products (e. g., creatine) which result from their disintegration. At any rate it has been found that creatine is excreted in the urine when no carbohydrates are available. In the case of man living on an average diet, although the daily nitro- gen excretion is about 15 grams, it can be lowered to about 6 grams provided that in place of the protein that has been removed from the diet enough carbohydrate is given to bring the total calories up to the normal daily requirement. If an excess of carbohydrate over the energy requirements is given, the protein may be still further reduced with- out disturbing the equilibrium. It has been found that it is not the amount of carbohydrate alone that determines the ease with which the irreducible protein minimum can be reached; the kind of protein itself makes a very great difference. This has been very clearly shown by one investigator, who first of all determined his nitrogen excretion while living exclusively on starch and sugar, and who then proceeded to see how little of different kinds of protein he had to take in order to bring him- self into nitrogenous equilibrium. He found that he had to take the following amounts: 30 gm. meat protein, 31 gm. milk protein, 34 gin. rice protein, 38 gm. potato protein, 54 gm. bean protein, 76 gm. bread protein, and 102 gm. Indian-corn protein. The organism is evidently able to satisfy its protein demands much more readily with meat than with vegetable proteins. This variability in the food value of different proteins depends on their ultimate structure — that is, on the proportion and manner of linkage of the various ammo acids that go to build up the molecule. In no two proteins are these building stones, as they are called, present in exactly the same proportions, some proteins having a preponderance of one or more and an absence of others, just as in a row of houses there may be no two that are exactly alike, although for all of them the same build- ing materials were available. Albumin and globulin are the most im- portant proteins of blood and tissues, so that the food must contain the necessary units for their construction. If it fails in this regard, even to the extent of lacking only one of the units, the organism will either be unable to construct that protein, and will therefore suffer from partial starvation, or it will have to construct for itself this missing unit. It STARVATION 573 is therefore apparent that the most valuable proteins will be those that contain an array of units that can be reunited to form all the varieties of protein entering into the structure of the body proteins. Naturally, the protein which most nearly meets the requirements is meat protein, so that we are not surprised to find that less of this than of any other protein has to be taken to gain nitrogen equilibrium. The most exact information regarding the "food value" of different proteins has been secured by observations on the rate of growth of young animals. This method yields more reliable information than can be secured by studies on the nitrogenous balance, because it is not usually possible to keep up the latter observations for a sufficient period of time, or to secure an adequate number of data. During growth the building-up processes are in excess of the breaking-down, so that the effect is an increase in bulk of the tissues, thus permitting us, by the sim- ple expedient of observing the body weight, to draw conclusions as to the influence of various foodstuffs on tissue construction. CHAPTER LXIV NUTRITION AND GROWTH In the growth of animal tissues two factors are concerned, one being the property of the cell to grow, the growth factor; and the other, the availability of suitable material to grow upon, the food factor. Concern- ing the groAvth factor little is known; its variability in different species of animal, its irregularity despite proper adjustment. of the food factors, its abnormality leading to tumor formation, etc., are all well-known but apparently inexplicable facts (Mendel8). THE FOOD FACTOR OF GROWTH Our knowledge is constantly increasing concerning the food factor of growth, and many facts of extreme practical importance have been ac- cumulated in recent years. In seeking for the relationship of food to growth, we must first of all consider whether this process entails a greater expenditure of energy than is necessary for mere maintenance in adult life. Important results bearing on this question have been se- cured by observations on the basal metabolism of young children. In computing the energy supply of fasting adult animals of different sizes, it will be remembered that the smaller the animal, the greater is the energy exchange in relationship to the body weight, although when computed in relationship to body surface tolerably constant values are obtained. When the calorie output per square meter is determined in growing children, there is, as we have already seen, clear evidence of greater energy expenditure (see page 541), particularly marked in boys just before puberty. An increased energy metabolism has also been de- scribed in the case of infants, but the uncontrollable muscular activity, the psychic disturbances, etc., may explain the result. Even after dis- counting these factors, however, it is possible that there may be a cer- tain influence, depending probably on the active mass of growing proto- plasmic tissue, which stimulates the energy expenditure. The question is not yet finally settled. The Relationship of Proteins to Growth and Maintenance of Life. — Since protein constitutes the fundamental chemical basis of the cell, it is natural to devote attention in the first place to this food principle. 574 NUTRITION AND GROWTH 575 In the pioneer investigations, studies on the nitrogen balance in young animals yielded results from which it was concluded that the conditions for the disintegration of protein are less developed in young animals than in adults, so that the growing organs rapidly withdraw circulating protein and build it into tissue protein. In consideration of the accumulation of data extending over several decades, Rubner denied these conclusions, and showed that the diet of the growing infant is by no means relatively rich in protein. He con- cluded that "growth is not proportional to the quantity of protein in the diet." Important though this pioneer work may have been in the de- velopment of our present-day conception, the viewpoint of the men who carried it out was very much narrowed on account of the paucity of knowledge concerning the structure of the protein molecule. No allow- ance was made for the fact, which has recently been firmly established, that the protein molecule may vary extremely in regard to the units of which it is composed, and that the growing tissues may demand, not so much an abundance of protein as such, but rather a proper supply of all the building stones which are required for growth (Mendel). QUANTITATIVE COMPARISON OF AMINO ACIDS OBTAINED BY HYDROLYSIS OF PROTEINS* (Compiled by T. B. Osborne, 1914) t CAS3IN OVAL- BUMIN GLIADIN ZEIN EDESTIN LEGUMIN ox MUSCLE Glycocoll Alanine 0.00 1.50 0.00 2.22 0.00 2.00 0.00 13.39 3.80 3.60 0.38 2.08 4.0 8.1 Valine 7.20 2.50 3.34 1.88 6.20 ? 2.0 Lcucine 9.35 10.71 6.62 19.55 14.50 8.00 14.3 Proline 6.70 3.56 13.22 9.04 4.10 3.22 8.0 Phenylalanine Glutaminic acid 3.20 15.55 5.07 9.10 2.35 43.66 6.55 26.17 3.09 18.74 3.75 1.3.80 4.5 10.6 Aspartic acid Serine 1.39 0.50 2.20 ? 0.58 0.13 1.71 1.02 4.50 0.33 5.30 0.53 22.3 ? Tyrosine Cystine Histidine 4.50 ? 2.50 1.77 ? 1.71 1.61 0.45 1.84 3.55 ? . 0.82 2.13 1.00 2.19 3.55 ? 2.42 4.4 ? 4.5 Arginine Lvsine 3.81 5.95 4.91 3.76 2.84 0.93 1.55 0.00 14.17 1.65 10.12 4.29 11.5 7.6 Tryptophane, about Ammonia 1.50 1.61 present 1.34 1.00 5.22 0.00 3.64 present 3.28 present 1.99 present 1.07 65.49 • 48.85 85.68 88.87 82.28 57.43 102.87 *These analyses are combinations of what appear to be the best determinations of various chemists. tThe figures for the more recent analyses of gliadr'n are inserted. From the accompanying ' table giving the percentage of the various amino acids, etc., present in certain proteins, it will be evident that there are very marked variations in the units of which different proteins aro composed. If any one of these units should be essential for growth and 576 METABOLISM the organism be unable to manufacture the missing unit for itself, it is clear that growth could not .proceed- however much protein not contain- ing the necessary unit we might feed to the animal. It is an application of the law of the minimum, and is analogous with the failure of growth which has long been known to ensue when certain inorganic substances are withheld from the growing animal. A diet might be perfectly bal- anced as judged by comparison of the nitrogen intake and output, and yet if it should fail to contain even one of the essential units and the organism should be incapable of supplying this unit, then would the diet be inadequate for growth. These important facts are the outcome of modern work, and they have been established by observations on the growth of young animals fed with a ' ' basal ration ' ' to which were added mixtures of amino acids Days Each division = 20 days. Fig. 183. — Curves of growth of rats on basal rations plus the various proteins indicated. The normal curve may be taken as that with casein (I). (Adapted from Lafayette B. Mendel and T. B. Osborne.) or various proteins which differ considerably from one another in the nature of the units entering into their make-up. In such experiments the periods during which growth is observed must be prolonged, since a transient increase in weight might depend merely on repair processes occurring in tissues which had previously for some reason been brought below par. Among the most important observations have been those of Lafayette B. Mendel and T. B. Osborne8 and of McCollum and his collaborators. The animals chosen for Mendel and Osborne 's experiments were young white rats. Large batches of these animals were fed on a basal rat-ion consisting of protein-free milk (containing the inorganic salts, the sugars, traces of protein, and unknown substances having an important influence on NUTRITION AND GROWTH 577 growth — vitamines?), to which were added more carbohydrate, purified fat, and the protein whose influence on growth it was desired to study. The same diet was fed at regular intervals to a given batch of rats, and the weight of each rat was periodically taken, the observation being pro- longed until the animals grew to maturity and produced young, and these again grew to maturity, reproduced, and so on. By plotting the re- sults in curves, with the time periods along the abscissae and the average weight of the rats of each batch along the ordinates, the extent of the influence of a given diet on the curve of growth was obtained. A normal curve of growth is shown in No. 1 of Fig. 183. It was obtained from re- sults secured by adding liberal amounts of casein to the basal diet. Wt 100 Mo HO VH Each division -*o dayi Days Each division -20 days Lafayette B. Mendel and T. B. Osborne.) Similar curves were obtained with lactalbumin of milk and ovalbumin and ovovitellin of egg. Perhaps the most interesting substances capable of producing the normal curve of growth are certain of the proteins that T. B. Osborne has succeeded in separating in crystalline form from vegetable foodstuffs. These are edestin (hempseed), globulin (squash seed), excelsin (Brazil nut), glutelin (maize), globulin (cottonseed), glutein (wheat), glycinin (soy bean), cannabin (hempseed). That growth proceeds normally with any one of these proteins when fed abundantly does not, however, necessarily indicate that each con- tains in adequate proportion all of the necessary units to meet the pro- tein demands of growing tissues. In the case of casein, for example, one of the units, namely, glycocoll, which is the simplest of all the 578 METABOLISM amino acids, is entirely missing, and another, cystine, which is a sul- phur-containing amino acid, is present only in small amount. The ab- sence of glycocoll, however, is not of importance, because the organism can manufacture it for itself (see page 630). In the case of cystine, which the tissues can not manufacture themselves, the deficiency has to be made up for by feeding an excess of casein so as to cover the needs of the tissues for this amino acid. By so doing a superabundance of most of the other units will be ingested, and this superabundance will entail the destruction and excretion of the useless amino acids, a process, however, which is conducted in such a way as to permit of the utilization, by the organism, of a part of the energy which the cast-off amino acids contain (see page 667). It is, therefore, not entirely a wasteful process. When the supply of casein is limited, on the other hand, the curve of growth becomes subnormal, because an insufficient supply of cystine is thereby offered (Fig. 184). Similar results have been obtained in the case of edestin, a protein from hempseed. This contains an insufficiency of the diamino acid, lysine. Fed in abundance, edestin gave a normal curve of growth, but when fed in insufficient amount the curve failed to ascend properly, which, however, it could be made to do by adding some lysine to the edestin. There is a large group of proteins which fail to permit of any growth no matter in what amounts they may be added to the basal ration. These include: legumelin (soy bean), vignin (vetch), gliadin (wheat or rye), legumin (pea), legumin (vetch), hordein (barley), conglutin (lupine), gelatine (horn), zein (maize), phaseolin (kidney bean). The adequacy to maintain growth of any of these pure proteins varies according to the deficiency in their amino acids. In the case of gliadin of wheat or rye, glycocoll is lacking, and lysine is present only in small amount (see table). The absence of glycocoll can not, however, as we have already seen in the case of casein, explain the inadequacy of gliadin as a foodstuff for growth (Curve II in Fig. 183) . It must be the lysine that is at fault. A still more deficient protein is the zein of maize. With this as the only protein added to the basal diet, the curve of growth actually descends (Curve III of Fig. 183), thus indicating that the animal is starving and must soon succumb. The missing units in this protein are glycocoll, lysine and tryptophane (see table on page 575), and it is very signifi- cant that if the latter two amino acids are supplied along with zein, an almost normal curve of growth will result. Some improvement can even be brought about by giving tryptophane alone; that is to say, the curve assumes a horizontal x line instead of descending, indicating that, although inadequate for growth, the diet is now sufficient for the main- tenance of life. NUTRITION AND GROWTH 579 The important fact demonstrated by these experiments, is that cer- tain diets are adequate for the maintenance of life although they are inadequate for growth. In conformity with this conclusion, it was found when young white rats were fed with gliadin alone for periods of time ex- Fig. 185. — Photographs of rats of same brood on perfect diet (uppermost picture) ; on a main- tenance diet but inadequate for growth (middle picture) ; and on a diet that was inadequate both for maintenance and growth. (From Mendel and Osborne.) ceeding those in which they should have become full grown, that they remained in an ungrown stunted condition. The capacity to grow had not, however, been lost, for when the gliadin was replaced by milk, the animals resumed growth at a very great rate. The capacity to grow 580 METABOLISM had only been inhibited by the inadequate diet, and there was nothing really abnormal about the stunted animals. For example, the reproduc- tive function developed normally, as was shown in the case of a young female rat which, after being fed with gliadin as the sole protein sup- ply for 154 days, was mated and produced four young. Although the mother was still maintained on the gliadin diet, the young rats pre- sented normal growth, for they were living on the milk supplied by the mother, and this milk, because it contained either casein or some other necessary accessory factor (vide infra), was an adequate food. After removal from the mother, three of these rats were fed on an arti- ficial diet of casein, edestin and the basal ration, and continued the nor- mal course of growth, but when one of them was placed on the gliadin food mixture it immediately failed to grow properly. It would appear from these experiments that, of the two amino acids that are missing or deficient in gliadin — namely, glycocoll and lysine — it must be the lysine that is essential for growth. This very important conclusion was fully corroborated by finding that, in young rats stunted by previous gliadin feeding, growth immediately started when lysine was added to the diet and ceased again when the lysine was removed, and so on, the experi- ments being often repeated in various modifications. Mendel and. Os- borne call attention to the relatively high percentage of lysine in all those proteins that are concerned in nature with the growth of young animals; thus, it is present in large amounts in casein, lactalbumin and egg vitellin. It is particularly in protein of vegetable origin that indispensable units are likely to be missing, the best known of these units being the aromatic amino acids, tyrosine and tryptophane; the diamino acid, lysine; and the sulphur-containing acid, cystine. Some animal proteins, Such as gelatine, also fail to contain aromatic groups, and are therefore utterly inadequate as protein foods. That the absence of one or two units should render a protein utterly incapable of maintaining life suggests that a specific role may be taken by certain amino acids in the maintenance of nutritional rhythm; thus, they may be necessary for the elaboration of some hormone or other in- ternal secretion essential to life, such as epinephrine, the active principle of the suprarenal gland. This is an aromatic substance not far removed in its chemical structure from tyrosine (see page 734). It is therefore natural to suppose that the absence of the tryptophane unit in zein is the reason that this protein is incapable of maintaining the in- itial body weight. • In attacking the problem from this viewpoint, Hopkins and Willcock10 made observations on the survival period of young mice; that is, the period during which the animals survived when fed on a basal diet NUTRITION AND GROWTH 581 mixed either with zein alone or with zein plus small quantities of tryp- tophane. It was found that, with zein alone, the mice were unable to maintain growth; they lost in weight and died in from about a week to about a month. Other mice fed on the same amount of basal diet and zein, but to which was also added some tryptophane, although they did not grow, were capable of maintaining their body weight and lived in some instances for nearly a month and a half. There were other indica- tions of the difference in the efficiency of the two diets. The mice fed on the zein alone were very inactive, and remained for a considerable period of the time in a condition of torpor. The hair was ruffled, the eyes were half closed, and the ears, feet and tail were cold. The ani- mals, however, gave evidence of having a good appetite. On the other hand, the mice to which tryptophane was also given manifested a strik- ingly different behavior, being active and more or less normal until just before death. That both groups of animals failed to live more than forty-four or forty-eight days is probably to be accounted for by the absence in the zein of the other unit, lysine. Had this been added along with the tryptophane it is probable, in the light of Mendel and Osborne's observations, that the animals would have survived much longer. To supply the missing unit, besides using the pure amino acid, we may employ other proteins which contain the required amino acid (Curve III of Fig. 184). That mixtures of protein foodstuffs are desirable has long been apparent to those who have studied practical dietetics. We must com- bine the unsuitable protein with others which, although in themselves perhaps also unsuitable, yet furnish us with a mixture which contains all the essential units both for maintenance and growth. As Mendel points out, these considerations suggest that we may be able to utilize certain of the low priced protein by-products of the cereal, meat and milk in- dustries. The test of the adequacy of the corrected diet must, however, be determined by experiments of the type which we have just described. It is probably in stock-raising rather than in connection with human nu- trition that these facts will prove of practical value; for, not only is the diet of man more varied, but it contains animal proteins in which the deficien- cies are not so common. Most important work of this character is being conducted by McCol- lum and his collaborators.12 It would take us beyond the confines of this book to discuss the results in detail, but it may be mentioned that they have shown that, since the adequacy of the diet depends on a multiplicity of factors besides the amino-acid make-up of proteins, — some of which we shall discuss immediately, — very extensive observa- tions with various food mixtures must be conducted over long periods of time. The nutritive values of the common cereals added to a stand- ard diet that had brought the animals (rats) to the threshold of death, 582 METABOLISM were found to be as follows: With, cornmeal there was immediate recov- ery and rapid growth, both of which were also secured in considerable degree by wheat embryo and entire wheat kernel ; with rye and oats, on the other hand, there was little if any improvement. Much work is, of course, yet to be done before we can determine the exact role which each unit plays in the physiologic development of young animals. To sum up what we already know, it may be said that glycocoll is not essential, since it can be manufactured by the animal itself; that tryptophane is essential for maintenance, probably because it is required for the production of certain essential hormones, for the make-up of which in its absence other tissues must become disintegrated, leading therefore to a diminution in body weight; and that lysine ap- pears to be essential for growth. Tissues can be maintained without lysine, but they can not grow, for the slight trace which most food con- tains of this important amino acid may be sufficient for maintenance purposes, but utterly inadequate for growth. That the young rats in the experiments of Mendel grew normally while living on milk supplied by the stunted mother indicates that the requisite lysine must have been produced in the mother's body. In the application of the foregoing principles to human dietetics, it is undoubtedly safe to follow Bayliss's advice to take care of the calo- ries and allow the proteins to take care of themselves.11 For example, in the case of milk the deficiency of cystine in its chief protein, casein, is corrected by the presence of lactalbumin, which, though present in only small amounts, contains sufficient quantities of this amino acid to meet the demands of the growing tissue. • These observations on maintenance and growth suggest very interest- ing applications in connection with the growth of tumors. Is it possible that we might retard the growth of tumors by a diet that was insufficient for growth while sufficient for maintenance. In an experiment devised to test this proposition mice were fed on a diet of starch, lard, lactose and gluten on which they could merely maintain existence but failed to grow. Some of these rats were inoculated with a rapidly growing tumor at the same time as another batch of mice kept on normal diet, and it was found that the tumor grew much more slowly in the stunted mice than in the others. One mouse, for example, on the restricted diet had a scarcely visible tumor 52 days after the inoculation. When this mouse, however, was placed on a normal diet of bread, milk, etc., the tumor immediately began to grow at a very great rate.13 Too much importance should not be placed on this experiment. We shall now pass on to consider some of the factors besides the pro- tein content which have an important bearing on dietetic efficiency. CHAPTER LXV NUTRITION AND GROWTH (Cont'd) THE RELATIONSHIP OF OTHER FACTORS THAN PROTEINS The Relationship of Carbohydrates. — As we have seen elsewhere, car- bohydrates are almost certainly essential for normal metabolism. If they are not given with the food, they must be manufactured out of protein by the organism itself. It is not surprising, therefore, that their absence from the diet of growing animals should lead to abnormality in the rate of growth. Pediatrists have not infrequently insisted that one form of carbohydrate is more advantageous for growth than another. This no doubt in the main is true, but the whole question of adequacy probably depends on the digestibility of the carbohydrate and not upon its essential chemical nature. It is likely that the only carbohydrate required by the tissues is glucose. The readiness with which the car- bohydrate of the food becomes converted into this monosaccharide is pr-obably the only determinant of its efficiency as food material. The Relationship of Fats.— Although fats are an invariable constit- uent of practically every diet, it is yet a debatable question as to whether they are essential to the maintenance of a healthy normal organism. Difficulties standing in the way of a solution of this problem are that it is not only technically very difficult to remove fat entirely from the common foodstuffs, but also that the simple fats are usually associated with substances having similar solubilities and physical properties: namely, the lipoids, phosphatides, cholesterol, pigments, etc. Since these substances are present in practically every cell, it is almost certain that they can be manufactured by living protoplasm. Indeed, experimental evidence is not wanting to show that this is actually the case. Although the cell can manufacture lipoids, a young animal can apparently not grow when these substances, as well as simple fat, are entirely absent from the diet. This has been shown by feeding young mice on a diet from which all traces of fat and lipoids had been removed by extraction with alcohol and ether (Stepp)14. On such a diet the mice lived only a few weeks. They could be kept alive much longer when some of the alcohol-ether extract was mixed with the diet, but not so when neutral fat instead of the alcohol-ether extract was added. The 583 584 METABOLISM addition of the ash of the lipoid extract failed to maintain the mice, so that the lacking substance could not be inorganic in nature. More recent and extended observations, however, have shown that neutral fat is also necessary for the adequate and continued growth of the animal. For a period of two months or so an animal may, as we have seen from Osborne and Mendel's experiments, grow in apparently nor- mal fashion on an artificial fat- and lipoid-free diet composed of casein, carbohydrate and inorganic salts, but sooner or later the great majority of these animals begin to show failure of adequate growth. The in- adequately growing animals often manifest indications of malnutrition other than the failure to increase in weight; for example, inflammation of eyes, roughening of the fur, etc. When certain fats are added to the inadequate diet, normal growth is immediately resumed. Fats pro- ducing this normal growth are such as butter fat, or the fat extracted from egg yolk, or cod-liver oil, added to the extent of 5 per cent of the ration. On the other hand, vegetable oils, such as olive oil or almond oil, are inefficient in promoting growth. That all oils or fats do not suffice to produce growth, and that one dose of an adequate oil or fat may be sufficient to stimulate it, indicate that something other than the mere presence of the comparatively simple fat molecule — that is, some acces- sory material — must be the agency responsible for the growth. This conclusion is further supported by the interesting observation of McCollum and Davis that vegetable oils can be rendered efficient for growth by shaking them with a solution of soap prepared by com- pletely saponifying butter fat with potassium hydroxide in the absence of water. ACCESSORY FOOD FACTORS, VITAMINES In searching for the nature of the accessory food factors, the im- portant observations which have been made in recent years concerning the so-called vitamines must be considered. These are substances essential in the diet for the proper maintenance of nutrition in adult animals. The existence of such substances was suggested by observations on the disease beriberi, which is caused by exclusive feeding on polished rice; that is, on rice from which the pericarp had been removed by the process of polishing. When patients suffering from this disease were given unpolished rice, the symptoms immediately disappeared. Further investigation of the exact nature of these substances was greatly facil- itated by the discovery that a similar condition is readily induced by feeding fowls on polished rice. The birds develop a polyneuritis, from which, however, they very promptly recover if some rice polishings or, NUTRITION AND GROWTH 585 better still, an extract of rice polishings, is added to the polished rice diet. The extract is made by means of slightly acid 91 per cent alcohol, and from it Funk has succeeded in separating a substance in crystal- line form apparently related to the pyrimidines, which it will be re- membered are a characteristic constituent of the nucleins. Doses as small as 0.02 to 0.04 gm. of this material given by mouth were adequate to cure the polyneuritis of fowls in from six to twelve hours; indeed, in some cases the bird seemed quite well after three hours. A similar sub- stance has also been extracted from yeast, milk, brain and lime juice, and it has been called, for want of a better name, vitamine.15 It is quite likely that other diseases, such as scurvy, may also be due to the absence of some vitamine in the diet — some substance, namely, which in the case of this particular disease would seem to be absent in preserved food, the continued taking of which is so frequently its cause. Fresh fruit and other foods added even in small amounts to such a diet would appear to supply the necessary vitamine. It is not the higher animals alone that suffer from the want of some such substance as vitamine. It has been shown, for example, that, when a normal artificial culture medium is inoculated with yeast in very small amounts, it fails to grow, whereas the same quantity will grow luxuriantly in a medium to which sterilized beer wort has been added. Vitamine is not of the nature of a ferment, since it withstands heating to 120° C. for more than an hour. The addition of yeast to dietaries that are deficient in vitamines is an excellent corrective. Returning now to the accessory substances that seem to be adherent to certain forms of fat, we see at once that they can not be exactly the same as the so-called vitamine of Funk, for they contain no nitrogen. There are, therefore, probably two accessory factors concerned in ade- quate growth. One of these must be present in the protein-free milk which serves as a constituent of the basal diet used in Osborne and Mendel's experiments, for we have seen that animals will grow on this for a certain period, provided the proper amino acids are present. Later, however, they pass into a state where there is no growth but adequate maintenance. If now the other accessory factor is added, as, for example, butter fat or a small amount of milk itself (i. e., in place of protein-free milk), then growth will be resumed at its normal rate. "Either of the determinants may become curative. Both are essential for growth when the body store of them becomes depleted." McCollum suggests that these accessory factors should at present be called the "fat-soluble A" and "water-soluble B." The latter is present in yeast cells, in fat-free milk, and in many other animal foods, and is probably the same as Funk's vitamine. The former is soluble in the fat solvents, 586 METABOLISM being present in most animal fats, but not in all; for example, it is absent from the fat surrounding the pig's heart. By using such a nomenclature it is recognized that the subject 'is as yet only in an early state of development. "We may sum up the main facts of this chapter by stating that growth and maintenance are more than a mere problem of energy supply. Granted that this is sufficient, we must also have a suitable admixture of building units of protein and the presence of extremely small quantities1 of some unknown accessory substances. These are present in some natural foods but not in others, and some are soluble in water and others in fats. They are found, for example, in animal fats but not in those of vegetable origin. Both fat- and water-soluble factors are present in large quanti- ties in milk. Both accessory food factors are necessary, as is illustrated in the fol- lowing summary of experiments from Lusk's "Science of Nutrition," (third edition). Purified protein -f carbohydrate + vegetable fat + inorganic salts — no growth. " -f " + butter fat 4- " " — no growth. " " + " + vegetable fat + " " +vitamines (accessory factor B) = no growth. " " + " + butter fat + " " + vitamines = growth. The Relationship of Inorganic Salts. — Inorganic salts ale also an es- sential ingredient of the diet. McCollum found that young animals soon ceased to grow when fed on a diet of corn and purified casein, but that rapid growth returned when a suitable salt mixture was added. Oats, wheat, and beans have also been shown to require some adjustment of their ash content to make them adequate for growth. Most of the animal foods contain in themselves sufficient inorganic material, as is evidenced among other things by the adequacy of milk alone as diet for growing animals and the abhorrence of salt that is shown by strictly carnivorous animals. In the usual mixed diet of man there is almost always enough inorganic material, the salt which he adds being largely for seasoning purposes. When a preponderance of vegetable food is taken, however, the salt comes to have a real dietetic value. The practical application of the results of these numerous and at present somewhat bewildering observations to the nutrition of man, and particularly to the dietetic treatment of disease, is undoubtedly very great. This is especially so in infants and growing children, in whom the correction of some slight inadequacy in the diet may have the most pronounced results, not only on growth and nourishment, but also on the power of resistance against disease and infection. The bene- ficial influence of cod-liver oil, for example, may depend on some fat- NUTRITION AND GROWTH 587 soluble accessory food factors, while the miraculous benefit which scorbutic children derive from the addition of the juice of limes, lemons, etc., to the food is undoubtedly due to such influences. The accumu- lating mass of evidence as to the faulty nutrition in animals fed on single kinds of food that fail to contain both kinds of food factors emphasizes the necessity in the dietetic treatment of such diseases as diabetes, nephritis, etc., of seeing to it that the diet is sound, not only in calories, protein content, and palatability, but also with regard to the presence of accessory food factors. CHAPTER LXVI DIETETICS THE CALORIE REQUIREMENT In the application of the important facts that have been reviewed in the preceding chapters to the science of dietetics, the question arises as to how we may determine with scientific accuracy just exactly how much food should ~be taken under varying conditions of bodily activity. In a general way, we know that the amount of food that we require to take is proportional to the nature and amount of bodily exercise that is being performed at the time ; and that, if the food supply is inadequate, the work before long will fall off not only in quantity but in quality as well. "Horses (also men) work best when they are well fed, and feed best when they are well worked," is an old adage and one the truth of which can not be overestimated in the consideration of all questions of dietary requirements. An ill-fed beggar will rather suffer from the pain and misery of starvation than attempt to perform a piece of work that the well-meaning housewife bargains should be done before she gives him a meal. The spirit may be willing but the flesh is weak. If he could be trusted, he should be fed first and worked afterwards. Besides the amount of work, two other factors are well known to influence the demand for food — namely, growth and climate. A young, growing boy will often demand as much if not more food than would appear to be his proper share, from a comparison of his body weight with that of his seniors; and, other things being equal, it is well knoAvn that we are inclined to eat much more heartily of food during the cold days of winter than during the sultry days of July and August. That we know these facts in a general way, indicates that the first step to take in the exact determination of dietetic requirements is to find out how mueh energy the body expends under varying conditions of activity. This, as we have seen, may be done by having the person live for some time in a respiration calorimeter, so that we may measure the calorie output. To the conclusions drawn from results of observa- tions made under such artificial and unusual conditions of living, the objection might, however, quite justly be raised that they need not apply to persons going about their ordinary routine of life. To meet 588 DIETETICS 589 this objection another method, which we may call the statistical, is avail- able. It consists in taking the average diet of a large number of indi- viduals and comparing the calorie value with the average amount and type of work that they are meanwhile called upon to perform, and can best be used where the diet is accurately known, as in public institu- tions, the army, the navy, etc. The total food supplied is then divided by the number of individuals, this giving the per capita consumption. Obviously some get more than others, but when a sufficient number of individuals is included, such errors become eliminated by the law of averages. The reliability of this method is testified to by the remarkable corre- spondence in the calorie value of the food consumed by farmers in widely different communities: Calories Farmers in Connecticut 3,410 " " Vermont 3,635 " ". New York 3,785 " " Italy 3,565 " " Finland 3,474 Average 3,551* *Lusk: The Fundamental Basis of Nutrition. The average inhabitant of various cities: London 2,665 Paris 2,903 Munich 3,014 Konigsberg 2,394** «*Rubner. Individuals in different callings: Farmers ' families (U.S.A.) 3.560 Mechanics' families (U.S.A.) 3,605 Professional men's families (U.S.A.) 3,530 Army (U.S.A.) 3,851 Navy (U.S.A.) 4,998t tAtwa'er. In general, it is usually computed that a man weighing 70 kg. requires in calories: 2,500 for a sedentary life, 3,000 for light muscular work, 3,500 for medium muscular work, 4,000 and upwards for very hard toil.t JMcKillop. These figures apply to the average man, but in calculating the calorie requirements of a family or a community we must make allowance for the lesser requirements of women and children. Several dietitians have compiled tables showing how many calories are expended according to age and sex, and the German authorities have recently taken these figures and from them calculated a generalized mean, which shows in comparison 590 METABOLISM with men the percentage that should be allowed for women and children. The figures are as follows: Man 100 Woman 83 Boy over 16 92 Boy 14-16 81 Girl 14-16 74 Child 10-13 64 Child 6-9 49 Child 2-5 36 Child under 2 23 In calculating the calorie requirement of the population as a whole, the necessity of making allowance for the varying needs of men, women, and children would obviously make the calculations far too complicated for practical purposes. It is necessary to have a factor by which we may multiply the total population in order to determine its ' ' man value. ' ' This factor is based on the relative proportion of men to women and children, and it amounts very nearly to 0.75, i. e., three-quarters of the total population gives "the man value." Knowing the total population, say, of a city, we must therefore multiply this by 0.75 in order to ascer- tain for how many men doing moderate muscular work (3000 C.) food has to be provided. THE PROTEIN REQUIREMENT The facts considered in the previous two chapters lead to the question: To what extent may the proportion of protein in the diet be reduced with safety? It is evident that there must be a minimum below which every one of the necessary building materials of protein could not be supplied in adequate amount to reconstruct the worn-out tissue protein. The extent to which the protein content of the diet of man can be lowered with safety depends on several factors, of which the most im- portant are: first, the nature of the protein; second, the number of non- protein calories ; and third, the extent of tissue activity. Where so many factors must be taken into consideration, the only method by which the actual minimum can be determined consists in what may be called "cut and try experiments." Of the many investigations of such a nature, probably the best one for us to consider, is that recently published from the Nutrition Laboratory of Copenhagen. The subject, an intelligent laboratory servant, lived a perfectly normal and active life for a period of five months on a diet of potatoes cooked with margarine and a little onion, and containing 4000 C., with a total protein content of 29 grams. During another period he did outdoor work as a mason and laborer, and took 5000 C. daily, and 35 grams of protein. DIETETICS 591 It is important to contrast these results with the following based on municipal statistics of gross consumption. MUNICIPAL FOOD STATISTICS PROTEIN FAT CARBOHYDRATES CALORIES Konigsberg Munich gm. 84 96 gm. 31 65 gm. 414 492 2394 3014 Paris 98 64 465 2903 London 98 60 416 2665 It is certain that man can lead a normal existence and remain in good health on very much less protein than the 100 grams which statistical studies show to be the amount he actually takes. This discrepancy be- tween the amount which experiment demonstrates to be adequate and that which habit and custom demand, raises the question as to whether, after all, our instincts may not have erred and so made us unnecessarily extravagant in our protein intake. It has been suggested that such pro- tein extravagance will in various ways have a dele'terious effect on the organism; thus, that the excretory organs, such as the kidneys, will be overtaxed in eliminating the unused amino acids, that the constant pres- ence of these bodies in excess in the blood will cause degeneration and sluggish metabolism, and that the excess protein in the intestine will lead to the production of ptomaines, whose subsequent absorption into the blood will cause toxemic symptoms. Important support to such views appeared to be supplied some dozen years ago by Chittenden, who was able to show that he himself and many other persons doing different kinds of work could be supported on daily amounts of protein that were not more than from one-third to one-half of the amount usually taken. Not only so, but it was averred that dis- tinct improvement was experienced in the general sense of well-being and of mental efficiency as a result of the lesser protein consumption. Taking these results as a whole, it is quite clear that man can get along under ordinary conditions with much less protein than he usually takes; but that really proves nothing, for the question is not can he but should he, so deprive himself? Are instinct and custom wrong and is Chittenden right ? That is the question. To answer it many studies have been made of the condition of peoples who for economic or other rea- sons are compelled to live on less protein than the average. Are these people healthier, less prone to infections and degenerative diseases, and more efficient mentally than others? In such studies great care must be exercised to see that conditions other than diet, such as climate, exercise, etc., are properly allowed for. It would not be fair, for example, to compare the mental and bodily condition of pepples living in the tropics 592 METABOLISM and who take comparatively little protein, with those living in temperate zones, who consume much more. After discounting all of these other factors, it has been quite clearly shown that, when the protein allowance is materially reduced, the people as a whole are less robust, mentally in- ferior, and, instead of being less prone to the very diseases which are usually supposed to be due to overloading of the organism with useless excretory products, are more liable to suffer from them. That a decided reduction in -protein weakens the defense of the organ- ism against infection is probably due to the fact that the fluids of the body normally contain a great variety of so-called antibodies — that is, of highly complex substances that are largely protein in nature. When bacteria, or the poisons produced by them, enter the body, they are met by one or more of these defense substances and destroyed or neutralized. Now it is clear that there should always be a surplus of protein-building materials from which the antibodies may be constructed. Such an excess will constitute a "factor of safety" against disease. And there are fac- tors of safety of another nature to be provided for, two of which we are in a position to appreciate. In the first place, there must always be an adequate supply of tryptophane, of lysine, and of cystine, not only to meet the bare necessities of the protein constructive processes that go on under normal conditions, but also to make good the larger amount of protein wear and tear that greater degrees of tissue activity will entail. Although moderate muscular exercise does not appear to cause any im- mediate consumption of protein (carbohydrate and, later, fat being the fuel material used to produce it), yet it does throw a greater strain on the tissues and causes a greater wear and tear of the machinery, and hence a demand for more protein-building material. In the second place, there are certain of the internal secretions of the body, such as epineph- rine (adrenaline), that are essential for life, and as crude materials for the manufacture of which certain amino acids are essential. Tyro- sine is one of these, and since proteins, as we have seen, differ from one another quite considerably in the amount of this amino acid which they contain, it is advisable to provide an excess, so that an adequate supply of tyrosine may always be available. The answer to one of the most important practical questions in die- tetics— namely, "What proportion of protein should the diet contain?" depends on these scientific principles. The source of the protein is the important thing. With animal protein there is no doubt that we could get along with perfect safety by taking daily not more than 50 or 60 grams, which is about half of what we actually consume. If the protein is of vegetable origin and part of it of the first quality, as wheat and Indian corn preparations, more should be taken so as to allow for the DIETETICS 593 deficiency of certain amino acids. When vegetable proteins of the sec- ond quality, such as those of peas, beans, lentils, etc., are alone available, much larger amounts are necessary. Such proteins are inadequate in the case of growing children at least, and even in adults it is undoubtedly advisable that other proteins should supplement them. To insure safety, therefore, it is almost imperative that the diet should contain proteins of various sources. If for economic reasons the main source must be proteins of vegetable origin, then some animal protein, such as is contained in milk or meat or eggs, should be added to at least one of the daily meals. When peas and beans are mainly depended on for the protein supply, they should be taken either with milk or one of its prep- arations, or with a thick gravy or sauce made from meat and containing the finely minced meat. This must not be strained off, for if it is, the sauce will contain only the meat extractives but not any of the protein, which is coagulated by the boiling water. Meat extract, in other words, contains no proteins; it is not a food but merely a condiment of no greater dietetic value than tea or coffee. ACCESSORY FOOD FACTORS Little need be added to what has already been said regarding this subject. The practical point to be remembered is that there are at least two accessory factors concerned, one of them soluble in fat and present in adequate amount in butter and other animal fats but not in vegetable oils, and the other soluble in water and present in wheat, vegetables, fruits, etc. Milk contains both of these factors, so that its inclusion in a diet is a safeguard not only against inadequacy in suitable protein, but also against the absence of accessory food factors. There is little danger of the diet being inadequate with regard to food factors if it contains some fruits or green vegetables or unheated fresh milk. The food fac- tors are destroyed by prolonged cooking. DIGESTIBILITY AND PALATABILITY We have seen that practical dietetics depends on several factors, the exact relative importance of which can not perhaps be gauged in every case, but preparation of the food so as to make it appetizing must un- doubtedly rank high. The importance of good cooking will now be ap- parent. It is the act of making food appetizing and therefore digestible. It is really the first stage in digestion, the stage that we can control, and one therefore to which much attention must be given, especially when it becomes necessary to make attractive articles of diet ordinarily considered common and cheap. Most people can cook a lamb chop so as to make it 594 METABOLISM reasonably appetizing, but few can take the cheaper cuts of meat and con- vert them into cooked dishes that are as popular and attractive. And there are still fewer who can take the left-overs and trimmings and convert them in the same way. This is the real art of cooking, and too much encourage- ment can not be given to the effort which our cooking experts are making to show people how these things can be done. The waste of good food in a large city is truly appalling. Cooking has other advantages than making the food appetizing. The heat loosens the muscle fibers of the meat so that it is more readily masticated; it destroys microorganisms and parasites in the meat; it de- stroys antibodies which might interfere with the action of the digestive ferments. Thus, untreated raw white of egg is not digested in the stom- ach because it contains one of the antibodies which prevent the pepsin from acting on it; but boiled egg white, if properly chewed, can be di- gested, and whipping the egg white into a foam partly destroys the in- hibiting substance. Before concluding, something should be said about the laxative quali- ties of food, for it is often in this particular alone that one food is more satisfactory than another. A diet of meat, milk, eggs, and white bread is apt to be unphysiologic because there is nothing in it to act as what has been called intestinal ballast; that is, a material which will keep the intestines sufficiently filled to stimulate their muscular movements. This ballast is best furnished in the shape of cellulose, the most important constituent of green food. Peas, beans, cabbage, salad, and many fruits, especially apples, should always occupy a place in the daily menu. An- other valuable food yielding this ballast is the outer grain of wheat, oats, etc. So much must not be taken as to produce a constant intestinal irritation, and each person must determine for himself where this limit lies. The difference among various breads is almost entirely in the de- gree to which they supply ballast. The all-important subject of food economies can receive no attention here, except to point out that it is one which must be most carefully con- sidered in the solution of all problems of dietetics. An admirable ac- count of the subject will be found in Graham Lusk's "Science of Nutri- tion" (third edition) and in McKillop's "Food Values."16 . CHAPTER LXVII THE METABOLISM OF PROTEIN Introductory. — The older physiologists believed that the protein taken with the food was brought into a soluble condition by the digestive en- zymes, and that it was then absorbed into the blood and directly incor- porated with the tissues. The discovery of the enzymes trypsin and erepsin and of free amino acids in the gastrointestinal contents clearly showed that this simple theory of Liebig could not be correct. It was, furthermore, found that when an excess of proteins such as egg albumin gains entry to the blood, part of the protein appears in an unchanged condition in the urine ; and that enzymes capable of digesting this protein but not other varieties make their appearance in the blood. After the injection of foreign proteins into the blood, symptoms of varying severity often develop, from the almost instantaneous death produced by snake venom to the slowly developing anaphylactic reac- tions which follow the injection into the blood of many proteins chemi- cally indistinguishable from those of the blood serum itself. When pro- tein is taken in the usual amounts by mouth, these poisonous reactions do not supervene, — even snake venom is harmless when swallowed, — nor is it "possible during digestion of a protein meal to detect food protein in the blood by means of the precipitin reaction. Finally it was discovered that the very slow intravenous injection of completely digested flesh did not produce on the part of the body any of the reactions that injected protein itself produces, indicating that perfect assimilation had occurred. From these and similar observations it soon became clear that protein can not be absorbed as such from the alimentary canal, but must first of all ~be completely broken down into the amino acids, which are then rebuilt into the protein of the organism. The direct evidence for this important change in belief concerning protein metabolism has been gained by the discoveries that: (1) nitrogen equilibrium can be maintained in animals fed with completely digested protein mixtures; and (2) amino acids can be isolated from the blood. The experiments of the first group consist, in principle, in breaking down protein until there is no longer the characteristic biuret test, and then feeding this digestion mixture to animals and observing them from day to day, using as criteria of their nutritional condition the body weight 595 596 METABOLISM and the nitrogen equilibrium, (Page 571.) It has been shown that suc- cess in maintaining nutritional efficiency depends partly on the nature of the process used for digesting the protein, and partly on the presence or absence of carbohydrate in the digestion mixture. It was found that acid hydrolytic products failed to maintain equilibrium, and it was believed that this was owing to the fact that the acid had more completely disrupted the protein molecule, and had left no polypeptides, which, it was imagined, remained intact during enzyme action and were essential for proper protein metabolism. This view has now been considerably altered, since it has been shown that the acid actually destroys certain ammo acids which the enzyme leaves intact. The amino acid particu- larly concerned is tryptophane. Thus, when animals were fed with three diets, consisting of (1) fully digested casein, (2) fully digested casein from which the tryptophane had been removed, and (3) fully digested casein from which the tryptophane had been removed and then the proper amount of pure tryptophane added, it was found that nitrogen equilibrium could not be maintained on the second diet, which contained no tryptophane, whereas it was maintained on the first and third diets. That this explanation is correct is further supported by the fact that, if the protein is only partly digested by acid — that is, not digested enough so as to break up all the tryptophane — it can efficiently maintain nitrogen equilibrium. Eegarding the necessity for carbohydrates, it is possible that under certain conditions these may be produced from the protein itself. At least, it has been possible for Abderhalden, who has done a large share of this work, to maintain an animal in nitrogen equilibrium with a diet of digestion products and fat containing no carbohydrate. These results obtained in different classes of animals have also been confirmed for the human subject. A boy suffering from a stricture of the esophagus, when fed by rectum for fifteen days with digestion products resulting from the action of trypsin and erepsin on flesh, gave evidence of nitrogen retention. Concerning the second type of evidence, many investigators attempted to separate the amino acids themselves from the blood, particularly dur- ing the digestion of a large amount of protein, but the results were at first entirely negative because of the lack of methods that were suffi- ciently delicate to make it possible to detect the slight increase that could be expected even when a maximum absorption of nitrogen had occurred. The very large flow of blood through the portal vein causes such extensive dilution of any substances added to it that the concentra- tion of the substance in an isolated sample of the blood can be only trivial. THE METABOLISM OF PROTEIN 597 To account for the indisputable disappearance of the amino acids from the intestine during protein digestion, coupled with the impossibility of detecting any of them in the blood, two views were current for many years. One of these was that the amino acids become deaminated (NH2 group split up as NH3) by the intestinal epithelium, and the other, that these cells are endowed with the power of reconstructing the amino acids into protein, which then passes into the blood. Justification for the de- amination hypothesis seemed to be obtained by the observation that there is more free ammonia in the blood of the portal vein than in that of the systemic circulation. The falsity of this evidence was, however, defi- nitely established by Folin and Denis,32 who found by means of delicate quantitative methods for the estimation of ammonia and urea in the blood that the amount of neither of these substances became increased in the portal blood during absorption of amino acids from the intestine. They made the further important discovery that the ammonia in the portal blood is really very little in amount, and represents that absorbed as such from the intestinal lumen, where it is produced chiefly by the action of putrefactive bacteria. Nor could any evidence be obtained in favor of the hypothesis that the absorbed amino acids become built up in the intestinal epithelium into proteins, which are then transformed or carried away by the blood. This hypothesis was based entirely on negative findings, and had there- fore to be dropped when discovery was made of the actual presence of amino acid in the blood. This brief historical survey of the subject brings us to a position where we may proceed to discuss the present-day teaching regarding protein metabolism. Briefly stated, this teaching is to the effect that the protein molecule is broken down into its ultimate building stones, the amino acids, by the digestive enzymes of the gastrointestinal tract, and that these amino acids are absorbed into the blood, by which they are carried to the various organs' and tissues, which sift out the amino acids and use those of themi which they require for the reconstruction of their broken-down protein. The amino acids not required for the process, along with those wliich may be liberated in the tissues themselves by disintegration of tissue proteins, are then split into two portions, one represented by ammonia and the other by the remainder of the amino acid molecule. The former is excreted as urea and the latter is oxidized to produce energy. CHEMISTRY OF PROTEIN Before proceeding to discuss the evidence upon which the above con- clusions depend, it will be necessary to consider some of the most important facts concerning the chemistry of the protein molecule. We shall require 598 METABOLISM this information not only to understand the history of protein in the animal body, but also to follow intelligently the important work that has already been discussed concerning the relative value of different proteins as food. A knowledge of protein chemistry has come to be essential in practically all branches of medical science. Proteins, like starches, are composed of numerous smaller molecules, In the case of starch these molecules are the various monosaccharides — glucose (dextrose), levulose and galactose; in the case of proteins they are the amino acids. The breaking apart of the links that hold the mole- cules together is effected in both cases by the process of hydrolysis, so called because of the fact that the reaction consists in the taking up of a molecule of water at each of the places where the chain falls apart. This hydrolysis may be effected either by the action of mineral acids or alka- lies, or by enzymes, the only difference' in the action of these reagents being that in the former case the breaking apart takes place more or less indiscriminately, whereas in the latter it proceeds according to a definite plan, which varies somewhat with the type of enzyme employed. Just as a chemical knowledge of the structure of sugar or monosac- charides is the basis of carbohydrate chemistry, so is that of the amino acids the basis of protein chemistry. Amino Acids. — There are, so far as known, eighteen different amino acids concerned in the constitution of protein, but they are all alike in their characteristic structure. The most striking characteristic depends on the presence in the molecule of: (1) an amino group with a basicity comparable to that of ammonia, and (2) an acid group with an acidity comparable to that of acetic acid. Let us take in illustration one of the simplest fatty acids — namely, acetic. It has the formula CH3COOH. The COOH group is called carboxyl, and on it depend the acid properties of the compound. The CH3 group is known as methyl, and the amino group (NH2) is attached to it in place of one of the hydrogen atoms, thus giving the formula CH2NH2COOH, which is aminoacetic acid or- gly- cocoll. If we take the next higher acid of the fatty acid series, having the name propionic and the formula CH3CH2COOH, its amino acid, called alanine, has the formula CH3CHNH2COOH. Now let us place the formu- las of these two acids side by side in the following manner: H CH3 NH2-C-COOH NH2-C-COOH (amino group) H (acid group) (amino group) H (acid group) Aminoacetic acid Aminopropionic acid (glycocoll) (alanine) THE METABOLISM OF PROTEIN 599 It will be observed that the only difference between the two acids is dependent upon a change in the group that is attached to the upper verti- cal valency bond of the central carbon atom, which therefore must be considered as the center of the entire molecule. The various amino acids produced from protein differ from one another solely with regard to the chemical nature of the group that is attached to this vertical valency bond. Evidently, then, the reactions that amino acids possess in common must depend on the terminal groups containing the carboxyl and amino radicles, whereas the characteristic reaction of each of the eighteen amino acids must depend upon the differences in the radicles attached to the upper vertical bond. This may be represented thus: Any radicle NH,-C-COOH H Any amino acid The end groups endow the amino acids with the power to combine with both acids and bases. With acids they behave like substituted ammonias to form salts, which can ionize into the amino acid, as the cation, and the acid group, as the anion. With bases the carboxyl group reacts to form salts, which yield amino acid as the anion. A most important reaction con- sists in the condensation of aldehydes with the amino group. This occurs particularly readily with formaldehyde, water being eliminated in the re- action, and the basic nature of the amino acid being thus destroyed. Upon this reaction depends the method of Sorensen for determining the amount of amino acid in a mixture (see page 606). The titration is per- formed by rendering the solution of amino acids neutral, then adding formaldehyde and titrating with standardized acid, using phenolphtha- lein as the indicator, and thus finding to what degree the acidity of the mixture has become increased as a result of adding the formaldehyde. Since this increase in acidity must depend upon the number of amino groups, it furnishes us with an indirect estimate of the concentration of the amino acids. The reaction is illustrated by the equation: radicle H radicle NH.-C-COOH -1-11-0 = 0 = CH, = N-C-COOH + H2O i A (amino acid) (formaldehyde) Another reaction of amino acid of physiologic interest is that known as the carbamino reaction, consisting in a union of the amino acid with calcium and carbonic acid. Finally, it is important to note that the amino 600 METABOLISM group is very firmly attached; it remains intact in acids and alkalies and is removable only by a process of oxidation. This can be accomplished by treating the amino acid with such reagents as hydrogen peroxide or per- manganate, when the amino group is displaced and a so-called ketonic acid formed. The reaction will be evident from the accompanying equation: CH3 CHS O + NH,-C-COOH ?± O = 0-COOH+ NH3 H ( alanine ) ( pyruvic acid ) To illustrate this reaction we have chosen aminopropionic acid or ala- nine, because the substance formed by its oxidation and known as pyruvic acid is of very great importance in intermediary metabolism. It serves as the common substance from which proteins, carbohydrates or fats may be formed, and therefore as the intermediary substance through which one of them may pass on being transformed into another (page 666). The use of two arrows pointing in opposite directions in the above equation indicates that the reaction may proceed readily in either direction. The ammonia set free from amino acids may be oxidized to free nitrogen by using nitrous acid according to the general equation: NH3-f-HONO= 2H20+N2. Upon this reaction depends another extremely important quantitative method for measuring the number of amino groups present in protein (Van Slyke). To make the estimation, nitrous acid is allowed to act on the amino acids, and the volume of nitrogen gas set free by the reaction is measured, the principle being similar to that used for the de- termination of urea by the hypobromite method. The apparatus employed for decomposing the substance and collecting and measuring the evolved nitrogen consists essentially of a mixing bulb, connected below through stop- cocks with two small burettes, one containing a solution of sodium nitrite and glacial acetic acid, and the other a solution of the substance to be investigated. The upper end of the mixing bulb is connected through a three-way cock with a graduated gas burette and with another bulb containing potassium permanganate solution. By allowing some nitrite and acid solution to run into it and shaking, the mixing bulb is first of all filled to a certain mark with nitrous oxide gas. A measured quantity of the amino solution is then allowed to mix with the nitrite; the apparatus is shaken for five minutes at 15 to 20° C., and the evolved nitrogen and nitric oxide are dviven over into the permanganate, which absorbs the nitric oxide, leaving the nitrogen, which is then measured in the burette. The apparatus has now been so perfected that numerous analyses may be made with it in a very short time and with a degree of accuracy that is scarcely surpassed in any other biochemical estimation. From the point of view of protein chemistry, the most significant reac- tion of the amino acids is their ability to link together to form compounds THE METABOLISM OP PROTEIN 601 called peptides. This linking occurs between the amino group of one amino acid and the carboxyl group of the other. When alanine and glyco- coll, with which we are familiar, are thus linked together, the reaction takes place according to the equation: H CH H /!H + HO: OC-C-NH, — HOOC - C -NH - CO- C- NIL + H2O HOOC-C-N | . I ] | ^ \H H H H (alanine) (glycocoll) (alanyl - glycocoll) In this manner, then, a so-called dipeptide is formed, in which it will be observed there still remain free carboxyl and amino groups, thus per- mitting the linking on of other amino-acid groups to form tripeptides or tetrapeptides or other polypeptides. Indeed, this process of condensa- tion may go on practically indefinitely, a polypeptide containing eighteen amino-acid groups — namely, three leucine and fifteen glycocoll groups — hav- ing actually been synthesized. The resulting polypeptides have the proper- ties of derived proteins like the proteoses; thus, they give the biuret and other reactions characteristic of proteins and are precipitated by such reagents as mercuric chloride and phosphotungstic acid. Some are also digested by trypsin and erepsin. They have the same optical" activities as proteins. One of them has been prepared which produces a typical anaphylactic reaction. So far a polypeptide that can be coagulated by heat or is in other ways identical with the naturally occurring proteins, has not been synthesized ; but there is no doubt that it is only a matter of time before this will be accomplished. Eighteen distinctly different amino acids have been isolated from pro- tein, and it may assist in the conception of our problem if we place these amino acids side by side and link them together in the manner described above. This is done in the accompanying chart compiled by D. D. Van Slyke, in which also various other important facts concerning the chem- istry of the amino acids are incidentally added. At the lower part of each formula will be seen the characteristic car- boxyl and amino groups of neighboring acids linking together the ter- minal carbon atoms. The upper vertical bond of this carbon atom is con- nected with the characteristic group of the amino acid, which may be very simple and represented only by hydrogen, as in glycocoll, or highly com- plex and including a ring formation, as in tryptophane. It will further be observed that there may be other amino groups connected in various positions in this radicle. This is particularly the case in the first three of the amino acids in the table — namely, the basic amino acids. In lysine the extra amino group reacts with nitrous acid, liberating free nitrogen 602 METABOLISM s P 1 M 1 V. o 1 1 I 2 | ft X | " 113 8— •— o— o— -w £•'§ + i 1 1 IA = •i 53 > x" = "•? z •• - f 'i'g O X X ^ „ 2 ee 8 ^1 + NEUTRAL AMINO ACIDS CONTAIN ONE AMINO OROUP AND ONE CARBOXYL GROUP. WIIIOH NEUTRALIZE EACH OTHER 1 I 1 C :S I 0 *s ^ X e § f 1 « o— B-S + » g"X8 "' X = X— o— X^ + i o o I j-xl * I?XI » o £ tt" f 5 §— 0— X| + i 8 * X J_ g g— — ,JJ 1 , J - £/"" ""I"*! ' + X ... , 'f. « X X X X 1 ijje-S^ ^~°~° 1 *l~ .1 1 '5 I_l3 5= BCLr*J * • I 1 1 x — o^ y x *f 2«> x=o o 8 X B r T J ^S-^""~^ 8 f XX . 'is, •— « X— O ^o O — O^ti * g t 4- Batio Amino Acidi Two or more basic groups to one acid OOOH group The three "hexone bases." Resemble X g g !E B ,o— o— B! i + x x" s" x" x" B" i „ SB— o—o— o o— o— X « i i B" * B B" B" B" f 1 1 f 2_0_js-o-o 3— o— a.S'S+S , a fc * THE METABOLISM OF PROTEIN 603 by the Van Slyke method; but in other cases, as in arginine, it fails to give this and the other characteristic reactions of the amino group. That the extra amino group in lysine reacts directly with nitrous acid explains why various proteins when examined for amino nitrogen yield an amount that is equal to half of the lysine nitrogen. It will further be observed that the amino acids are arranged in three main groups: one basic, another neutral, and the third acid. The acids of the basic group are three in number and have an alkalinity similar to that of ammonia. They have been called the hexone bases, because each contains six carbon atoms. They are alone present in certain forms of pro- tein called protamines. The neutral amino. acids contain one amino group and one carboxyl group, which exactly neutralize each other. This is the largest group of amino acids, and is further subdivided into three: one containing aromatic or benzene rings and including the very im- portant amino acids, tyrosine and tryptophane; another containing the so-called pyrrolidine ring; and the third, the largest of all, containing the so-called aliphatic chains; that is, the chains characteristic of the fatty acids and which may be either straight or branched. "When the chains are branched, the substance is called an isosubstance, as in isoleucine. The acid amino acids, including glutamic acid and aspartic acid, are characterized by containing two carboxyl groups and only one amino group. They therefore resemble acetic acid in acidity. It may be of assistance to some if we restate these chemical facts from a slightly different standpoint as follows: Glycine, or glycocoll, is aminoaeetic acid, CH2NH2COOH. NH2 / Alanine is glycine plus a methyl group, CH3CH ; it is therefore amino- COOH OH / propionie acid and is closely related to lactic acid, which is CH3CH . Many of COOH the other amino adds may ~be considered as derivatives of alanvne,* thus : 1. Serine is alanine with an "OH" (hydroxyl) group in place of one of the "H" NH2 / atoms of the methyl group, CH2OH - CH COOH 2. Cysteine is alanine with an "SH" (thio) group in this position, NH2 / CH2SH-CH COOH 604 METABOLISM Two cysteine molecules united at the " S " groups give cystine. NH., OILS - CH COOH NH, 3. Phenylalanine has a C6H6 (phenyl) group, CH2C6H5-CH \ COOH NH 4. Tyrosine has a C6H4OH (phenol) group. CH,CCH4OH - CH COOH C \ 5. Tryptophane has a C6H4 CH (indole) group: \ / NH C — CH2 - CH - NH2 - COOH. / \ C.H4 CH \ / NH CH / \ N NH I ! 6. llistidine has a CH — C - (imidazole) group: CH / \ N NH I ! CH = C.CH2 . CH. NH3-COOH. The last two are also called keterocyclic compounds, of which there is another, viz.; Proline (and oxyproline), which is a-pyrrolidine carboxylic acid: CH2 — CH2 I ' I CH2 CH.COOH \ / NH Other amino acids are: (1) Valine Leucine Isoleticine thus : CH. CH, CH, CH3 CH3 C2H5 \ / \ / \ / CH CH CH CH.NH2 COOH (valine.) CH2 I CH.NH2 COOH (leucine) CH.NH2 COOH (isoleucine) THE METABOLISM OF PROTEIN 605 (2) The amino dibasic acids: Aspartic, which is aminosuccinic acid, CH2COOH I CHNH2COOH; and Glutaminic, which is aminoglutaric acid, CH2 CH2 - COOH CHNH2 COOH. Lastly there are the diamino acids, in which two groups exist : Lysine a e-diaminocaproic- acid, NH2 NH2CH2 - CH2 - CH2 - CH2 - CH2 - CH COOH. Arginine o-amino — 8-guanidine-valerianic acid, NH2 HN — C NH2 NH.CH2 - CH2 - CH2 - CH COOH The guanidine group in this acid is of interest because of its close relationship to NH2 rl~ / urea, which is O = C "NH, CHAPTER LXVIII THE METABOLISM OF PROTEIN (Cont'd) AMINO ACIDS IN THE BLOOD AND TISSUES In the Blood. — Furnished with the general facts concerning the chem- istry of proteins, we may now proceed to consider the more precise knowledge recently acquired concerning the history of this substance in the animal economy. Although no one has succeeded in separating ammo acids in pure condition from drawn blood even during the height of digestion, it has nevertheless been possible to do so from circulating blood by a method of dialysis, known as vividiffusion, elaborated by Abel33 and his pupils. The method consists in connecting a long tube of collodion with the two ends of a cut artery in an anesthetized animal. The tube, coiled many times, is then immersed in a solution containing approximately the same salt content as the blood plasma of the animal. The diffusible constituents of the blood plasma dialyze into the saline solution ; or any one of them may be prevented from dialyzing by adding that particular substance to the saline in such amounts as will make its concentration in plasma and saline alike. In some ways, it will be seen, the apparatus may be considered as an artificial kidney. Its possible clinical application, for the purpose of removing poisons from the blood is under investigation. It has been possible in this way to isolate several of the amino acids and other ammonia-yielding substances from blood. Thus, alanine and valine have been obtained as crystalline salts, and histidine and creatine (see page 622) shown to be -present by their reac- tions. All of the amino substances, however, do not dialyze, and these exceptions are further characterized by the fact that they do not readily give up their ammonia on the addition of sodium carbonate, as do the diffusible substances (Rohde). Although amino acids can thus be sepa- rated in a pure state from circulating blood, their concentration in a drawn specimen is too low to make direct quantitative estimation possible. By the methods of Van Slyke and Sorensen, already described, however, it has been shown among other things that the blood always contains a certain concentration of amino acids ; thus, in that of fasting animals from 3 to 5 mg. per 100 c.c. of blood are usually found present. During the absorption of a protein meal, the amino content of the blood undergoes 606 THE METABOLISM OF PROTEIN 607 a marked increase, becoming doubled or more; and a similar result has been obtained by placing pure ammo acids in the small intestine. After 10 grams of alanine, for example, the amino nitrogen of the mesenteric blood rose from 3.7 to 6.3 mg. per cent.* In the Tissues. — After entering the circulation, the amino acid very quickly disappear from it again. This has been demonstrated by ob- serving the amount of amino acids in the blood after intravenously injecting a solution of amino acid into an anesthetized animal. After injecting 12 gm. of alanine into the vein of a dog, 90 per cent was found Fig. 186. — Vividiffusion apparatus of J. J. Abel. to have disappeared from the circulation within five minutes. The ques- tion is, What becomes of the amino acids that rapidly disappear? Are they decomposed in the blood, or do they become absorbed by the tis- sues ? This problem has been attacked by analyzing portions of various organs and tissues removed before and some "time after the injection into an animal of amino acid solutions. In the case of the muscles it has been found that the amino-acid content increases until from 60 to 80 mg. per cent of amino acid has accumulated. Beyond this point, however, the muscles do not seem to be able to take up any more amino acid. The capacity of the intestinal organs, however, is more elastic; "This is a convenient way of stating per 100 c.c. of blood. 608 METABOLISM for example, the amino nitrogen of the liver has been observed to become increased to 125 or 150 mg. per cent of the original amount. Although this absorption of amino acids by the tissues is extremely rapid, it never proceeds to such a point that the blood becomes entirely free of them. Even after many days' starvation the blood contains its normal quota of from 3 to 10 mg. per 100 gm. of moist tissue (Fig. 188). This indicates that a certain equilibrium must become established between the amino-acid content of the blood and that of the tissues, the concentration in the tissues being approximately from five to ten times greater than in the blood. 150 100 50 Injectio Muscl e z Hours Fig. 187. — Curves showing the amount of amino nitrogen taken up by different tissues after the cutaneous injection of amino acids. The lowermost curve shows the urea concentration of the blood. (From D. D. Van Slyke.) The absorbed 'amino acids are very loosely combined with the tissues, for they can be extracted by such feeble reagents as water or dilute al- cohol. Their presence can not, however, be merely due to diffusion; for if it were, the concentration could not become greater in the tis- sues than in the blood. The further fate of the amino acids is difficult to follow. We know that they do not remain in the body for a long time, because most of the protein nitrogen in the food is excreted as urea within twenty-four hours after ingestion; and when single amino acids are fed, they quickly reappear in the urine as urea. THE METABOLISM OF PROTEIN 609 The tissues can therefore be only a stopping-place for the amino acids. When the latter are determined in blood collected from different parts while absorption of protein from the intestine is in process, it has been found, as shown in Fig. 188, that during the passage of the blood through the liver there is a greater fall in the concentration of amino acids than during its passage through the entire remainder of the body. It will be seen that the above conclusions are drawn from estima- tions made on blood taken from the vena cava, portal vein, and hepatic S 3 8 O _j OQ o 8 I O Fig. 188. — Curves showing the concentration of amino-acid nitrogen in the blood during fasting and protein digestion. (From D. D. Van Slyke.) artery, the upper curves in the chart being from animals during digestion and the lower from fasting animals. The results show that the liver must be particularly greedy of amino acids, which, however, must Rapidly be- come transformed into other substances, since no conspicuous varia- tion has been found to occur in the amino-acid content of the tissues according to whether the animal is fasting or is digesting protein food. This result, it is to be noted, is quite different from that which is ob- tained after the intravenous injection of amino acids, and the results of 610 METABOLISM the two experiments taken together, indicate that the amino acids after their absorption can not remain in the tissues in a free condition for a long time. It means that the amino acids during natural digestion must ~be disposed of at a rate which is practically the same as that at wh'ich ab- sorption is proceeding. THE FATE OF THE AMINO ACIDS To follow the metabolism of the amino acids further we must deter- mine the end product into which they are converted. This is urea, whose estimation can nowadays be made with considerable accuracy on account of the discovery, by Marshall, of the action of urease in con- verting its nitrogen into ammonia, which can then be estimated by com- paratively simple methods (Folin). When the viscera are compared before and at various periods after the intravenous injection of amino acids, the immediate increase in amino nitrogen remains undiminished in all of them except the liver, in which a very rapid reduction is observed to occur. At the same time the percentage of urea in the blood steadily rises. These facts are illus- trated in Fig. 187. The simplest interpretation of these results is that the liver converts the amino acids into urea and discharges this urea into the blood. This conclusion, however, it must be observed, is not inevitable; for it is pos- sible that the amino acids may be condensed into polypeptides in the liver, just as sugar is condensed by this organ into glycogen, and that the increase in urea is merely coincident (Fiske). It must not be imagined that the conversion of the amino acids into urea is exclusively a function of the liver. On the contrary, it is well known that this process may occur in animals from which the liver has been entirely removed. It is probably safe to conclude, however, that the liver is the most active center for amino-acid transformation and urea formation. When urea is estimated in samples of blood removed at short inter- vals of time after the ingestion of a large amount of protein, it is found that the increase becomes very early established. In one experiment, before the food was taken the concentration of urea nitrogen in the blood was a little over 10 mg. per cent; one hour after taking 500 grams of meat, it had risen to about 18, and in two hours to nearly 25. Evidently the increase had occurred about the same time as the passage of food •from the stomach into the duodenum. These facts indicate that urea formation in the liver becomes stimulated long before the other tissues, such as the muscles, have had time to take up their full quota of amino THE METABOLISM OF PROTEIN 611 acids. During digestion of protein the liver does not appear to wait until the other tissues have become saturated with amino acids before it begins to destroy the unnecessary excess by conversion into urea; on the contrary, this process sets in with the very first installment of amino acid that reaches the liver by the portal blood. This conclusion is in harmony with the well-established fact that, when protein is given to a starving animal, the greater part of its nitrogen is soon excreted as urea, leaving only a small fraction to be used for rebuilding the wasted tissues (see page 643). The amino acids that are absorbed by the extrahepatic tissues become very quickly converted into formed protein, as is evident from the fact that the concentration of free amino acids in the tissues of an animal during absorption of protein is not perceptibly greater than in those of a fasting animal, and the question remains to be considered, What be- comes of the protein thus formed? The answer is, that it is gradually used up in the metabolic processes, so as to liberate again the amino acids, which add themselves to those absorbed from the intestine and be- come used again or excreted, according to the demands of the tissues at the time for amino acid. This process of liberation of amino acid from the breakdown of body protein goes on of course irrespective of absorption of amino acid from the intestine. It goes on, for example, during starvation; indeed, in this condition the percentage of free amino acids in the muscles is, if anything, somewhat higher than that observed in an ordinarily fed an- imal. In starvation also the migration of amino acid is going on among the various organs, of which those whose activity is essential to the maintenance of life, such as the heart and the respiratory muscles, are supplied with amino acids from tissues that are less vital, such as the skeletal muscles (see page 568). These experiments further show that free amino acids can not serve to any significant extent as food reserves in the same way as glycogen and fat. If amino acids were of value as food reserves, we should expect the store of them to be depleted by starvation. As to how long a period of time elapses between the incorporation of the absorbed amino acids into tissue protein and their subsequent liberation again by autolysis, we are entirely ignorant. The researches which we have just been considering do not throw any light on the relative value of different proteins in tissue metabolism. They do not inform us as to which of the amino acids must be absorbed ready-made from the digested food, and which of them may be dispensed with since the organism can manufacture them for itself. We know that the higher animals can synthesize some amino acids, such as glycocoll, but not others, such as tryptophane; but which amino acids belong to 612 METABOLISM the glycocoll and which to the tryptophane groups, can not as yet be definitely stated. The investigation of this problem, has to be under- taken by experiments of an entirely different type — namely, by observing the welfare and growth of animals fed on proteins of varying amino- acid composition. A full discussion of these experiments is given in the chapters on Nutrition and Growth. CHAPTER LXIX THE METABOLISM OF PROTEIN (Cont'd) THE END PRODUCTS OF PROTEIN METABOLISM Introductory. — So far we have approached the problem of protein metabolism by studying the behavior of the absorbed products of pro- tein breakdown, and we have seen that these become gradually assimilated by the tissues and used by them in their metabolic processes. "We have been unable, however, to offer any facts regarding the exact chemical changes which each amino acid undergoes during this process of tissue metabolism. At first sight it might appear an easy matter to collect such information by direct examination of the tissues themselves, either by searching in them for amino derivatives which might be derived from absorbed amino acids, or by studying the changes which occur when the amino acids are subjected to the action of the isolated tissue en- zymes that must be responsible for the change. Such methods of in- vestigation are, however, fraught with technical difficulties so great that very little can be learned from them, and for the present at least we must be content to piece our information together from facts derived by less direct methods. Such a method is offered by investigating the behavior of the end products of protein metabolism. The main end product is urea along with traces of its precursor am- monia, but these are not the only ones, for some amino acids after being incorporated with the tissue proteins break down into products that are no longer members of the amino-acid series, although they may be closely related to certain amino ocids. Such substances are creatine and its anhydrid creatinine. A part of the amino acids during their pres- ence in a free state in the blood may also be excreted unchanged by the kidney. Our list so far therefore includes urea, ammonia, creatine, creatinine, and amino nitrogen, of which the last is usually included in metabolism investigations in the fraction designated undetermined nitrogen. Another group of closely related substances coming, not from the general protein metabolism of the tissues, but from the metabolism which is peculiar to the nuclei, consists of the so-called purine bodies. Furthermore, so as to serve as a check on results obtained by examining these nitrogenous metabolites, it is important to observe the manner of 613 614 METABOLISM excretion of the sulphur moiety of the protein molecule, for it will be remembered that it is in protein alone that sulphur is usually taken into the animal body. The excretion of sulphur therefore runs more or less parallel with the intensity of protein metabolism. After selecting the end products that are most likely to be of signif- icance, the first question concerns the amount of each of them excreted during twenty-four hours on diets that are either rich or poor in pro- tein. The possibility of conducting such investigations obviously de- pends on the use of quick and yet reliable methods for the estimation of the nitrogenous metabolites. Such methods have been furnished by the painstaking and careful work of Folin, an example of whose results are given in the accompanying table. NITROGEN-RICH DIET NITROGEN-POOR DIET Volume of urine Total nitrogen Urea nitrogen Ammonia nitrogen Uric-acid nitrogen Creatinine nitrogen Undetermined nitrogen Total SO3 Inorganic SO3 Ethereal SO3 Neutral SO3 1170 C.C. 16.8 grams 14.7 grams = 0.49 gram := 0.18 gram = 0.58 gram = 0.85 gram = 3.64 grams 3.27 grams = 0.19 gram — 0.18 gram — 87.5% 3.0% 1.1% 3.6% 4.9% . 90.0% 5.2% 4.8% 385 c.c. 3.60 grams 2.20 grams 0.42 gram 0.09 gram 0.60 gram 0.27 gram 0.76 gram 0.46 gram 0.10 gram 0.20 gram = 61.7% = 11.3% = '2.5% = 17.2% = 7.3% = 60.5% = 13.2% = 26.3% The general conclusions which may be drawn from these results are as follows: 1. With a protein-rich diet much more urine is excreted in twenty- four hours than with one that is protein-poor. Evidently the nitrogenous metabolites act as diuretics. 2. The total or absolute amounts of nitrogen and of all the other nitrogenous metabolites, save creatinine, become diminished during the starvation period. The same is true of the sulphur derivatives, except in the case of the neutral sulphur, which behaves like creatinine. 3. The decrease in the portion of nitrogen excreted as urea is relatively greater than the decrease in total nitrogen, this fact being shown in the table by the percentage figures, which were secured by calculating the proportion of nitrogen in the various substances as a percentage of the total nitrogen excreted during the periods. The inorganic sul- phate behaves in a manner similar to the urea — that is, the percentage of total sulphate excreted in the inorganic form becomes much less during starvation. 4. The relative amount of all the other nitrogenous metabolites, as well as that of the ethereal and neutral sulphates, becomes increased during starvation. THE METABOLISM OF PROTEIN 615 The most striking results of the above investigation are that ereatinine remains unchanged during starvation, but that urea becomes relatively increased. The former must be derived from metabolic processes going on in the tissues independently of the supply of foodstuff carried to them, whereas the latter must depend, if not entirely, yet very largely, on the protein content of the food. Creatinine may therefore be called an end product of endogenous metabolism, and urea an end product of exogenous metabolism. Other metabolites — namely, ammonia, uric acid and the undetermined nitrogen, as well as the ethereal sulphates — must represent processes of metabolism that are partly exogenous and partly endogenous. . Having made ourselves acquainted with the general nature of the changes that occur in the nitrogenous metabolites when protein metab- olism is stimulated by the taking of food or depressed by starvation, we may now proceed to take up each of the metabolites separately and see what other information can be obtained regarding their source and origin in the animal body. UREA AND AMMONIA For various reasons it is important to consider these two metabolites together. During the intermediary metabolism of the majority of the amino acids, the amino group becomes broken off as ammonia, which immediately combines with the available acids to form neutral ammonium salts. The most available acid for this purpose is carbonic acid; there- fore ammonium carbonate is formed in large amounts. A small propor- tion of the ammonia may combine with other acid radicles, such as chlorine, to form ammonium chloride. The fate of these two types of salt is very different. The ammonium carbonate becomes quickly trans- formed into urea, whereas the ammonium chloride is excreted in the urine. The process of urea formation may therefore be considered as having the function of preventing the accumulation of ammonium car- bonate in the animal body. It is the means by which a harmful substance is converted into an innocuous substance— a detoxication process, in other words. Regarding the nature of the chemical process involved in this trans- formation of ammonium carbonate into urea, reference to the formulas below will show that the ammonium carbonate that is formed by the union of carbonic acid with ammonia, by losing one molecule of water becomes ammonium carbamate, -which by repetition of the process be- comes transformed into urea ; 616 METABOLISM OH ONH, ONH4 NH, CO + 2NH3?±CO -H2O?±CO -H2O — CO \ \ \ \ OH ONH4 NH2 NH2 (carbonic (ammo- (ammonium1 (ammonium (urea) acid) nia) carbonate) carbamate) Some of the urea may come from metabolic processes of an entirely different type. One of these at least is known ; namely, the splitting-off of urea from arginine, which it will be remembered is guanidine-amino- valerianic acid (see page 605). An enzyme called arginase, having this action, has been isolated from various organs and tissues. The diamino- valerianic acid, or ornithine, which remains after the urea is split off, may be further used in protein metabolism. The reaction is shown in the following equation: NH2 - C - NH - CH2 - CH2 - CH2 - CHNH2 - COOH + H2O ' I! NH (arginine) = NH2-CO | + NH2-CH2-CH2-CH2-CHNH2-COOH NH2 (urea)' (ornithine) On an ordinary diet, as we have seen, a man excretes somewhat more than 90 per cent of his total nitrogen as urea and about 3 per cent as ammonia, the remainder of the nitrogen appearing in the other nitrog- enous metabolites. Influence of Acidosis on Ammonia-Urea Ratio. — It sometimes happens that a large proportion of the ammonia is not converted into urea, but is used for the purpose of neutralizing abnormal acids present in the organism. "When mineral acids are given to an animal, or when acids are produced in the organism itself by some faulty type of metabolism, the ammonia excretion by the urine immediately rises. In diabetes, for example, where considerable quantities of /3-oxybutyric acid are pro- duced (see page 683), a decided increase in the ammonia excretion by the urine is observed. A milder type of acidosis may also be induced in normal persons by withholding carbohydrates from the diet, and here again the ammonia excretion is relatively increased. In such cases it is quite evident that ammonia is used as an alkaline reserve of the body; that is, as a substance which is capable of prevent- ing acidosis by neutralizing the acids. It does not appear, however, that all types of acidosis entail the utilization of ammonia as reserve alkali, and an increase in the relative amount of ammonia in the urine does not necessarily indicate a condition of acidosis. In the pernicious THE METABOLISM OF PROTEIN 617 vomiting of pregnancy, for example, a relatively high excretion of am- monia has been found associated with no greater a degree of acidosis, as determined by the power of the plasma to absorb carbonic acid, than in normal cases of pregnancy. Influence of Liver on Ammonia-Urea Ratio. — Experimental Observa- tions: (1) REMOVAL OF LIVER. — There are several facts which indicate that other causes than acid-production may interfere with the conversion of am- monia into urea. What are these causes? Since, as we have seen, the liver is the organ which most actively converts amino acids into urea, it would be natural to expect that, when the functions of this organ were interfered with, relatively more of the nitrogen excre- tion would occur as ammonia and relatively less as urea. In order to determine the exact significance of the liver as a urea-forming organ, two types of investigation' have been used; namely, (1) observation of the changes produced in- the ammonia-urea ratio in the urine by partial or total removal of the liver; and (2) observation of the urea-forming power of a liver perfused outside the body. To remove the liver from the circulation the portal vein is brought in apposition with the vena cava, the two are sewed together, and a passage opened between them, after which the portal vein is ligated above the anastomosis (forming the so-called Eck fistula). The portal blood then passes directly into the vena cava, and the liver is now supplied only by the hepatic artery. The animals live for a considerable time after the operation, and the urine frequently contains relatively less urea and more ammonia than normal. The results are, however, not nearly so striking as would be expected if the liver were the main seat of urea formation. The experiments have nevertheless brought to light a fact of considerable clinical interest — namely, although the animals may thrive if kept on a diet not containing an excess of flesh, they im- mediately begin to develop peculiar symptoms, not unlike those of ec- lampsia or uremia, when they are fed with large amounts of flesh food. Most of the symptoms can be referred to abnormal stimulation of the central nervous system, and examination of the urine has shown a large increase in the excretion of ammonia and a change from the normal acid reaction to an alkaline one. At one time it was assumed that the toxic symptoms were caused by the presence in the blood of ammonium carbamate, since large quantities of the calcium salt of this substance could be- separated from the urine. It is now known, however, that the ammonium carbamate is present only because of the excess of ammonium carbonate, the two salts existing to- gether in solution according to the laws of mass action. That the intox- ication is not due to ammonium carbamate does not exclude the pos- 618 METABOLISM sibility that it may be due to ammonia itself, although it is more likely that other nitrogenous metabolites, produced when excess of flesh food is taken, are the responsible agents. If the liver is entirely removed by ligating the hepatic arteries in an animal with an Eck fistula, a more pronounced decrease in urea and increase in ammonia occur during the short period of time that the animal survives the operation. The results observed after the removal or diminution of liver function fail to occur when other viscera are removed from the animal, which would at least tend to indicate that the liver is very important in the manufacture of urea out of ammonia. This does not, however, warrant the conclusion that the liver is the only place in the animal body in which such a process occurs. In corroboration of these observations on mammals, it may be of in- terest to note that when the liver is removed from ~birds, which is a com- paratively simple operation on account of a natural anastomosis between the portal and renal veins, there is a marked decrease in the excretion of uric acid and, a corresponding increase in the excretion of ammonia during the twelve hours or so that the birds survive. In birds and reptiles urea is excreted as uric acid, being produced by a synthetic process in the liver (see page 644). The changes in this experiment are of considerable magnitude ; thus, before the operation the amount of ammonia nitrogen relative to total nitrogen has been found to vary be- tween 10 and 18 per cent; after the operation it may be increased to between 45 and 60 per cent. The uric-acid nitrogen normally varies be- tween 60 and 70 per cent of the total nitrogen; after the operation it may fall to between 3 and 6 per cent. In animals with an Eck fistula and with the hepatic artery ligated, an increase in the urea output occurs when amino acids* are injected under the skin. This result corroborates the conclusion that the liver can not alone be responsible for the conversion of ammonia into urea. (2) PERFUSION OF ORGANS. — This method consists in removing the or- gan into a warm chamber or bath and perfusing it, through cannulse inserted in its main artery and vein, with a solution of defibrinated blood or of defibrinated blood mixed with saline solution. The perfusion liquid is kept at body temperature and is saturated with oxygen. By means of a pump it is made to circulate in a pulsatile flow, and the total amount of urea or other metabolite in the circulating fluid is determined before and after the fluid has been circulated several times through the organ. When the liver is perfused, urea gradually accumulates in the fluid, particularly after the addition of one of its known precursors — for example, ammonium carbonate. When other organs or viscera are THE METABOLISM OP PROTEIN 619 perfused, no urea is formed. The evidence shows that the liver is an important seat of urea formation, but not necessarily that other organs are unable to form it in the intact animal, for there are many sources of inaccuracy in perfusion experiments. Even though we exercise the greatest care, we can not hope to maintain the organ in other than a slowly dying condition. It is certainly far removed from the normal state, as is revealed not only by histologic examination, but by the fact that edema almost invariably sets in and the blood vessels become ex- tremely constricted, thus necessitating a gradual increase in the per- fusion pressure as the perfusion goes on. Furthermore, the organ being isolated from the nervous system, there can be no control of the rela- tive blood supply of different parts. In the intact animal the circula- tion is more or less distributed according to the particular needs of the different viscera, and such conditions obviously can not be simulated in a perfusion experiment. Another objection depends on the fact that the well-being of the organs in the intact animal is largely dependent on hormones conveyed to them from other organs. Such hormones are frequently quite labile in nature, and soon disappear from the perfusion fluid. Notwithstanding these objections, there can be no doubt that many of the functions of an organ are retained much longer than they would be if the organ were not perfused ; for example, the contractility of the muscle or the power of forming urea in the liver. Perfusion experiments are of value therefore when they yield positive results. Negative re- sults may indicate either that the organ does not perform the particular function that is being investigated or that it has lost this function as a result of partial death. That a perfused muscle retains its power of contraction does not necessarily indicate that it maintains all of its metabolic functions; neither does the fact that the liver forms urea prove that it is capable of performing its other functions. It is easy to show that the liver dies piecemeal; some functions, such as glycogen- formation, die early, while others, such as urea-formation, remain for a long time intact. The use of perfusion experiments for the settling of questions of metabolism should therefore always be very carefully con- trolled and never used as the sole line of evidence on which to base impor- tant conclusions. (3) Before leaving this subject it may be well to point out that the method which at first sight might appear to be the simplest for settling such questions — namely, the examination of the inflowing and outflowing blood of different parts or organs — is not applicable in most cases. This is be- cause of the extremely small changes in concentration which may occur even although large amounts of the particular substance in question 620 METABOLISM are being absorbed or produced. As we shall see later, this criticism is particularly applicable in the case of sugar. Even during the injection of considerable quantities of sugar into the portal vein, no difference in percentage can be demonstrated between the blood of the two sides of the liver, although we know that sugar is being retained to form glycogen. For the same reasons, differences in the percentage amounts of amino acids or of urea are often difficult to demonstrate in the blood entering and leaving the liver even when we know that large quantities of them are being added to or removed from it. Clinical. — Since the liver is an important seat of urea formation, the question arises as to whether the relative percentage of urea and am- monia in the urine will become altered by disease of the liver. Many observations with this point in view have been undertaken, but it can not be said that the results are very striking. In extreme destruction, such as that produced by phosphorus poisoning, there may indeed be a great increase in the relative amount of ammonia and a decrease in that of urea. The same is true in acute yellow atrophy of the liver, in which disease the nitrogen excreted as ammonia may amount to as much as 70 per cent of that excreted as urea. In milder forms of liver dis- turbance, however, such as cirrhosis, the figures are much less striking. When an increased ammonia excretion is observed in such cases, we must be cautious in drawing the conclusion that it is due primarily to abolition of the hepatic function. It may just as well be caused by the development of acids in the organism that require the ammonia for their neutralization. It is significant, for example, that considerable quantities of acids are produced in phosphorus poisoning. Although the urea and ammonia excretions become altered by exten- sive destruction of liver tissue, it is a remarkable fact that very little if any change occurs in the amino nitrogen, either of the urine or of the blood. In experimental necrosis of the liver produced by chloroform or by phosphorus, it is only in the latest stages of the condition and when it is of the very severest type that an amino-acid increase has been found to occur in the blood and urine. The conditions seem to be some- what different in man, abnormally high amounts of amino nitrogen hav- ing been observed in the blood in a considerable proportion of patients with impaired liver function. In very severe cases of diabetes, for ex- ample, figures that are distinctly higher than normal have been observed (Van Slyke, etc.). In eclampsia the marked pathologic changes in the liver might be expected to be associated with an upset in the metabo- lism of amino acids. Losee and Van Slyke35 have, however, recently shown by the most accurate methods that neither in the blood nor in the urine is any excess of amino acids to be found in this condition, although THE METABOLISM OF PROTEIN 621 in cases of pernicious vomiting of pregnancy, there was a relative in- crease in the ammonia excretion. We have already seen that this increase did not bear any relationship to the acid-absorbing power of the blood plasma (see page 617). The importance of the kidneys in removing the urea from the ~bldod is readily seen from the change in the percentage of urea in this fluid after the partial or complete removal of the kidneys. Animals sur- vive nephrectomy for about three days, and during this time urea rapidly accumulates in the blood and begins to make its appearance in the saliva and the intestinal secretions. In man also where the kidneys are extensively diseased, a similar accumulation of urea occurs in the blood, some of the excess being got rid of through the sweat and to a certain extent through the intestine. The importance of encouraging perspiration and a free movement of the bowels in cases of nephritis is thus indicated. It must not be concluded that the accumulation of urea in the organism is the direct cause of the symptoms. Urea itself is comparatively inert, and it is generally believed that other metabolic products with which the urea runs parallel in amount are the toxic agents. Hewlett has found, however, that very large injections of urea do produce symptoms in animals.34 CHAPTER LXX THE METABOLISM OF PROTEIN (Cont'd) CREATINE AND CREATININE Creatine and creatinine are very largely products of endogenous metab- olism ; they are mainly derived from chemical processes occurring in the tissues although some of the creatine and creatinine present in the food may appear as creatine in the urine. Essential Chemical Facts Before we proceed further with a discussion of the metabolism of these important substances, it will be necessary to refer briefly to some points in their chemistry. The simpler of the two bodies is creatine, which is methyl-guanidine-acetic acid; creatinine is its anhydrid, being formed from creatine by the removal of a molecule of water, so that the NH2 groups become joined together in the same way as they do in the formation of pep tides from amino acids (page 599). The relationships are illustrated in the following formulas: (methyl) CH3— N / \ / CH..COOH NH — C - H20 — \ (acetic acid) (guanidine) NH2 (creatine) CH3-N-CH-CO I NH = C \ \ I NH (creatinine) It should be noted that guanidine is closely related to urea NH2 / (0=0 ), and that when creatinine is formed from creatine a ring NH, THE METABOLISM OF "pROTEUST 623 formation occurs, giving what may be regarded as an imidazole deriva- tive (see page 604). Creatine is also related to one of the important diamino acids, arginine, since both contain guanidine radicles, NH2 (NH=C ), and to histidine and the purines (see page 634), both NH2 of which contain the imidazole ring. The close relationship which 'creatine bears to urea is illustrated by the fact that urea is formed when creatine is subjected to- the action of boiling barium hydrate. When it is oxidized by means of potassium permanganate, urea is also formed, the remainder of the molecule, more or less intact, being split off as NH-CH3 / methyl-ammo-acetic acid (CH2 ), also known as sarcosine. COOH The conversion of creatine to creatinine goes on slowly in aqueous solutions, but is much accelerated by heating with acid. Heated in an autoclave at a temperature of 117° C. for thirty minutes, with half nor- mal hydrochloric acid, the creatine goes over almost quantitatively into creatinine. It will be noted that the creatinine ring is partly oxidized. This renders it unstable, so that creatinine in the presence of alkalies has the power of reducing metallic oxides. Like glucose it can reduce alkaline solutions of copper, silver and mercuric salts; it also reduces picric acid in weakly alkaline solution to picramic acid, which, being red, furnishes us with a solution the strength of which can be estimated colorimetrically. Quantitative Estimation. — Although the presence of creatinine in the urine has been known for many years, there being from 1 to 2 grams of it in the twenty-four-hour urine, little progress was made in the study of its metabolism because of the absence of a reliable method for its estimation. The elaboration by Folin of a colorimetric quantitative method for creatinine, depending on the reduction of picric acid, has furnished the starting point for the modern work which has been done. To estimate the creatine by this method, it is usual to proceed as fol- lows: The creatinine content is first of all determined, another portion of urine being then heated with acid in the autoclave until all of its creatine has been converted into creatinine. A second determination of creatinine is then made, and the difference between the two is calculated as creatine. 624 ' METABOLISM It should be pointed out that, since the creatine is estimated by an indirect method, there are considerable chances for inaccuracy. Indeed, it has been shown that errors may have been incurred in some of the recent work on account of the fact that when acetoacetic acid is present in the urine it prevents the creatinine from developing its full reducing power on picric acid in the cold, so that when subsequently the urine is heated with acid for the purpose of converting the creatine into creati- nine, the destruction of acetoacetic acid allows the reducing power of the creatinine to develop to full intensity. It is obvious that this would make it appear as if creatine had been converted into creatinine. It is par- ticularly in the urine of diabetic patients, in which acetoacetic acid is present that mistakes are likely to be made. Metabolism When we come to consider the metabolism of creatine and creatinine, we find that there are remarkably few facts definitely known concerning it. The average amount excreted daily, expressed as the number of milli- grams of creatinine in twenty-four hours per kilogram body weight, is known as the creatinine coefficient (Shaffer).36 For a lean person this is about 25 mg. ; for a corpulent person, about 20 mg., the difference in- dicating that muscle mass, and not body weight, is the important factor determining the coefficient. Further evidence that this relationship ex- ists is furnished by the fact that in the muscular atrophies creatine ex- cretion is distinctly below normal. It must be the mass of the muscles rather than their activities that is the determining factor, for the creatine excretion does not become increased by muscular exercise. Influence of Food, Age, and Sex. — Although creatine and creatinine are endogenous metabolites, it must be remembered that, under ordinary dietetic conditions, a part of each is derived from these substances pres: ent in the food. It is important therefore to consider the conditions under which the creatine and creatinine in the food appear in the urine. Regarding creatinine, it is pretty well established that practically all that is taken with the food reappears as creatinine in the urine. Shaffer has, for example, succeeded in recovering 76 per cent of ingested creat- inine in the urine excreted during twenty-one hours following the in- gestion of 0.7 gm. creatinine. The conditions for the excretion of creatine are more complex. It is present in the urine of children in considerable amount, but in that of adults only as traces. In the first years of life the creatine in boys' urine may amount to one-half of the total creatine and creatinine, but it becomes gradually less and practically disappears at about seven THE METABOLISM OF PROTEIN 625 years of age. Girls, on the other hand, continue to excrete creatine until about puberty, after which, although ordinarily absent, it reappears in the urine at each monthly sexual cycle, and is present during pregnancy and for some days after delivery. Feeding creatine to children causes it to appear in the urine, accompanied usually by a slight increase in the creatinine. The same results can be observed in women during the monthly periods, when as much as 0.1 gm. may be present, and during pregnancy. Creatine is also present in the urine of most if not all of the other mammalia. Some of these facts are shown in the following table : AGE CREATININE-N CREATINE-N EXCRETED IN 24-HR. URINE 2 0.025 0.023 -3 0.057 0.022 Boys 5 8 0.112 0.163 0.025 0.0 11 0.157 0.0 15 0.378 0.0 5 0.069 0.005 6 0.032 0.003 Girls 7 0.157 0.066 10 0.147 0.020 12 0.201 0.011 (From Mathews.) When creatine is given to an animal that has been kept in a starved condition, most of it seems to disappear. It can not be recovered in the urine either as creatine or as any other nitrogenous metabolite. It seems to functionate more as a food than as a useless substance. The possi- bility that some of it can be destroyed by the intestinal bacteria being admitted, there is nevertheless some justification for the view that the creatine finds a useful function in the anabolic process of the muscles. Influence of Complete and Partial Starvation. — Although, as we have seen, the creatinine excretion remains constant when the amount of pro- tein in the diet is greatly reduced, yet it does not remain constant during complete fasting or when carbohydrates are entirely withheld from the diet. In fasting it has been found that creatine appears in place of the creatinine which has disappeared, so that if both creatine and creatinine are determined, very little if any diminution will be found to have oc- curred. Fasting, therefore, causes the adult creatine and creatinine metabolism to become like the juvenile metabolism. As pointed out by Mathews, it would be interesting in the light of this observation to see whether other substances, passed in the urine of young animals but ab- sent in that of the adult, would reappear in the urine when the animals were made to fast. In the case of man, for instance, allantoin would be worth investigating in this regard (page 641). 626 METABOLISM A similar replacement of some of the creatinine by creatine appears when carbohydrate is entirely withheld from the diet, or in diabetic animals, either in the disease diabetes mellitus in man or in the experi- mental condition induced in animals by giving phlorhizin. Unfortu- nately, in a considerable part of the work that has been done on this phase of the subject a method of estimation was employed which did not take sufficiently into account the influence of acetoacetic acid on the creatine estimation; but even after allowing for this possible source of error, there can be no doubt that creatine appears in the urine when carbohydrates are improperly metabolized. If carbohydrates are given to a starving animal, for example, the creatine is replaced in its urine by creatinine, although this will not occur when either protein or fat is fed. The general conclusion which may be drawn from these observations is that carbohydrates in some way are required for the proper conversion of creatine into creatinine in the animal body (Cathcart)37. Origin of Creatine and Creatinine Notwithstanding the amount of excellent work that has recently been done on the metabolism of creatine and creatinine, w,e know very little indeed regarding the origin of these bodies in the animal organism. It would be profitless to discuss this problem to any great extent, but a few of the most important facts so far established may be of interest and of value. The first step in attacking such a problem is to compare the amounts present in the various organs and tissues, in the blood, and in the excreta. Of the approximately 12D grams of creatine and creatinine in the body of an average adult, a very large proportion is in the muscles, the voluntary muscles containing the largest percentage, 'the heart con- taining a medium percentage, and the involuntary (intestinal) muscles containing relatively a small amount (Myers and Fine)38. Next to the skeletal muscles, and containing more than the involuntary mus- cles, come the testis and brain. The liver, pancreas, thyroid, kidneys, spleen, etc., contain traces, the smallest amount of all being found in the blood. In all these places by far the greatest proportion of the total creatine- creatinine exists as creatine, which is exactly the reverse of the condi- tion obtaining in the urine of adults, where practically all is excreted as creatinine. The close chemical relationship between creatine and creat- inine, considered along with the above facts regarding their quantitative distribution in the body, indicates that the creatinine of the urine is de- rived from the creatine of the tissues. The question is, How does the creatine come to be converted into creatinine? Such a transformation is THE METABOLISM OF PROTEIN 627 probably effected by many of the tissues of the body and certainly by the blood, the active agency in all cases being no doubt an enzyme. That the blood contains such an enzyme is indicated by the fact that creatine is transformed to creatinine by blood serum more quickly than it is when merely dissolved in water. Even heated blood serum possesses some of this power. The liver also probably brings about the transfor- mation, as has been shown by perfusion experiments, and by the fact that in cases of phosphorus or hydrazine poisoning creatine displaces creatinine in the urine. The problem therefore narrows itself down to the question of the origin of creatine. In the light of chemical knowledge there are several precursors from which creatine might be formed. One, for example, is arginine, which it will be remembered is guanidine-amino-valerianic acid (see page 605). By oxidation this might become changed into guani- dine-amino-acetic acid, which by methylation would then be changed into creatine. That such a process of methylation may actually occur in the animal body is definitely known, for it happens when such substances as pyridine or naphthalene are given with the food. They appear in the urine as methyl derivatives. The possibility of the derivation of creatine from arginine is not, however, borne out by the result of the injection of arginine, for such injection does not increase the creatinine in the urine. The closely related substance, guanidine-acetic acid, when fed to animals (rabbits) does cause a slight increase in the excretion of creatine (Jaffe), and also, it is said, an increase in the creatine content of the muscle. Even in this case, however, by far the largest proportion of the admin- istered guanidine-acetic acid is excreted in the urine unchanged. The large percentage of creatine in muscle tissue leads one to expect that some relationship must exist between muscular metabolism and the amount of creatine present either as such in the muscles or as creatinine in the urine. Regarding the latter point it is definitely established that muscular exercise leads to no increase in the creatinine excretion, al- though it is said that an increase occurs following a tonic contraction of the muscles. With regard to the creatinine in the muscles, no definite results indicating that muscular metabolism changes its amount are on record. In the light of the fact already stated regarding the presence of creatine in other organs than the muscles, it seems probable that the substance has really little to do with muscular contraction as such, but rather is concerned in some way in the formative metabolism of the cell, with its general growth or maintenance. Indeed, it is a question whether creatine is an actual constituent of the living tissue. It may rather, as has been suggested by Folin, be a postmortem product, represented dur- ing life by creatinine. 628 METABOLISM Creatine appears in the urine in phosphorus poisoning, in carcinoma of the liver and during postpartum involution of the uterus. It is not de- rived from the disappearing uterine muscle, however, for creatinuria also occurs after cesarean section with removal of the uterus. Creatine elimination is not an index of cellular destruction, for it has been found large in a dog injected with phlorhizin and maintained in constant weight by feeding with washed meat (S. E. Benedict). Muscular fatigue also leaves the creatine content of muscle unchanged. In late stages of nephritis, creatinine accumulates in the blood and serves as an index of the gravity of the condition (page 651). CHAPTER LXXI THE METABOLISM OF PROTEIN (Cont'd) UNDETERMINED NITROGEN AND DETOXICATION COMPOUNDS In the present chapter we shall refer briefly to the groups of urinary substances styled undetermined nitrogenous compounds and to the com- pounds that are excreted in the urine as the result of the combination in the body of certain toxic bodies with chemical substances that render them harmless (detoxication compounds). Undetermined Nitrogen Included under undetermined nitrogen are amino acids, peptides and basic substances. The amount of amino acids and peptides in normal urine is very small but may become considerable in disease, especially of the liver, when leucine and tyrosine may appear. The presence of traces of amino acid and peptone in normal urine is to be expected, for although the actual concentration of amino acids .in the blood is never very great, a certain leakage of amino acids must occur into the urine. The peptide is sometimes known as oxyproteic acid. It becomes dis- tinctly increased in phosphorus poisoning and in such conditions as are accompanied by excessive protein metabolism. The basic constituents include such substances as trimethylamine, ethylamine, putrescine and cadaverine (page 502), and there are probably many more of a similar nature. Many of these substances are similar to the so-called ptomaines found in meat, etc., and they have been called the ptomaines of urine, from which they can be isolated by rendering the urine alkaline and shaking out with ether. It is probably to the presence of these sub- stances that urine mainly owes its toxic action. The Detoxication Compounds Certain nocuous substances are produced in the intestine during the digestive process (see page 501), and others may result from the meta- bolic processes in the tissues. To guard against the harmful action of these substances on the organism, they become detoxicated in various 629 630 METABOLISM ways, mainly by forming inert compounds with other substances, par- ticularly with glycocoll, sulphuric acid or glycuronic acid. The com- pound thus formed is then excreted in the urine. Hippuric Acid. — Glycocoll is used mainly to detoxicate the benzoic acid which results from the oxidation of the aromatic substances pres- ent in large quantities in vegetable food and fruit (particularly in cran- berries). Some benzoic acid may also be produced by the breakdown of the aromatic group of the protein molecule; phenylalanine, for ex- ample, gives rise to benzoic acid by bacterial decomposition. The com- pound formed is hippuric acid, this name indicating that it is present in large quantities in the urine of the horse, as it is also in the urine of all herbivorous animals. Hippuric acid is benzoyl-glycine (C(iH5.CO.NH.CH2COOH), and it can readily be produced in the laboratory by bringing together benzoyl chloride with glycocoll, thus: C6H5 . CO i Cl + H~ i HN . CH2COOH = CtH5CO ..NH . CH2COOH + HC1. (benzoyl chloride) (glycocoll) (hippuric acid) Under ordinary dietetic conditions only a trace of hippuric acid is present in the urine of man, but much larger quantities, 2 grams a day for example, may appear when the diet contains a large proportion of fruit or vegetables. It is not known to undergo any characteristic varia- tions in disease. The benzoic acid which is contained in certain canned foods as preservative also combines in the body with glycocoll, so that any toxic effect which it might produce is practically negligible. There is certainly no very evident reason why canned foods containing benzoic acid should be tabooed, for in so far as the berzoic acid is concerned, they can be no more toxic than a diet composed largely of vegetables and fruit. This detoxication of benzoic acid requires the presence in the organ- ism of a constant supply of glycocoll, which, it will be recalled, is the lowest in the series of amino acids, being aminoacetic acid (CH,NH2COOH). It is present in greatest amount in the protein of the connective tissues. It is said, however, that not more than from 2 to 3.5 per cent of glycocoll is available in the proteins of the body. Al- though this amount of glycocoll would amply suffice to detoxicate the benzoic acid produced by the metabolism of the food in carnivora, it is quite inadequate for this purpose in the case of herbivora, and the question naturally presents itself as to where the glycocoll in these animals comes from. It is said, for example, that of the total nitrogen excretion in herbivora 50 per cent may appear as glycocoll under cer- tain conditions. These facts indicate that the organism is capable of THE METABOLISM OF PROTEIN 631 producing new glycocoll for itself, and it is interesting to consider how this glycocoll may be derived. A very probable source is by synthesis between ammonia and glyoxylic acid (CHO. COOH). That glyoxylic acid or. its aldehyde, glyoxal, is readily produced during metabolism from car- bohydrates and that ammonia is always available would seem to lend some support to this view (see page 665). 'The synthesis of glycocoll from glyoxal and ammonia occurs thus: H.COCHO + NH3 = CH2NH.,COOH. (glyoxal) (glycocoll) The linking up of glycocoll with benzoic acid occurs in the kidney. If the kidney is removed from the circulation in the majority of animals that produce hippuric acid in large amount — the rabbit being a.n excep- tion— no hippuric acid will accumulate in the blood. On the other hand, an isolated perfused preparation of the kidney produces hippuric acid provided benzoic acid is added to the perfusion fluid, and the latter also contains an abundance of oxygen, which is best secured by using de- fibrinated arterialized blood instead of artificial serum (Locke's solu- tion). The necessity of a plentiful supply of oxygen is further shown by the fact that, if the hemoglobin of the blood is rendered incapable of carrying 02 by bubbling carbon monoxide gas through it, no synthe- sis of hippuric acid will result from perfusing the blood through the kidney. The actual chemical process by which the synthesis occurs (de- .hydration) is similar to that by which polypeptides are formed by the union of amino acids, or ereatinine from creatine. (C6H5CO IOH + H; HNCH2COOH). Glycocoll may be used for detoxicating other substances than benzoic acid, particularly cholic acid, forming the glycocholic acid of the bile (see page 494) and phenylaeetic acid. In birds the benzoic acid be- comes combined with diamino-valerianic acid or ornithine (NH2-CH2- CH2 - CH2 - CH - NH2 - COOH) in place of glycocoll, so that in the urine of these animals in place of hippuric acid a compound called ornithuric acid occurs. It is of importance to point out here that this pairing of aromatic toxic substances with certain of the metabolic products of the organism has frequently been found an excellent experimental method for demon- strating the presence of intermediary metabolic substances that other- wise \vould not have appeared in the excreta. These substances are thus diverted from their normal course in metabolism so as to form neutralization or detoxication compounds. Glycuronic acid is an example. 632 METABOLISM Ethereal Sulphates and Glycuronates. — The other substances used for detoxication purposes are sulphuric and glycuronic acids. Phenol, and its derivative cresol, after being absorbed from the intestine, in the contents of which they are produced by the bacterial decomposition. of protein (see page 501) become combined in the body, probably in the liver, with sulphuric acid or with glycuronic acid to form the sulphate or glycuronate. The aromatic sulphate further combines with potassium to form the so-called ethereal sulphates, ias which the substance is excreted in the urine. A small amount of phenol may however appear in the urine unchanged. As we have already seen, the sources of the phenol in the intestine are tyrosine and phenylalanine (see page 530), and since these amino acids are also present in the tissues, it might be sup- posed that some of the phenol sulphate of potassium present in the urine could come from the tissues. It is usually assumed that, however, derivation from the tissues does not occur. Another ethereal sulphate is indoxyl sulphate of potassium, which re- sults from the absorption into the blood of the indole and skatole pro- duced by intestinal putrefaction from tryptophane (see page 502). Immediately after absorption indole is- oxidized to indoxyl, which then combines with sulphuric acid and with potassium to form indoxyl sul- phate of potassium, which is the well-known indican of the urine. As in the case of phenol sulphate of potassium, none of the urinary indican seems to come from the normal metabolism (of the tryptophane) of the tissue proteins. It is a much more reliable indicator than phenol sul- phate of potassium of the extent of intestinal putrefaction, but it also becomes increased in amount during putrefaction in the body itself, as for example in abscess formation. The amount of indican in the urine may be roughly gauged by oxi- dizing the urine by means of hypochlorite and then shaking out with chloroform. If the resulting extract is more than light blue in color, it indicates excessive putrefaction. A negative test does not neces- sarily mean that intestinal putrefaction is absent, but a marked positive test always indicates that it is occurring. Skatole, the methyl deriva- tive of indole, may undergo similar processes and appear in the urine during excessive intestinal putrefaction. Its presence in the blood some- times confers on the breath a distinct fecal odor, for this body, as its name indicates, is that to which the odor of the feces is due. Glycuronic acid, the other substance used for detoxication processes, is of the nature of a dextrose molecule with the one end-group oxidized to carboxyl (CHO - (CHOH)4 - COOH). It is probably produced under normal processes of metabolism in the animal body, but is destroyed unless when such poisonous substances as camphor, chloral hydrate or THE METABOLISM OF PROTEIN 633 certain aromatic alcohols are given, when it is used for the purpose of detoxicating them. The resulting glycuronates have reducing powers and may be confused with glucose when present in large amount. Gly- curonates may be distinguished from glucose in the urine (1) because they are levorotatory, and (2) because they do not ferment. The free acid itself, however, is dextrorotatory. CHAPTER LXXII Introductory. — The participation by highly trained organic chemists in the investigation of biochemical problems has brought our knowledge of the history of the purine substances in the animal body from a state of chaos and guesswork to one of system and scientific accuracy. The peculiar solubility reactions of uric acid and its salts and the discovery of urates in gouty deposits served to make uric acid metabolism one of the earliest research problems in both the medical clinic and the bio- chemical laboratory, but the earlier results were practically valueless, partly because they were inaccurate and partly because their interpretation was impossible in the absence of even the most elementary facts concerning the chemistry of uric acid. Before any real progress was possible, a clean sweep had to be made of all the old speculations and hypotheses, such as that dignified by the high-sounding name of "uric-acid diathesis," and a foundation of ac- curate chemical knowledge established. This foundation is now wonder- fully complete, and a superstructure of biochemical fact is already beginning to grow upon it. In the present chapter we shall examine some of the most important contributions that have made this progress possible. As in the study of any other problem of metabolism, we must, however, make ourselves familiar with the main facts concerning the chemistry of the purine bodies and of the tissue constituents into the composition of which they enter, before proceeding to the more strictly biological aspect of the subject. The Chemical Nature of the Purines By an examination of the empiric formulas of the purines of biochem- ical interest, it will be observed that they are all derivatives of a sub- stance purine, which although in itself of no importance is interesting, since it serves as the basic substance from which the others are derived. The list is as follows: Purine . C5H4N4 Hypoxanthine C5H4N4O Monoxypurine 1 Adenine . C5H3N4.NH2 Amino-purine I Purine Xanthine . C5H4N4O2 Dioxypurine I bases. Guanine . C5H3N4O.NH2 Amino-oxypurine J Uric acid . C5H4N4O3 Trioxypurine 634 URIC ACID AND THE PURINE BODIES 635 The first oxidation product of purine is hypoxanthine, which has long been known as a constituent of meat extract. Adenine, the amino derivative of hypoxanthine, occurs in combination with other substances in the nuclear material. The second oxidation product is xanthine and its amino derivative, guanine. They occur in the same places as hypo- xanthine and adenine. The highest oxidation product of all is the well- known urinary constituent, uric acid, which may therefore be chemically designated as trioxypurine. In addition to the purines of animal origin, there are also certain ones of vegetable origin — the methyl purines, which exist as the alkaloids of tea and coffee — namely, caffeine, theobromine, and theine. To understand the chemical structure of this group of substances, it is perhaps simplest to start with that of uric acid. This consists essentially of two urea molecules linked together by a central chain of three carbon atoms, as will be evident from the accompanying structural formula : HN-CO OC C-NH ! I! \ . co I II / HN-C-NH (urea) (urea) \ (central chain) This structure can be shown by methods both of decomposition and of synthesis. When uric acid is decomposed by oxidizing it with nitric acid, it yields urea and a residue called alloxan ; or it can be synthesized from urea and trichlorlactamide, a derivative of lactic acid, which it will be remembered contains three carbon atoms. The changes involved in this synthesis will be made clear by examination of the accompanying structural formula, in which the manner of production of the by- products of the reaction (NH3, H20 and HC1) are shown by dotted lines: NH. ! H NH.. ; - C = O / I CO \ H J-C ! OH H ! NH Cl i || CO \ C- i~~Cl~H~~i NH . (urea) (urea) NH. j ~ II Cl j ( trichlorlactamide ) 636 METABOLISM By milder oxidation by means of potassium permanganate in the cold, uric acid becomes quantitatively converted to allantoin: C5H4N403 + H20 + 0 = C4H6N403 + C02. (uric acid) (allantoin) The importance of this transformation lies in the fact that in most animals, man and the higher apes being exceptions, uric acid is thus decomposed in the animal body. The structural formulas for the other purine bodies in relationship with those of purine and uric acid are given below. Purine itself has the following structural formula: IN Ce H H2 - » - NH7 •\ 08- H sN — C4 - NS (For convenience of description the atoms in purine are numbered as shown.) HN-C=O H- C C-NH HN-C=O O = C C-NH \ / C-H N - C - N (hypoxanthine) (6-oxypurine) N = C-NH, \ HN - C - N (xan thine) C-H (2-6-oxypurine) H- C C-NH \ H,N = HN-C=O I- NH C-H -C- N \ N-C- N C-H (adenine) (6-amino-purine) (guanine) (2-amino-6-oxypurine) HN-CO Jc ' C-NH CO HN- C -NH Uric acid (2-6-8-trioxypurine) The substances with which the purine bases are most closely related are the pyrimidine bases. Three of these are known: thymine (NH-CQ cytosine ( N = C-NH2 and uracil (NH-CO I ! I II CO C.CH8 CO CH CO CH NH-CH ); NH-CH); NH-CH). URIC ACID AND THE PURINE BODIES 637 From an examination of the structural formulas, it will be seen that they are more or less related to purine (having one of the urea radicles omitted), although it can scarcely be doubted that they exist as separate constituents of the nucleic acid group in the animal body, and are not derived from purine. They are primary products. The Chemical Nature of the Substances in Which Purine and Pyrimidine Bases Exist in the Animal Body. — In general it may be said that the amino purines — adenine and guanine — together with the pyrimidine bases — thymine and cytosine — occur combined with phos- phoric acid and a carbohydrate in the various nucleic acids, each of which is again combined with some simple protein to form nuclein, the essen- tial constituent of the chromatin of the nucleus. One of the oxypurines, hypoxanthine, may also exist combined with phosphoric acid and carbo- hydrate to form a substance present in muscle and known as inosinic acid. The general scheme of construction of a nucleic acid of animal origin is illustrated in the following formula suggested by Levene and Jacobs :39 HO O = PO — C6H]004 — C5H4N50 / (hexose) (guanine group) O HO | \ O = PO — C6H8O2 — C5H5N202 / | (hexose) (thymine group) HO O HO \ O — PO — C6HSO2 — C4H4N3O / | (hexose) (cytosine group) HO O Phosphoric acid \ groups O = PO — C6H10O4 — C5H4N5 / (hexose) (adenine group) HC/ According to this formula nucleic acid may be considered as a com- pound of polyphosphoric acid, containing carbohydrate groups, which serve to link the phosphoric acid molecules to those of purine or pyrimi- dine. In nucleic acids of animal origin, such as the example given above, the carbohydrate is a hexose, (i.e., contains 6 C-atoms), whereas 638 METABOLISM in those of plants (e. g., yeast), it is a pentose (5 0-atoms). It has been found necessary to introduce some terms to designate the different parts of the nucleic acid molecule ; thus, the whole molecule is called a tetra- nucleotide, each mononucleotide molecule of which is composed of a phosphoric acid molecule plus a nucleoside, which again is composed of a purine or pyrimidine nucleus attached to pentose or hexose. The nucleoside is so named because it is similar in structure to a glucoside. Apart from differences in the carbohydrate group, it appears that there is a close similarity in the structures of nucleic acids from dif- ferent cells. This would indicate a common function for them all, which may be either of a skeletal or of a physiologic nature; that is, nucleic acid may have to do with the sustentacular material that builds the nucleus, or it may have to do with some physiologic function common to all cells, such as irritability, or growth, or respiration. If nucleic acid is merely a sustentacular material, then the study of the behavior of chromosomes and chromatine in cells can not have the significance that it would have were nucleic acid concerned in the more vital activ- ities of the nucleus. All the so-called nuclear stains owe their specific staining properties to the fact that they are of a basic nature and com- bine with nucleic acid. Until we know more definitely what the exact function of nucleic acid may be, it is unwise to place too much weight on the behavior of the chromosomes in cytologic researches. The History of Nucleic Acid in the Animal Body. — We shall first of all study the manner in which nucleic acid may be broken down. As is to be expected from its complex structure, various types of enzymes are concerned in this process. The first to act are known as the nucle- ases. They split the tetranucleotide molecule into two dinucleotides, which immediately afterward split further into mononucleotides. Four nucleotides, two of purine and two of pyrimidine, are thus formed from each molecule of nucleic acid. Each nucleotide molecule may now un- dergo decomposition in one of two ways: (1) either by the splitting off of phosphoric acid, leaving a nucleoside (guanosine or adenosine), or (2) by the splitting off of both phosphoric acid and carbohydrate, leaving free purine bases. Nucleuses have been found which specifically effect either of these decompositions, and they have been called phospho- nucleases* (1), and purine-nucleases (2), respectively. In the decompo- sition of nucleic acid all of the four purine compounds — guanine, guano- sine, adenosine and adenine — may be formed. This is illustrated in the accompanying schema, in which the nucleic acid is represented as a purine nucleotide: •The numbers refer to the enzymes indicated in the schema. URIC ACID AND THE PURINE BODIES 639 NueteicAcid (without the pyrimidine group) / tf (Action of nucleuses) d ' \ GuanineAdenine euses) \ \ \ d en o sine 8— >Ad (4) (5) (6) (Action of deamvnizing enzymes') >Jr ss- 60 tr TO rs so tr to is- /oo Fig. 189. — Curves showing the percentage of glucose in blood after a constant injection of an 18 per cent solution into a mesenteric vein. V.C., vena cava, continuous line; P.O., pan- creaticoduodenal vein, broken line; /, iliac, dotted line. It will be observed that, so far as can be judged from changes in the concentration of sugar in the blood, the sugar-retaining power of the liver is about equal to that of the muscles — a conclusion which is, how- ever, contrary to the usually accepted one that the liver has such pro- nounced sugar-retaining powers that under ordinary circumstances it removes from the portal blood all the excess of sugar added to it by absorption and which is not required by the organism. One objection which may properly be made to these observations is that the animals on which they were made were under anesthesia, and that the anesthetic had a paralyzing effect on the sugar-retaining power of the liver. In view of this criticism it is important to examine the results obtained on animals that are not under the influence of anes- THE METABOLISM OF THE CARBOHYDRATES 659 thesia. Such observations have been made on rabbits, and a few on man himself. By collecting blood from the ear veins of rabbits, it has been found that, after giving from two to ten grams of glucose by stomach, the glucose concentration of the systemic blood begins to rise in fifteen minutes, attaining a maximum in about an hour and then returning to the normal level in about three hours. Similar results have been obtained by examination of the venous blood in man. After giving 100 grams of glucose by mouth, for example, there is commonly an increase in blood sugar amounting to from 30 to 34 per cent of the normal and lasting for from one to four hours. The existence of this postprandial hyperglycemia, as we may call it, indicates that the sugar-retaining powers of the liver and muscles are not sufficiently de- veloped to prevent the accumulation of some of the absorbed sugar in the systemic blood. Whenever this increase exceeds a certain limit, some of the sugar begins to escape through the kidney into the urine, producing glycosuria — postprandial glycosuria. The percentage of blood sugar above which glycosuria occurs is, in the case of man, probably about 0.10 to 0.11 gm. per cent. After damage to the kidney, as in nephritis, or in long-stand- ing cases of mild diabetes, the percentage may probably rise considerably higher in the blood without evidence of glycosuria. Value of Blood Examination in Diagnosis of Diabetes. — The determina- tion of the amount of ingested carbohydrate required to bring about post- prandial glycosuria constitutes, as we have already seen, the so-called assimilation limit for sugar, which is often taken as an index of the sugar- metabolizing power of the organism. It is evident, however, that the time of onset, and the extent and duration of postprandial hyperglycemia must serve as a more certain index of the sugar-retaining power of the liver and muscles; and now that a simple and rapid clinical method exists (Lewis-Benedict method) for the accurate determination of sugar in small quantities of blood, there is no reason why this index should not be used for the detection of failing powers to metabolize carbohydrate. In no disease, probably not even in tuberculosis, is it more important than in diabetes that an early diagnosis should be made. Thus, if we find that the postprandial hyperglycemia after a certain amount of carbo- hydrate develops to an unusually high degree and persists for an unusual length of time, we are justified in curtailing the carbohydrate supply so as to hold these values down to their level in normal individuals. It is almost certain that the first sign of diabetes is an unusual degree and duration of postprandial hyperglycemia. At first the excess of sugar leads to no damage and it is insufficient to cause any evident glycosuria, although it is quite likely that if the urine in such individuals were collected at very frequent intervals after eating carbohydrate-rich food, glucose would be 660 METABOLISM found present in at least some of the specimens. In incipient diabetes, however, the condition progresses, until the postprandial hyperglycemia after one meal has not become entirely replaced before the next is taken, so that the increase in sugar produced by the second meal becomes super- added on that following the first meal. The curve of blood sugar rises ever higher and higher, until at last permanent hyperglycemia is estab- lished, or rather the normal level from which the postprandial rise occurs has become permanently raised, so that in blood collected at any time a higher percentage of sugar is found. The Relationship Between the Sugar Concentration of the Blood and the Occurrence of Glycosuria. — Claude Bernard first pointed out that the percentage of sugar in the blood may rise considerably above its normal level without the appearance of any of the sugar in the urine, or at least without a sufficient amount to give the usual tests for sugar. Even when this limit is reached, as we have seen, the sugar which appears is not all of the excess but only a small part of it. This overflow hypothesis, as it is called, has not been universally accepted because of the many results which are not in conformity with it. Many of these exceptional results have been explained as due to alterations in the permeability of the kidney for sugar, and in general it is probably safe to accept Claude Bernard's hypothesis with certain reservations. Strong support has been lent to a modified form of the hypothesis by the recent work of Woodyatt and his collaborators, who have shown by continuous intravenous glucose injections that as much as 0.8 gm. of glucose per kilo body weight can be injected during an hour into an animal without any glycosuria, although under such conditions a very distinct increase occurs in the percentage of sugar in the blood. To explain the failure of glucose to pass into the urine under normal conditions, it has been supposed by several investigators that the glucose exists in some form of chemical combination in the blood. This compound is believed to behave like a colloid. One of the recent supporters of this view is Allen, who has observed that, when glucose is injected intrave- nously, it causes diuresis as well as glycosuria; whereas glucose injected subcutaneously or taken by mouth causes neither of these conditions to become developed ; indeed it causes for some time after the administration of the sugar a distinct anuria. To explain these differences in behavior between glucose administered intravenously and that taken in other ways, it is supposed that the glucose molecule in passing through the intervening wall of the capillaries combines with some substance to form a compound which becomes available for incorporation into and utilization by the tissues, glucose in a free state being incapable of utilization. This com- pound is supposed to be of a colloidal nature, and the substance which THE METABOLISM OF THE CARBOHYDRATES 661 combines with glucose to form it is believed to be related to the internal secretion of the pancreas (see page 676). The difficulty in explaining why the glucose of the blood does not con- stantly leak into the kidney is, however, the only evidence upon which the hypothesis of a blood sugar compound rests. No chemical evidence can be offered in support of such a view. On the contrary, all experimental work indicates that the sugar exists in a free state; but unfortunately even this evidence is not convincing. Thus, it has been found that, when speci- mens of perfectly fresh blood are placed in a series of dialyzer sacs sus- pended in isotonic saline solutions, each solution containing a slightly dif- ferent percentage of glucose, diffusion of glucose, in one or other direction, occurs in all of them save one — namely, that in which the percentage of glucose in the fluid outside the dialyzer is exactly equal to the total sugar content of the blood. Such a result can be explained only by assuming that all of the sugar in the blood exists in a freely diffusible state. In its general nature this experiment is analogous to that by which the tension or partial pressure of C02 is determined in blood (see page 338). It has been assumed by many clinicians that glycosuria may sometimes become developed because the kidney fails to hold back the blood sugar even when the percentage is not above the normal — so-called renal dia- betes. For the diagnosis of this condition a comparison must be made be- tween the sugar concentration of the blood and that of the urine. In order to do this at least two samples of blood must be taken, one of them at the beginning and the other at the end of a period during which urine is being collected. Merely to find that one sample of blood collected before or after or during the period of urine collection contains a normal percentage of sugar, does not necessarily indicate that at some other period while the urine was being produced a temporary hyperglycemia may not have ex- isted. CHAPTER LXXV THE METABOLISM OF THE CARBOHYDRATES (Cont'd) FATE OF ABSORBED GLUCOSE. GLUCONEOGENESIS We may now consider what becomes of the sugar that is retained by the liver and muscles. Two things may happen to it: It may become stored, or it may become oxidized or split up. Of these processes, storage occurs in both the liver and muscles, whereas oxidation occurs mainly if not entirely in the muscles, although a certain amount of splitting of the glucose molecule may also occur in the liver. Storage of Sugar. — For the present we shall consider the process of storage of sugar and defer a consideration of its utilization until after we have studied, not only the nature of the process by which the storage occurs, but also the immediate destiny of the stored sugar. The storage of sugar by the liver is brought about by its conversion into a polysac- charide called glycogen. After an animal has been absorbing large quan- tities of glucose, an acidified watery extract of a portion of liver mad? immediately after death will be found to contain no more sugar than that of a normal liver. On the other hand, it will be observed that the extract is highly opalescent and yields on the addition of alcohol a copious precip- itate, which on further purification can readily be shown to consist of a polysaccharide— that is to say, of a starch-like substance which on hydrol- ysis with mineral acid becomes entirely converted into sugar. If instead of removing the liver immediately after death, it is allowed to stand for some time, the yield of glycogen will greatly diminish, and in its place will appear large quantities of glucose, indicating that some enzyme must exist which attacks the glycogen after death and converts it into sugar This enzyme is called glycogenase. The existence of postmortem glyco- genolysis, as it is called, would seem to indicate that during life also there is a constant tendency for the glycogen in the liver to be attacked by glycogenase, but that this is prevented by conditions which depend on the vital integrity of the liver cell. It is evident that if anything should happen during life to interfere with this inhibiting influence, the glycogen will become converted into glucose, which on escaping into the blood will produce hyperglycemia and glycosuria. Sources of Glycogen. — In studying the source of sugar in the animal body it is important therefore that we should first of all know exactly the 662 THE METABOLISM OP THE CARBOHYDRATES 663 conditions under which glycogen may be formed in the liver; that is, whether it is formed from absorbed sugar alone, or from other substances also, such as protein and fat. The importance of such knowledge rests in the fact that in severe diabetes, sugar still continues to be added to the blood, although no sugar is being taken with the food. To check the hyperglycemia in such cases it becomes necessary, therefore, to curtail the diet not only with regard to its carbohydrate content, but also with regard to whatever other foodstuff may be capable of causing glycogen formation. The practical question therefore is, "What are these foodstuffs? There are two methods by which the problem may be investigated. The first, which we may call the direct method, consists in rendering the liver free of glycogen and then some time afterward feeding the animal with the foodstuff in question, afterward killing it and examining the liver for glycogen. The other, which we may call the indirect method, con- sists in first of all rendering the animal incapable of oxidizing glucose — that is, making it diabetic — and then proceeding to see whether the in- gestion of a given foodstuff causes an increase in the sugar excretion in the urine. The methods for rendering an animal experimentally diabetic will be considered later ; for the present it is important to note that, if a diabetic animal excretes more glucose while fed on a given foodstuff, we may infer that the normal animal would convert it into glycogen. The results of the direct method are much less reliable than those of the indirect for the reason that it is extremely difficult to remove all traces of glycogen from the liver. The methods employed for this pur- pose have consisted in: (1) starvation of the animal; (2) muscular ex- ercise; (3) exercise and starvation combined; and (4) the production of certain forms of experimental diabetes — for example, that produced by phlorhizin. Starvation alone is unsatisfactory, for it has been found that, although at certain stages of this condition the liver may become al- most entirely free from any trace of glycogen, at a later stage glycogen may again make its appearance. It is therefore most difficult to decide at what stage in starvation the animal should be considered as glycogen- free. If the starving animal is made to perform muscular exercise, complete removal of glycogen from the liver can be depended upon. The exercise may be produced by the administration of strychnine in such dosage as just to produce convulsions of the voluntary muscles without permanent contraction of those of respiration. The most useful method, however, consists in starving the animal for a few days and then placing it in a cold, damp room, after giving it a cold bath. The evaporation of mois- ture from the surface so cools the body down that the glycogen store all becomes used up in the attempt to supply fuel for the production of 664 METABOLISM sufficient heat to maintain the body temperature. This method can he rendered still more certain in effecting a removal of all carbohydrate from the body by giving the animal phlorhizin every eight hours. This drug, as we shall see, renders the animal diabetic. After removing the glycogen, further deposition in the liver can be readily shown to occur when any of the ordinary sugars or starches are given as food. It does not occur, however, when chemical substances closely related to ordinary sugar, such as the wood sugars (pentoses) or the alcohols and acids corresponding to dextrose, are contained in the diet. Nor does it occur with cellulose or with inulin, a polysaccharide built up from pentose sugar. When proteins are fed the results are not so definite, although many observers have claimed that glycogen is formed. With fat, on the other hand, no glycogen formation can be shown to occur, although we know that a trace of carbohydrate must be formed out of the glycerine of the fat molecule. The results of the direct method, even when the conditions are per- fectly controlled, are very unreliable, because any new sugar produced by the ingested substance instead of being stored as glycogen may be directly used by the tissues as it is formed. Where only a slight degree of gluconeogenesis is occurring, it is not likely that any of the glucose will be retained in the body as glycogen. The methods employed for producing experimental diabetes in investi- gation of these problems by the indirect method are (1) the entire removal of the pancreas, and (2) the continuous administration of the drug phlorhizin. The animal rendered diabetic by either of these methods is first of all observed for several days to determine the normal daily ex- cretion of sugar. At the same time the nitrogen excretion for the day is determined, the ratio between the total nitrogen and the glucose being known as G to N ratio, and being about 1 to 3.65 when complete diabetes has become established. The foodstuff in question is then fed to the animal, and the amount of extra glucose excreted thereby is taken to represent that which has been derived from the ingested food. By this method it has been possible to show that, not only the above mentioned carbohydrates, but protein as well produce a very considerable quan- tity of glucose in the animal body. Fats, however, yield only negative results. The indirect method has another great advantage over the direct in that the results are much more quantitative in character; for example, Lusk and his pupils have been able to determine the amount of glucose which can be produced by feeding certain of the building stones of the protein molecule. The great practical importance of such results in THE METABOLISM OP THE CARBOHYDRATES 665 the therapy of diabetes makes it advisable for us to go into the subject a little more in detail here. Dogs are rendered diabetic by phlorhizin followed by a cold bath and exposure in a cold room. When all of the original glycogen in the body has been got rid of, as evidenced by the constancy of the G to N ratio in the daily quantities of urine excreted, the substance under in- vestigation is fed. If this substance contains no nitrogen and causes no change in the nitrogen excretion, any increase in that of glucose must obviously represent the extent to -which the substance has become con- verted into this sugar. On the other hand, if the substance itself con- tains nitrogen, or if it causes a change in the excretion of nitrogen, it becomes necessary to calculate how much of the excreted glucose might have been derived from the body protein, assuming that this can form glucose, and how much from the administered substance.* From the results of this method it has been an easy matter to show that the following substances are converted in the animal body into glucose: (1) Glycol aldehyde (CH2OH-CHO). By placing three mol- ecules of this substance together, a hexose molecule will be produced, a . synthesis which can be accomplished in the chemical laboratory. Glycol aldehyde may be formed in normal metabolism out of glycocoll (CH2NH2COOH). (2) Glycerol (CH2OH - CHOH - CH2OH) may also readily be con- verted into hexose in the laboratory, the possible intermediary products being dioxyacetone (CH2OH - CO - CH2OH) and gly eerie aldehyde (CH2OH-CHOH-CHO). Two molecules of either of these may be polymerized to form a hexose molecule, and when this process occurs in the animal body, the hexose formed is glucose. (3) Lactic acid (CH3CHOH - COOH) is completely converted to dex- trose in the diabetic animal, and the process must involve both a re- arrangement of the molecule and subsequent polymerization. The related substance, propyl alcohol (CH3- CH2-CH2OH) is also converted into glucose in the phlorhizinized dog. As to the exact nature of the chemical changes which occur as intermediary steps in the conversion of these substances into glucose, we are not as yet certain, but a step has been made in the discovery that a substance called methylglyoxal (CH3COGHO) may be obtained from lactic acid and also from glucose, and that this substance is converted into glucose when it is administered to phlorhi- zinized dogs. We shall find later an important role for this substance 'This calculation is made as follows: The amount of nitrogen in the administered substance is deducted from the nitrogen excretion, and the difference, which must represent the nitrogen of the body protein, is multiplied by the G to N ratio which prevailed on the day previous to that on which the substance was fed. We obtain in this way the glucose derived from the body. The glucose coming from the administered substance can then be ascertained by deducting that derived from the body protein from the total glucose excretion. 666 METABOLISM in the case of fat metabolism, and it can also readily be produced during the intermediary breakdown of certain of the protein building-stones, such for example as alanine (CH3CHNH2COOH). These chemical possibilities regarding the nature of the substances that serve as stepping stones between the above sugar-forming sub- stances and sugar itself may be translated later into probabilities on account of the discovery that the enzymes exist in various tissues which are involved in converting methylglyoxal into lactic acid: CH3 CH3 ! I CO + H2 -+ HCOH I O <- | CHO COOH (methylglyoxal) (lactic acid) These enzymes are called glyoxalases, and since the reactions which they mediate are undoubtedly reversible in character, it is probable that the conversion into sugar of lactic acid and alanine — to take those two as among the commonest of the sugar precursors of the animal body — occurs according to the following equation: CH3CHNH2COOH v (alarinc) CILCOCHO -H» C6H]2O6 CH,CHOHCOOH /* (lactic acid) (methylglyoxal) (hexose) The unique position of methylglyoxal, besides explaining the known resolutions of protein and fat and carbohydrate in intermediary metab- olism, is also of importance in explaining the synthetic production of glucose from fructose (or levulose). Fructose will first of all become converted into methylglyoxal radicles, and these will then become syn- thesized into glucose. The theory of the conversion of glucose into lactic acid as a stepping stone in the metabolism of carbohydrate meets with one objection; namely, that the lactic acid is not produced from carbohydrate in the organism, except in cases where there is oxygen deficiency or excess of alkali in the tissue fluids. Coming now to the amino acids, which, it will be remembered repre- sent the building stones of the protein molecule, it has been found that glycocoll, alanine, and aspartic and glutamic acids. increase the glucose excretion when given to phlorhizinized dogs, whereas leucine and tyro- sine have no such action. By the method described above, it is possible to determine the exact proportion of the carbon of each of those amino acids which becomes converted to glucose. This is shown in the accom- panying table. THE METABOLISM OF THE CARBOHYDRATES 667 TWENTY GRAMS OF THE VARIOUS AMINO BODIES WERE GIVEN TO PHLORHIZIN-DIABETIC DOGS ACID AND FORMULA AVERAGE AMOUNT OF GLUCOSE PRO- DUCED IN BODY PROBABLE GLUCOSE THAT CHANGE WOULD BE PRO- DUCED BY CHANGE Glycocoll CH2NH2COOH i. alanine CH3CHNH2COOH Aspartic acid COOH— CH2—CHNH,— COOH 13.43 one 18.77 12.42 (five dogs, gave 15.77) (tw.o dogs) (four dogs) All C converted to glucose it Three of the four C atoms converted 16.00 20.22 13.52 Glutamic acid COOH CH, 13.31 to glucose Three of the five C atoms converted to glucose CH2— CHNH2 COOH 12.24 It is of further interest to point out that these four amino acids constitute about 26 per cent of all the amino acids in flesh protein, and that the total yield of glucose from them could be 26.3 grams; thus accounting for nearly one half of the 66 grams which a diabetic animal produces from 100 grams of flesh. Gluconeogenesis in Normal Animals. — Although it has been clearly shown by the indirect method that not only protein but its decomposi- tion products as well, can be readily converted into glucose, yet this does not necessarily indicate that a similar conversion occurs in the nondia- betic animal. That such is the case, however, can be shown in various ways. Thus, at the end of a period of long starvation considerable quantities of glycogen are quite commonly found in the body, and the blood sugar, although lower than normal, never entirely disappears. Now, since no carbohydrate is being ingested, and the body stores of this foodstuff become exhausted early during starvation (cf. page 663), it is evident that the carbohydrate must be produced from the protein of the animal's body. A still more convincing experiment can be con- ducted by producing strychnine convulsions in a starving animal. If the animal is killed after the convulsions have lasted for a certain time, the tissues will be found almost if not entirely free of glycogen, but if the convulsions are made to disappear by giving chloral and the animal allowed to sleep for some time before killing it, glycogen again accumulates in the body. This glycogen must have been manufactured out of noncarbohydrate material. Corroborative evidence of a somewhat different nature is furnished by 668 METABOLISM an examination of the respiratory quotient, which, it will be remem- bered (page 54.7), varies according to the nature of the foodstuff or body constituent that is undergoing metabolism at the time, being about 1 with carbohydrate and about 0.7 with protein. If the quotient is observed during starvation, it will often be found to fall below 0.7, a figure which can be explained only by assuming that oxygen has been retained in the body beyond the quantity which is necessary for imme- diate purposes of oxidation (cf. equations on page 548). Since it is known that this retained oxygen can not exist in the body in a free state it must be concluded that it has become incorporated into substances having a high oxygen content. Such would be the case if protein or fat, which contains only from 12 to 20 per cent of oxygen, were converted to carbohydrate, ' which contains about 53 per cent. Utilization of inhaled oxygen for this purpose, as we have seen, becomes very striking in the case of hibernating animals during the winter sleep. CHAPTER LXXVI THE METABOLISM OF THE CARBOHYDRATES (Cont'd) FATE OF GLYCOGEN Having become familiar with the sources from which glycogen may be derived, we may now proceed to study the fate of the glycogen found in the liver cells and in the muscles. For the present we shall confine our attention to the glycogen of the liver. If a portion of liver removed from 'a well-fed animal is examined microscopically after staining either with iodine or with carmine by Best's method, it will be found that the cells of the lobules are filled with glycogen except for the nuclei, which are free from this substance. If, on the other hand, the liver is from an animal that has not been recently fed, the lobules will contain no glyco- gen except for an area bordering on the central vein and perhaps a narrow strip at the periphery of the lobule. When it is present the rela- tive amount of glycogen in different lobules, as determined chemically, is the same over the entire liver— that is to say, no one lobe is richer in this substance than another. Nothing definite is known as to how the glycogen is held in the protoplasm of the cells, although some histolo- gists suggest that it is combined with a sustentacular material especially provided for this purpose. The glycogen stored in the liver is gradually given up to the blood of the hepatic vein at such a rate as to maintain in the blood of the sys- temic circulation a more or less constant percentage of glucose. Under ordinary conditions this process of glycogenolysis is relatively slow, but when the requirements of the organism for fuel become increased, as during muscular exercise, it becomes very rapid. The glycogenic func- tion of the liver appears therefore to exist, in part at least, for the purpose of preventing the flooding of the blood of the systemic circu- lation with excess of sugar during absorption from the intestine and of maintaining the normal percentage at other times. This function is analogous to that occurring in plants, in which the sugar produced in the leaves, if not immediately required, is transported to various parts of the plant and there converted into starch, which, when the plant requires it, as during new growth, may again become transformed into glucose. The agency converting the glycogen into glucose is the diastatic 669 670 METABOLISM enzyme glycogenase, which is present, not only in the liver cell, but also in the blood and lymph. If is a difficult matter to explain why glycogen should be able to exist at all in the liver cells in the presence of this powerful enzyme. The following possibilities may be considered: (1) That glycogenase does not really exist in the living liver cells, but is a postmortem product; (2) that, although present, glycogenase is pre- vented from acting on the glycogen in the living liver cell on account of the latter being protected from its influence by combination with the sustentacular substance; or (3) that some chemical substance in the liver cell prevents the glycogenase from acting on the glycogen — an anti- glycogenase. Since the removal of any one of these inhibiting influ- ences would cause glycogenolysis to become excessive, and so bring about hyperglycemia, it is important, in searching for the possible causes of this condition, to examine the evidence that has been brought forward in support of each of these views. Against the view that glycogenase is a postmortem product may be cited the very rapid conversion into glucose that occurs when glycogen is added to living blood, as by injecting some into a vein. On account of the active glycogenolytic action of blood, it has been suggested that during life glycogen does not become transformed into glucose until after it has been discharged into the blood from the liver cell. When increased sugar must be mobilized, glycogen passes unchanged, or per- haps as seme dextrine, into the blood and lymph of the liver capillaries and lymphatics, the glycogenase of which converts it into glucose, the conversion being so rapid that, by the time the blood has traveled from the liver through the heart and pulmonary vessels to the arteries, all the glycogen has already become transformed into glucose. Postmortem glycogenolysis, according to this view, is due to the opposite occur- rence— the transference of glycogenase from the blood into the liver cell. Some facts supporting this view are as follows: (1) It has been found that the amount of free glucose in the blood of the vena cava is sometimes less than in that collected simultaneously from the carotid artery. (2) After giving certain substances, such as phosphorus or peptone, there is distinct diminution in the amount of glycogen in the liver, accompanied, however, by no increase in the amount of glucose in the blood. And (3) if the liver of an animal that has been rendered diabetic by stimulation of the splanchnic nerve or by puncture of the floor of the fourth ventricle is examined microscopically, after staining by the carmine method, masses of stained glycogen can be found present in the capillaries (sinusoids) that lie among the liver cells. According to the second view, the glycogen is removed from the influence of the intrahepatic glycogenase on account of its combination THE METABOLISM OF THE CARBOHYDRATES 671 with a sustentacular material. By disrupting this combination and thus exposing the glycogen to the action of glycogenase, glycogenolysis will occur. We may call this the mechanical hypothesis and it deserves serious consideration, for it has been shown that very little postmortem glycogenolysis occurs in the intact liver of frogs in winter, — even though at this time the organ contains an excess of glycogen, — but becomes marked when the liver is broken down by mechanical means. The third view depends on the well-known fact that enzyme activities become most markedly altered by slight changes in the chemical nature of the environment in which they act. Diastatic enzymes are partic- ularly susceptible to the reaction (CH) of their environment, a very slight degree of acidity favoring and a trace of alkalinity markedly depressing their activities. That a tendency to increasing acidity in the liver cells may retard the formation of glycogen is suggested by the depressing effect produced on the assimilation limit of sugars by administering acids, and by the observation that postmortem glycogen- olysis becomes marked in proportion as the dying liver becomes acid in reaction. It might be thought then that glycogenolysis in the liver cell could be set up by the local production of a certain amount of acid. Such a liberation of free acid could be brought about by a curtailment in the arterial blood supply of the hepatic cell, producing a local accu- mulation either of carbonic or of other less completely oxidized acids (e.g., lactic). It may be that asphyxia causes hyperglycemia by such a mechanism. Vasoconstriction and consequent curtailment of arterial blood supply occurs in the liver when the hepatic nerves are stimulated, and it is possible that the glycogenolysis which is also set up by such stimulation is due to the appearance of acids. The accelerating effect of epinephrine on glycogenolysis might also be explained as due to limitation of blood supply on account of vasoconstriction. THE REGULATION OF THE BLOOD SUGAR LEVEL The level at which the concentration of .sugar in the systemic blood is maintained represents the balance between two opposing factors: (1) the consumption of glucose bv the tissues, and (2) the production of glucose by the liver. Since this is the most readily oxidizable of all the proximate principles of food (page 652), muscular activity causes large quantities of it to be consumed, so that its concentration m the blood tends to fall below the physiologic level, a tendency which is immediately met by an increased discharge of glucose from the liver. The question therefore arises as to Iww the muscles or other tissues transmit their requirements for glucose to the liver. There are two 672 METABOLISM possible ways by which this could be done: (1) by means of a nervous reflex, or (2) by changes in the composition of the blood, either with regard to the percentage of sugar itself or because of the appearance in it of decomposition products of glucose or of some special exciting agent or hormone. In order to ascertain the relative importance of these methods of correlation between the places of supply and demand of glucose in the normal animal, it is necessary to investigate the conditions under which an excessive discharge of glucose occurs as a result of overstimulation of the nervous control, or because of the presence of exciting substances (hormones) in the blood. The glycogenic function can be excited through the nervous system in a variety of ways so as to produce hyperglycemia and glycosuria. This constitutes one form of experimental diabetes. In laboratory animals mechanical irritation of the medulla oblongata and stimulation of the great splanchnic nerves act in this way. Similar stimula- tion may also occur under certain conditions in man. Excitation as a result of changes in the composition of the blood can be produced experimen- tally by certain drugs (phlorhizin), or by the removal of certain of the ductless glands or the injection of extracts prepared from them, such as epinephrine. Nerve Control and the Nervous Forms of Experimental Diabetes. — The simplest experimental condition which illustrates the relationship between the nervous system and the blood sugar is electrical stimulation of the great splanchnic nerve in animals in which, by previous feeding with carbohydrates, a large amount of glycogen has been deposited in the liver. By examination of the blood as it is discharged into the vena cava from the hepatic veins, the increase in blood sugar is very evident in from five to ten minutes after the first application of the stimulus; but it is not until later that a general hyperglycemia becomes estab- lished. The conclusion which we may draw from these results is that the splanchnic nerve contains efferent fibers controlling the rate at which glycogen becomes converted to glucose in the liver. The center from which these fibers originate is situated somewhere in the medulla oblongata, for the irritation that is set up by puncturing this portion of the nervous system with a needle yields results similar to those which follow splanchnic stimulation. This "glycogenic" or diabetic center, as it has been called, must be provided with afferent impulses. Such im- pulses have indeed been described in the vagus nerves, but their dem- onstration is by no means an easy matter on account of the disturbance in the respiratory movements coincidently produced by the stimulation. The changes that such disturbances bring about in the aeration of the THE METABOLISM OP THE CARBOHYDRATES 673 blood may in themselves be responsible for the hyperglycemia (see page 332). It can at least be said that when the respiratory disturbances are guarded against, as by intratracheal insufflation of oxygen, vagal hyper- glycemia is much less marked, if not entirely absent. But this question awaits more thorough investigation. The increased glycogenolysis which results from stimulation of the efferent fibers in the splanchnic nerves may depend either on a direct control exercised over the glycogenic functions of the hepatic cells, or on the discharge into the blood of some hormone which excites the glycogenolytic process. It must furthermore not be lost sight of that the glycogenolysis may be secondary to local asphyxial conditions in the liver cells resulting from vasoconstriction. From their anatomic position, the adrenals are to be thought of as the source of the hormone, and evidence that splanchnic hyperglycemia isi due to hypersecretion from these glands has seemed to be furnished by the fact that after they are extirpated splanchnic stimulation no longer produces hyperglycemia, neither, indeed, does puncture of the medulla. There is also no doubt that the nervous system, acting by way of the splanchnic nerves, does exercise a control over the discharge of the internal secretion of the adrenal glands and that extracts of the gland, which we must suppose act in the same way as the internal secretion, cause hyperglycemia when injected intravenously (epinephine hyperglycemia and glycosuria). But on theoretical grounds alone, certain difficulties immediately pre- sent themselves in accepting this as the mechanism by which the nervous system controls the sugar output of the liver, for if increased sugar formation in the liver is dependent on a discharge of epinephrine, the question may be asked why this secretion should be caused to traverse the entire circulation before reaching the liver. There are, besides, certain experimental facts which do not conform with such a view. Thus, after complete severance of the hepatic plexus of nerves, stimulation of the splanchnic nerve does not cause the usual degree of hyperglycemia, whereas electric stimulation of the peripheral en'd of the cut plexus does cause it. On the one hand, therefore, there is evidence that stimulation of the efferent nerve path above the level of the adrenals has no effect on the sugar production of the liver in the absence of these glands; and on the other, we see that when they are present, stimulation of the nerve supply of the liver is effective, even though the point of stimulation is beyond them. There is but one con- clusion that we may draw — namely, that the functional integrity of the efferent nerve-fibers that control the glycogenolytic process of the liver depends on the presence of the adrenals, very probably because of the hormone which the glands secrete into the blood. This conclusion is 674 METABOLISM corroborated by the fact that stimulation of the hepatic plexus, even with a strong electric current, some time after complete removal of both adrenals, is not followed by the usual degree of excitement of the glycogenolytic process. These experiments demonstrate an important relationship between the nervous control, and at least one form of hormone control, of the sugar output of the liver. They indicate that when a sudden increase of blood sugar is required, the glycogenic center sends out impulses which not only directly excite the breakdown of glycogen in the he- patic cells, but also simultaneously influence the adrenals in such a man- ner as to produce more epinephrine in the blood and so augment the ac- tion of the nerve impulse. We are as yet quite in the dark as to the mechanism by which the nerve impulses or the hormone brings about increased glycogenolysis. It must consist of a removal of the influence that prevents glycogenolysis from occurring in the normal liver, for it has been shown by direct ob- servation that there is no increase in the amount of glycogenase present in extracts of the liver removed from diabetic animals over that present in extracts of the liver of normal animals. The possible nature of this influence has already been discussed (page 669). The change may con- sist either in a loosening of the combination between the glycogen and the protoplasm of the liver cell, or in a removal of the chemical influence that ordinarily prevents the glycogenase from attacking the glycogen. In the former case the glycogen liberated from its union with the sus- tentacular substances would either become attacked by the glycogenase present in the liver cell itself or it would first of all migrate, as glyco- gen, into the blood capillaries and there be attacked by the blood glycogenase. Evidence for the possibility of the occurrence of such a process has already been given (page 670). The chemical change re- ferred to under the second possibility might consist in an alteration in the hydrogen-ion concentration of the liver cell, a change, however, which for obvious reasons it is impossible to investigate. Nervous Diabetes in Man. — The main interest attaching to the inves- tigation- of these nervous forms of experimental diabetes depends on the insight which they afford us into the nature of the mechanism by which a prompt mobilization of glucose may be brought about in the normal animal. There is also some evidence that a relationship may exist be- tween certain of the clinical varieties of the disease in man and repeated excitation of glycogenolysis brought about by nerve stimulation. In- creased glucose output from the liver as a result of nerve excitation may be a normal process, but there is reason to believe that frequent repetition of this process tends to induce a permanent rise in the glucose THE METABOLISM OF THE CARBOHYDRATES 675 level of the blood and therefore a tendency to diabetes. There have recently been collected several facts which lend some support to this view. The frequent occurrence of diabetes in those predisposed by inheritance to neurotic conditions, or in those whose daily habits entail much nerve strain, and the aggravation of the symptoms which is likely to follow when a diabetic patient experiences some nervous shock, all point in this direction. Diabetes is common in locomotive engineers and in the captains of ocean liners — that is, in men who in the performance of their daily duties are frequently put under a severe nerve strain. It is apparently in- creasing in men engaged in occupations that demand mental concentra- tion and strain, such as in professional and business work. Cannon23 found glycosuria in four out of nine students after a severe examination, but only in one of them after an easier examination.* In the urine of twenty-four members of a famous football squad, sugar was found pres- ent in twelve immediately after a keenly contested game. Anxiety and excitement were responsible for its appearance, for five of the twelve were substitutes who did not get into the game. Although these nervous conditions, by excitement of hepatic glyco- genolysis, produce at first nothing more than an excessive discharge of sugar into the blood — a condition which is exactly duplicated in our laboratory experiments by stimulation of the nerve supply of the liver — their repetition may gradually lead to the development of a permanent form of hyperglycemia. To prevent the repetition of these transient hyperglycemias must be one of our aims in the treatment of early stages of the disease. Although there can be no doubt that the glycogenic function of the liver is subject to nerve control, it is probable that its control by hor- mones is of equal if not greater importance. This dual control of a glandular mechanism is by no means unique for the glycogenic function, for we have already seen it to exist in the case of the gastric glands and the pancreas, and it is probable that it also exists in the case of the thyroid. It may well be that the nerve control of the glycogenic function has to do only with those transitory changes in sugar produc- tion that would be demanded by sudden activities of muscle, and that the hormone control has to do with the more permanent process of build- ing up and breaking down of glycogen to meet the general metabolic requirements of the tissues. *We have been unable to confirm this observation even though the examinations were made unusually "nerve-racking." 676 ' METABOLISM HORMONE CONTROL AND PERMANENT DIABETES Nervous excitation can explain only transitory increases in blood sugar, the more permanent hyperglycemias being dependent upon some dis- turbance in the hormone control of carbohydrate utilization. This dis- turbance is a much more serious affair than that produced by nervous excitation. In the latter case the hyperglycemia ceases whenever all of the glycogen stores of the liver have been exhausted; whereas a dis- turbance in the hormone control, besides causing as its first step a breakdown of all the available glycogen, goes on to cause a production of sugar out of protein. A process of gluconeogenesis (new formation of glucose) becomes superadded on one of glycogenolysis. To ascertain the nature of this hormone and the mechanism of its action has been the object of most of the researches on those forms of diabetes produced by changes in certain of the ductless glands. The following possibilities may be considered: (1) that it is the concentration of glucose in the blood ; (2) that it is the presence in the blood of decom- position products of glucose; (3) that it is due to a hormone produced from some ductless gland. Regarding the first of these possibilities, it is supposed that the mechanism involved in the adjustment between the blood sugar and the glycogenic function is one explicable on the basis of the law of mass action; namely, that glycogen becomes converted into glucose whenever the blood flowing to the liver contains less than its normal concentration of glucose, and conversely, when this blood contains an excess of glucose, as during absorption, a glycogen-building process takes place. Although there can be little doubt that the process of glycogen formation or destruction will depend to a certain extent upon the amount of; glucose present in the blood flowing to the liver cells, yet it is impossible that this can be an important means in the control that exists between sugar production by the liver and sugar consumption by the tissues, because the sugar that is added to the portal blood during absorption would mask any depletion caused by sugar consumption in the -tissues. The second possibility — that the hormone is some decomposition prod- uct of glucose — would appear to have some support, if we consider this hormone to be an acid product (carbon dioxide or lactic) produced by sugar metabolism, for it is known that an increase in the hydrogen-ion concentration of the blood flowing to the liver cells excites a glycogen- olysis.. As we have already seen, however, it is difficult to secure ex- perimental evidence, in anesthetized animals at least, that glycogen- olytic activity is readily excited in this way. The third possibility — that some specific hormone may exist in the THE METABOLISM OF THE CARBOHYDRATES 677 blood exciting the glycogenolytic process — is investigated by producing disturbances involving various of the ductless glands, particularly the pancreas, the adrenals, the parathyroids and the pituitary. The influ- ence of certain, of these glands may be closely bound up with that exercised through the nervous control, as we have seen to be the case with the adrenal gland. Whether it is by the production of hormones directly necessary for proper carbohydrate metabolism, or by the re- moval from the blood of such substances as interfere with this process, that the ductless glands functionate, is one of the main problems we have to consider. Utilization of Glucose in Tissues. — Although the experimental diabetes induced by disturbances in the function of the ductless glands is at first dependent on an upset of the glycogenic function and later of glu- coneogenesis, the utilization of glucose in the tissues ultimately becomes interfered with. It is therefore important that we should digress for a moment to consider briefly what is knoAvn regarding the process by which sugar becomes utilized in the organism. That glucose becomes used up by active muscle there can be no doubt. Thus, if the muscles of one leg in the frog are tetanized, the glycogen content, compared with that of the other leg, will be found to be diminished. At first sight it migh.t appear that the easiest way to study the utiliza- tion of glucose in the muscles Avould be to compare its concentrations in the blood flowing to and coming from the muscle. The muscle that has been most successfully employed in studies of this kind has been the heart. Some years ago Starling and Knowlton24 examined the consump- tion of sugar by the excised mammalian heart, and in their earlier experiments seemed to be able to show that the extent to which this consumption occurred was 4 milligrams per 100 grams heart muscle per hour. A more thorough repetition of these experiments later by Pat- terson and Starling25 showed, however, that the results can furnish no criterion of the actual consumption of glucose by the tissue on account of the fact that the tissue itself may store away large quantities of carbohydrate in an unused state — i. e., as glycogen. Other investigators have thought to study the utilization of glucose by observing the rate at which it disappears from drawn blood kept in a sterile condition at body temperature for some hours after death. This process is called glycolysis, and it has been assumed that the process is similar to that which occurs in the tissues themselves — an assumption, however, for Avhich there is no warranty. Indeed, it may readily be shown that the glycolysis occurring in blood has very little if anything to do with the utilization of sugar in the tissues, for it has been found that glucose disappears from drawn blood very slowly indeed when 678 METABOLISM compared with the rate at which it disappears from the blood of animals in which the addition of glucose from the liver has been prevented by removal of this viscus (Macleod).26 A third method for studying the utilization of glucose consists in observing the respiratory exchange of animals. In normal animals the injection of glucose causes an increase in the carbon-dioxide excretion and a rise in the respiratory quotient, which it will be remembered is a ratio expressing the relationship between the amount of carbon dioxide excreted and of the oxygen retained in the organism. When carbohy- drate is undergoing combustion, the quotient is nearly 1, whereas with that of protein it is about 0.7 (see page 547). By observing the quotient under given conditions one can compute the proportions of carbohydrate and of fat and protein that are undergoing metabolism. In the hands of Murlin and others,27 this method has proved of some value in settling certain questions concerning the utilization of glucose in normal and diabetic animals ; but the results must be interpreted with great care on account of the fact that temporary changes in the blood may cause a greater or a less expulsion of carbon dioxide from it. Thus, if acids appear in the blood, they will dislodge carbon dioxide from it, and apparently cause the respiratory quotient to rise. Alkalies, on the other hand, apparently cause it temporarily to fall, and unless the observa- tions are done over a long period of time and with great care, faulty conclusions are very apt to be drawn from the results. Diabetes and the Ductless Glands We are now in a position to consider the forms of experimental dia- betes produced by disturbances in the ductless glands. Relationship of the Pancreas to Sugar Metabolism. — In no other of the many causes of diabetes has greater interest been shown than in that due to disturbance in the pancreatic function. Many of the earlier clinicians that followed cases of diabetes mellitus into the postmortem room, noted that definite morbid changes in the pancreas were a fre- quent accompaniment of the disease. Prompted by these observations, several investigators attempted experimental extirpation of the gland, but did not succeed in producing glycosuria in the few animals that survived the operation. Their failure, no doubt, resulted from incom- plete extirpation. To reduce the severity of the operation, Claude Ber- nard injected oil into the pancreatic duct, and tied it; but he succeeded in keeping only two dogs alive for any length of time, and these did not exhibit glycosuria. Neither were other investigators that adopted similar methods any more successful. It looked as if the pancreas had very little to do with the cause of diabetes. In the year 1889 Minkowski THE METABOLISM OF THE CARBOHYDRATES 679 and von Mering in Germany, and de Dominicis in Italy, by thorough extirpation of the gland, succeeded in producing in dogs a marked and persistent glycosuria, accompanied by many of the other symptoms of diabetes. The first two authors attributed the condition to removal of an internal secretion. The course of the diabetes thus produced is, however, somewhat differ- ent from that usually observed in man. It is extremely acute from the start, the G: N ratio being 1 :3.6 (see page 664), and it is unaccompanied by any of the classical symptoms seen in the clinical condition. Experi- mental pancreatic diabetes can, however, be made to simulate very closely the disease in man. This was first of all demonstrated by Sandemeyer, who found that if the greater part of the pancreas was removed, the animals for some months, if at all, were only occasionally glycosuric, but later became more and more frequently so, until at last the condition typical of complete pancreatectomy supervened. Similar results have more recently been obtained by Thiroloix and Jacob, in France, and by Allen in this country. These investigators point out that different re- sults are to be expected according to whether the portion of pancreas which is left does, or does not, remain in connection with the duodenal duct. When this duct is ligated, atrophy of any remnant of pancreas that is left is bound to occur, and this is associated with rapid emacia- tion of the animal, diabetes and death. When the remnant surrounds a still patent duct, a condition much more closely simulating diabetes in man is likely to become developed — one, namely, in wMch there is, for some months following the operation, a more or less mild diabetes, which, however, usually passes later into the fatal type. It is, of course, difficult to state accurately what proportion of the pancreas must be left in order that the above described condition may super- vene. Leaving a remnant amounting to from one-fifth to one-eighth of the entire gland is commonly followed by a mild diabetes, whereas if only one-ninth or less is left, a rapidly fatal type develops. As in clinical experience, the distinguishing feature between the mild and the severe types of experimental pancreatic diabetes is the tolerance toward carbohydrates. In the mild form, no glycosuria develops unless carbo- hydrate food is taken; in the severe form, it is present when the diet is composed entirely of flesh. It is thus shown that "by removal of a suitable proportion of the pancreas, it is possible to bring an animal to the verge of diabetes, yet to know that the animal will never of itself become diabetic. . . . Such animals, therefore, constitute valuable test objects for judging the effects of various agencies with respect to diabetes" — (Allen18). It therefore becomes theoretically possible to in- vestigate, on the one hand, other conditions which will have an influence 680 METABOLISM similar to removal of more of the gland, or, on the other, conditions which might prevent the incidence of diabetes, even though this extra portion of pancreas is removed. From the work which he has already done, Allen believes that he has sufficient evidence to show that the continued feeding with excess of carbohydrate food will surely convert a mild into a severe case, and in one experiment he succeeded in bringing about the same transition by performing puncture of the medulla — that is, by creating an irritative nervous lesion. By none of the other means usually employed to produce experimental glycosuria could the bordering case be made diabetic, although one such animal became acutely diabetic after ligature of the portal vein. To the clinical worker the value of these results lies in the fact that they furnish experimental proof that a so-called latent case of diabetes — that is, one that has a low tolerance value for carbohy- drates— may be prevented from developing into a severe case by proper control of the diet. Attempts to show whether or not there are any conditions which might bring about improvements in animals that were just diabetic have not as yet been sufficiently made to warrant any con- clusions that could help us in the treatment of human cases. The en- couragement of the internal pancreatic secretion by diminution of that discharge into the intestine may be of value. The certainty with which diabetes results from pancreatectomy in dogs, as well as the frequent occurrence of demonstrable lesions in the pan- creas in diabetes'in man, leaves no doubt that this gland must be in some way essential in the physiologic breakdown of carbohydrates in the normal animal, but how, we can not at present tell. All we know is that the first change to occur after the gland is removed, is a sweeping out of all but a trace of the glycogen of the liver, although the muscles may retain theirs ; indeed, in the cardiac muscle there may be more than the usual amount.28 Nor can any glycogen be stored in the liver when excess of carbohydrates is fed. After the glycogen has disappeared, gluconeogenesis sets in, so that the tissues come to melt away into sugar, and all the symptoms of acute starvation, associated with certain others that are possibly due to a toxic action of the excess of sugar or other abnormal products in the blood, make their appearance. So far it might be permissible to consider an overproduction of glu- cose as the cause of the hyperglycemia of pancreatic diabetes, just as we have seen it to be of these forms of hyperglycemia that are due to stimulation of the nervous system ; but this can not be the case, for another very definite abnormality in metabolism becomes evident — namely, an inability of the tissues to burn sugar. This fact is ascer- tained by observing the respiratory quotient. When glucose is added THE METABOLISM OF THE CARBOHYDRATES 681 to the blood in the case of a completely diabetic animal, no change oc- curs in the quotient. There are, therefore, two essential disturbances of carbohydrate metabolism in pancreatic diabetes — overproduction of sugar and aboli- tion of the ability of the tissues to use it. It becomes important for us to see whether the tissues exhibit this inability to use sugar when they are isolated from the animal ; for if they should, a much more searching investigation of the essential cause of their inability would be possible than is the case when they are functioning along with the other organs and tissues. The earlier experiments of Lepine and his pupils, which seemed to show that diabetic blood did not possess the glycolytic power of normal blood ; and those of Cohnheim, from which it was concluded that mixtures of the expressed juices of muscle (liver) and pancreas, although ordinarily destroying glucose, failed to do so when they were taken from a diabetic animal, are now known to be erroneous. The failure to show a depression of glycolytic power by these methods prompted Knowlton and Starling24 to investigate the question whether any difference is evident in the rate with which glucose disappears from a mixture of blood and saline solution used to perfuse a heart outside the body, according to whether the heart was from a normal or a dia- betic dog. In the first series of observations which these workers made, it was thought that the normal heart used glucose at the rate of about 4 mg. per 100 gm. of heart substance per minute ; whereas that of a dia- betic (depancreatized) animal used less than 1 mg. If such striking differences in the rate of sugar consumption could make themselves manifest for so relatively small a mass of muscular tissue as that of the heart, it is permissible to assume that a much more striking difference could be demonstrated when the perfusion fluid is made to traverse all or practically all of the skeletal muscles, as well as the heart. For this purpose an eviscerated animal may be employed — that is, one in which the abdominal viscera are removed after ligaticn of the celiac axis and mesenteric arteries, and the liver is eliminated by mass ligation of its lobes. Using such preparations, R. Gr. Pearce and Macleod29 found that the rate at which glucose disappears from the blood, although very irregular, is in no way different in completely diabetic as compared with normal dogs. They were thus unable to confirm any of Knowlton and Starling's earlier conclusions. Patterson and Starling subsequently pointed out that a serious error was involved in the earlier perfusion experiments, partly on account of a remarkable but irregular disap- pearance of glucose from the lungs, and partly because the diabetic heart may contain a considerable excess of glycogen, from which its 682 METABOLISM demands for sugar may be met without calling on that of the perfusion fluid. In spite of the failure to show that the isolated tissues of diabetic animals have a lower glucose-consuming power than those of normal animals, it is . important from a practical standpoint that we should know something regarding the possible nature of the disturbance which a removal of the pancreas entails. Even if we could not tell exactly how this disturbance operates, it would be of value to know whether it depends on the removal from the organism of some hormone that is essential to carbohydrate utilization, for, if this were proved to be the case, encouragement would be offered to seek for the chemical nature of this hormone so that we might administer it with the object of re- moving the diabetic state. The hope of a fruitful outcome of such an investigation is encouraged by the success of researches on diseases of other ductless glands, particularly the thyroid. The removal of some hormone necessary for proper sugar metab- olism is, however, by no means the only way by which the results can be explained, for we can assume that the pancreas owes its influence over sugar metabolism to some change occurring in the composition of the blood as this circulates through the gland — a change which is de- pendent on the integrity of the gland and not on any one enzyme or hormone which it produces. It is obvious that the results of removal of the gland could be explained in terms of either view, and indeed there is but one experiment which would permit us to decide which of them is correct. This consists in seeing whether the symptoms which follow pancreatectomy are removed, and a normal condition reestab- lished, when means are taken to supply the supposed missing internal secretion to the organism; if they should be, conclusive evidence would be furnished that it is by "internal secretion" and not by "local in- fluence" that the gland functionates. The experiments have been of two types: in the one, variously pre- pared extracts of the glands have been employed, and in the other, blood which is presumably rich in the internal secretion. The most recent work with pancreatic extracts has shown that injection of pan- creatic extracts into a depancreatized animal produces no change in the respiratory quotient, although injections of extracts of pancreas and duodenum may produce a temporary fall in the dextrose excretion in the urine on account of the alkalinity of the extract. Neither have experiments with blood transfusions yielded results that are any more satisfactory. In undertaking these experiments it is of course assumed that the internal secretion is present in the blood, and that if this blood is supplied to an animal suffering from diabetes because of the loss of THE METABOLISM OF THE CARBOHYDRATES 683 its pancreas, it will restore it to a nondiabetic state. The general con- clusion that may be drawn from the numerous researches of this nature, is that there is no evidence that the blood of a normal animal, even when it is from the pancreatic vein, contains an internal secretion that can restore to a diabetic animal any of its lost power to utilize carbo- hydrates. When the extent of glycosuria alone is used as the criterion of the state of carbohydrate metabolism, serious errors in judgment are liable to be drawn. The condition of the blood sugar and the extent and character of the respiratory exchange are the most reliable indexes. DIABETIC ACIDOSIS OR KETOSIS Nature and Cause. — Much confusion has existed in medical literature over the correct definition of acidosis, mainly because the term was first used for the particular variety of the condition observed in the later stages of diabetes mellitus. The acids which accumulate in the tissue fluids in this disease are acetoacetic and /?-oxybutyric, and they are oxidation products of acetone, which is again derived from fatty acids by a faulty metabolism (see page 709). The essential cause of the acidosis is therefore entirely different from that in nephritis; in dia- betes foreign acids are added to the blood, whereas in nephritis the acids of a normal metabolism accumulate because of faulty excretion through the kidneys. The usual signs of acidosis exist in both cases, because the surplus of acid depletes the store of bicarbonate and causes changes in the alveolar C02, in the C02-absorbing power of the blood, in the reserve alkalinity, and in the acid excretion by the kidney. It is important to recognize the special nature of diabetic acidosis by a separate name — ketosis. The chemical processes by which the ketone bodies are produced is discussed elsewhere (page 709). It remains for us to consider the general nature of the metabolic disturbance responsible for their ap- pearance in diabetes. For the thorough combustion of fat in the animal body a certain amount of carbohydrate must be simultaneously burned. Fat evidently is a less readily oxidized foodstuff than sugar; it needs the fire of the burning sugar to consume it. If the carbohydrate fires do not burn briskly enough, the fat is incompletely consumed; it smokes, as it were, and the smoke is represented in metabolism by the ketones and derived acids. Such a closing down of the carbohydrate furnaces may be brought about either by curtailment of the intake of carbohydrates, as in starvation (page 569), or by some fault in the mechanism of the furnace itself, as in diabetes. Besides fat, protein may also contribute 684 METABOLISM to the production of ketones when carbohydrate combustion is de- pressed. Fundamentally, therefore ketosis in diabetes is due to the same cause as in starvation — namely, an improper adjustment between the metabolisms of fat and carbohydrate. Bearing these principles in mind, it is easy to see how the intensity of acidosis which develops during starvation will depend upon the re- lative metabolism of carbohydrate, on the one hand, and of fat and protein, on the other; it will therefore depend on the amounts of these foodstuffs which have been stored in the organism, and this again will depend on the nature of the diet previous to the starvation period. For the first few days following entire abstinence from food in a healthy, well-nourished individual, very few if any ketones will be excreted in the urine, because the carbohydrate stored in the body as glycogen has sufficed during this time to maintain the proper proportion between fat and carbohydrate. Afterwards, however, their appearance is to be ex- pected, because the glycogen stores become exhausted long before those of fat. If starvation is still further prolonged^ a stage will come when the fat, as well as the carbohydrate, is used up so that the organism has now to subsist on protein alone. When this stage arrives, the ketones will diminish, for, although they might be derived from certain of the amino acids, yet this does not actually occur, because a large part of the protein molecule (nearly half) also becomes changed into glucose, which by burning, as above explained, prevents the formation of ketones from the other part of the molecule. For the same reasons, marked acidosis will not be expected to occur during any stage of starvation in lean persons, who from the start must utilize mainly their stored protein to supply the fuel upon which to live. In diabetes exactly the same principles apply, but to an organism in which the ability to metabolize carbohydrate has been depressed, so that "the maximum rate at which dextrose can be oxidized is fixed at some level which is absolutely lower than in health."30 Therefore, since a cer- tain proportionality must exist between the rates of combustion of fat and carbohydrate, the diabetic can thoroughly oxidize less fat; in other words, an amount of fat which could readily be burned in a healthy body is improperly burned by the diabetic, and ketones and their acids ac- cumulate. Starvation Treatment. — "In order to check a diabetic acidosis, it is necessary to restore the proper ratio of fatty acid to glucose oxidation," which can best be done by starvation, rest in bed and warmth. But this treatment may not at first suffice, because we have to deal not only with the acidosis bodies derived from fat, but with those which can be derived from protein on account of the diabetic organism having lost the power THE METABOLISM OF THE CARBOHYDRATES 685 even of burning the glucose which is derived from this foodstuff. By persistence in the starvation, however, the ability of the organism to utilize carbohydrate usually becomes so far restored that enough burns to prevent acidosis. Every case of diabetes can not, therefore, be expected lo react in the same way to starvation, the determining condition being the relation between the quantities of glycogen and fat stored in the body at the outset of the fasting period. This relationship depends on the nature of the previous diet. To sum up, ' ' fasting will lower acidosis either in. health or in diabetes, if it has the effect of stopping a one-sided metabolism and throwing the tissues on a more nearly balanced ration of fatty acids and glucose" — (Woodyatt). A practical point may be noted here — namely, that there is likely to be more danger of serious acidosis developing during starva- tion in fat than in lean diabetics. The importance of our appreciation of these facts in the starvation treatment of diabetes will be self-evident. CHAPTER LXXVII FAT METABOLISM Before considering, the physiology of fats, a few of the most essential points regarding their chemistry may be of assistance. THE CHEMISTRY OF FATTY SUBSTANCES It is usual to classify all substances that are soluble in ether as lipoids. They include fatty acids, neutral fats, cholesterols, cholesterol esters, and phospholipins. The fatty acids belong to two main homologous series, which differ from each other with regard to whether they are saturated or unsaturated. A saturated fatty acid is typified by palmitic, whose formula is CH3-CH2-CH2- CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-CH2-COOH, or CH3- (CH2)14-COOH; that is to say it is a higher member of the series to which acetic acid (CH3-COOH) belongs, differing from the latter in having four- teen extra methyl radicles, each joined to its neighbor by one bond or satu- rated linking on either side. Another member of this series is stearic, in which there are sixteen extra CH2 groups (CH3(CH2)16-COOH). An un- saturated fatty acid is oleic (CH3(CH2)7— CH29 = CH-(CH2)7-COOH). Its unsaturation is represented in the formula by the double bond or unsaturated linking, which it will be seen occupies a position in the mid- dle of the molecule, the other methyl radicles being linked together by single bonds. The fatty acids readily combine with alkali to form soaps; thus, CH3(CH2)14-COOH + KOH=CH3(CH2)]4-COOK + H20, the reaction being (palmitic acid) (soap) analogous to that by which acetic acid forms an acetate with alkalies. In place of being combined with alkali, the COOH (carboxyl) group of fatty acids may combine with alcohols to form substances called esters, Thus, acetic acid and ethyl alcohol form ethyl acetate, CH3COO JH + OHi C2H5=CH3COO-C2H5 + H20. When the alcohol thus (acetic (ethyl (ethyl acetate) acid) alcohol) united with fatty acid is glycerol (glycerine), in which there are three 686 FAT METABOLISM 687 OH (hydroxyl) groups, the resulting ester — called triglyceride — is neu- tral fat, Tripalmitin has the formula: CHj — OOC " C15H31 CH -OOC-G15H31 CH2-OOC-C15H31. By boiling neutral fats with alkali the fatty acid radicles are split off as soaps, leaving the glycerol. This process is called saponification, and it may be effected in many other ways, as for example by heating with steam or by the action of special enzymes called Upases, which are widely distributed in plants and animals. The natural fats are usually a mixture of triglycerides, and their dif- ferences in properties are dependent upon the relative amounts of fatty acids present. The three most important in animal fats are tripalmitin, tristearin and triolein. It is essential in the study of fat metabolism that we should know the most important methods ~by which the proportion of fatty acids present in a mixed fat is determined. These methods are as follows : 1. The melting point. Olein is liquid at 0° C.; palmitic acid melts at 62.6° C.; and stearic at 69.3° 0. The solidity of animal fats depends on the proportion of olein, palmitin and stearin present. Mutton fat, for ex- ample, is much stiffer than pig fat because it contains less olein and more stearin. The melting points of fats from different parts of the body may also vary. 2. The acid number indicates the amount of free fatty acid mixed with the fat, and is determined by titrating a solution of a weighed quantity of the fat in alcohol with a N/10 ' alcoholic solution of KOH, phenolphtha- lein being used as indicator. 3. The saponification value indicates the total amount of fatty acid present, both that which is free and that combined with glycerol. It is determined by heating a weighed amount of fat with an exactly known amount of alcoholic KOH (determined by titration with standard acid). After saponification is complete, titration of the mixture shows how much alkali has been used to combine with the fatty acid. This is the saponi- fication value. 4. The ester value indicates the amount of fatty acid combined with glyeerol, and is obtained by subtracting the acid value from the saponi- fication value. Besides these there are two values, known as the iodine and the Reichert- Meissl values, that are of importance because they depend on certain char- acteristics of the fatty-acid radicles. 688 METABOLISM 5. The iodine value indicates the amount of unsaturated fatty acids pres- ent, or the number of double bonds. It depends on the fact that iodine, like many other substances, is capable of directly attaching itself to the fatty-acid chain wherever double bonds exist. 6. The Reich ert-Meissl value indicates the amount of volatile soluble acid present in the fat. It is determined by first of all saponifying the fat, then decomposing the soap by mixing it with mineral acid and dis- tilling the liberated fatty acid, the distillate being collected in a known amottnt of standard alkali and titrated. It is a value that is not of very great use in physiologic investigations, but it is so in connection with food chemistry. Since volatile acids are present in butter, the Reichert- .Meissl value helps us to distinguish between butter and margarine. Fat is insoluble in water but soap is soluble, forming a colloidal solu- tion which presents the phenomenon of surface aggregation of molecules. This consists in the concentration of the soap both at the free surface of the liquid, where a skin may form, and at the interfaces between the soap solution and any undissolved particles present in it. This pellicle- formation around the particles prevents them from running together so that they remain suspended, thus forming an emulsion. An emulsion may therefore be formed either of neutral fat of any other physically similar substance. When fat itself is used, there is usually enough free fatty acid admixed with it to make it unnecessary in forming the emul- sion to do more than shake the fat with weak sodium-carbonate solution. With other substances not containing any free fatty acid, some soaps should be added. To preserve the emulsion it is often useful to add some mucilage. In the emulsified state, neutral fats are much more readily attacked by lipases than when they are present in an unemulsified state. Thus, emulsified fats are "digested" by the relatively small amounts of lipase present in the stomach, whereas neutral fats themselves are not so. Fatty acids also exist in nature in combination not with the triatomic alcohol, glycerol, but with monatomic alcohols such as cholesterol. These cholesterol fats differ from the glycerol fats in being very resistant to- wards enzymes and microorganisms. They are therefore used for pro- tective purposes in the animal economy; for example, they occur in the sebum, the secretion of the sebaceous glands, where they serve to moisten the hairs and skin. They are also present in cells, in which it is prob- able they take an important part in forming the skeleton of the cell. Cholesterol is absorbed from the intestine ; it is always present in the blood both in plasma and in corpuscles ; and it is an important constituent of bile, from which it may separate out in the bile passages and form calculi (gallstones). In the cells themselves the lipoids are represented mainly by compounds FAT METABOLISM 689 of a somewhat more complex structure — namely, the phospholipins. As their name indicates, these consist chemically of phosphoric acid combined with neutral fat and with a nitrogenous base, cholin. The best known of the phospholipins is lecithin, which is widely distributed in the animal body (present in blood and bile as well as in all cells). Other phospholipins present in nervous tissue are cephalin, cuorin and sphingomyelin. There are various lecithins distinguished from one another by the fatty-acid radicles which they contain. Distearyl-lecithin, for example, has the formula: CH2 - O - OC (CH2) M - CHS CH -0-OC(CH2)14-CH3 (stearic acid) CH2-0 O (glycerol) P / \ OH OCH2-CH2-]Sr(CH3)2 (phosphoric | acid) OH (choline) This complex molecule can readily be split up by hydrolysis (warming with baryta water) into: glycero-phosphoric acid, CH, - OH CH -OH CH2 - O O f C2H.OH \ / P ; choline, N < (CH3)3 (oxy-ethyl-ammonium / \ OH OH [OH hydroxide) ; and fatty acids. With hydrochloric acid, choline forms a salt which readily forms a double salt with platinic chloride. Since this double salt forms charac- teristic crystals, it is used to identify and separate lecithins. For quan- titative purposes, however, it is more suitable to determine lecithin in- directly by the amount of phosphoric acid present in an ethereal ex- tract of the organ or tissue. Evidence is constantly accumulating to show that lecithin is an ex- tremely important constituent of cells; indeed, it seems to be the inter- mediate stage in the utilization of neutral fats by protoplasm. Its phos- phorus also probably serves as the source of this element for the con- struction of nucleic acid (see page 637). In nervous tissues it is often associated with carbohydrate molecules (galactose), forming the sub- stance known as cerebrin. It may therefore have some role to play in carbohydrate metabolism. Some Avorkers also attribute to lecithin an 690 METABOLISM important function in the transference of substances through cell mem- branes. When mixed with water it swells up by imbibition, and if crys- talloids or other substances are dissolved in the water, a means is offered for bringing water-soluble and fat-soluble substances into intimate con- tact. DIGESTION OF FATS A certain amount of fat, especially when it is in an emulsified condi- tion, can be digested in the stomach by the lipase contained in the gas- tric juice. Most of it, however, is digested in the small intestine, into which as we have seen, it is gradually discharged suspended in the chyme. For this intestinal digestion of fat both pancreatic juice and bile are nec- essary. This is easily shown in the rabbit, in which the pancreatic duct enters the intestine at a considerable distance below the bile duct. If the mesentery is inspected during the absorption of fatty food, no fat in- jection of the lymphatics will be noted between the bile and the pan- creatic ducts but only below the latter. In the dog, in which both the bile and the main pancreatic ducts enter the intestine at about the same level, fat injection of the lymphatics starts at this point, but if the bile duct (or rather the gall bladder) is transplanted at some distance down the intestine, it will be found that the injection of the lymphatics with fat occurs only below the new point of insertion of the bile duct. Removal of the pancreas interferes very materially with the absorption of fat. In man, for example, absence of the pancreatic juice alone di- minishes the absorption of fat by 50 or 60 per cent. If the bile is also absent, the diminution amounts to 80 or 90 per cent, and in such cases, as is well known, the administration of bile or pancreas powder greatly improves fat absorption. In the dog, although ligation of the pancreatic duct apparently only slightly influences fat absorption, removal of the pancreas itself greatly interferes with the process; from which fact some observers have concluded that the pancreas, in addition to its external secretion into the intestine, must produce an internal secretion into the blood which has something to do with the efficient absorption of the fat (Pratt, McClure and Vincent'*8). It is, however, improbable that such a'n hypothesis is necessary, for it is very likely that the moribund condi- tion into which an animal is brought by extirpation of the pancreas, adequately accounts for the suppression of the fat-absorbing function. As to the relative roles of pancreatic juice and bile in the digestion of fat, we know of course that in the pancreatic juice there exists a lipolytic enzyme, lipase, which, under suitable conditions has the power of split- ting neutral fat into fatty acids and glycerine. If bile is examined, no lipolytic enzyme will be found in it. It is entirely inactive on fat, but PAT METABOLISM 691 if we mix bile with fresh pancreatic juice, which by itself only slowly digests fat, we shall find that the bile very materially increases the lipo- lytic activity of the pancreatic juice. It has been found that the salts of cholalic acid, the so-called bile salts, are the constituents of bile that are responsible for this activation of lipase, this fact having been demonstrated with bile salts prepared in such a way that there was no possible chance of any other biliary constituent being present as an impurity. It is important to remember, however, that lipase itself be- comes slowly activated on standing, which explains why it should be that bile added to pancreatic juice that has stood for some time, has a less evident activating influence than bile added to fresh juice. It is probable that the activating influence of bile salts is due to some physico- chemical change induced in the digestion mixture. One may ask how it happens that, when bile and pancreatic juice are both absent from the intestine, the fat which appears in the feces is not neutral fat but fatty acid. The reason is that the neutral fat that has escaped digestion in the small intestine becomes acted on by the intestinal bacteria, particularly in the large intestine. Under these conditions, however, the fatty acid that is split off is not absorbed, because the epithelium of the lower parts on the intestinal tract can not perform this function. Besides assisting the action of lipase, bile facilitates fat digestion in other ways. Thus, by its containing alkali and mucin-like substances it assists in the emulsification of fat. Although eiriulsification is no es- sential part of fat absorption, yet it greatly facilitates the process by breaking up the fat into small globules on which the lipase can act much more efficiently. The alkali also combines with the fatty acids, as they are liberated by the digestive process, to form water-soluble soaps, which are readily absorbed by the epithelial cells. The bile salts further assist in the solution of the fatty acids, and they lower the sur- face tension of fluids in which they are contained and so bring the fat and lipase into closer contact. ABSORPTION OF FATS After its digestion fat lies in contact with the intestinal border of the epithelial cells as fatty acid and glycerine. The fatty acid is combined either with alkali to form a water-soluble soap, or with bile salts to form a compound, which is also soluble. The glycerine and the dissolved fatty acids are separately absorbed into the epithelial cells of the in- testine, in the protoplasm of which — after the fatty acid has been set free from the alkali or bile salt — they become united or resynthesized to form neutral fat, which gradually finds its way by the central lac- 692 METABOLISM teals into the villi, and then by way of the lymphatics to the thoracic duct. The chemical explanation of the absorption of fat is very different from that formerly held by histologists who maintained that the fine particles of emulsified fat in the intestine penetrate by a mechanical process through the striated border of the epithelial cell into its protoplasm. The histologic evidence for this view seemed very convincing, for fine fat globules can readily be seen in the epithelial cells of the intestine after fatty food has been taken, while they are absent during starvation. These par- ticles seemed to have passed directly from the intestinal canal into the epithelial cells because, when the fat was stained with characteristic fat stains before feeding it to the animal, the globules in the epithelial cells were found to be similarly stained. The supporters of this mechanistic view of fat absorption maintained that the appearance of the stained fat globules in the epithelial cells could not be explained in any other way than by supposing that the fat globules had wandered unbroken into the epithelial cells. Such a conclusion is, however, unwarranted, for the stains that are soluble in fat are also soluble in soap, so that when the fat splits up, the stain will remain attached to the soap and be carried along with it into the intestinal epithelium. Absolute proof that the chemical theory is the correct one has been supplied by a large number of experiments. The following may be cited: (1) When the lymph flowing from the thoracic duct is examined after feeding with fatty acids instead of neutral fat, it is found to contain only neutral fat, indicating that a synthesis must have occurred between glycerine and fatty acid during the absorption. The glycerine for this synthesis is furnished from sources which will be described later. (2) When an emulsion made partly of neutral fats and partly of some hy- drocarbon, such as albolene, is fed and the feces are examined for these substances, it has been found that all the fat but none of the hydrocar- bon is absorbed; the feces contain all of the albolene but none of the fat. This experiment supplies very strong evidence against the mechanistic theory, for microscopic examination of the above described emulsion shows the particles of neutral fat and hydrocarbon to be of exactly the same size. (3) By examining the properties of the fatty substances in the thoracic lymph collected during the absorption of such an emulsion as that described above, nothing but neutral fat has been found present. (4) Similar results are obtained when wool fat, which is an ester of cholesterol and fatty acid, is fed. We may conclude that fatty substances which are insoluble in water or can not ~be changed by digestion into substances (soap) that are soluble in water, are not absorbed, however like fat they may be in other particulars. FAT METABOLISM 693 The chemical theory of fat absorption further explains why there should be such large quantities of soapy substances in the intestinal contents, and also why the globules of fat present in the epithelial cells of the intestine are so very much smaller than those which lie on the surface of the epithelium. It might be objected to the conclusions just stated that, although unde- tectable, there is really some essential physical difference between emul- sified fat and emulsified hydrocarbon. In order entirely to prove the case for the chemical theory, it is necessary to feed a neutral fat possessing some characteristic that depends on the manner of union existing between fatty acid and glycerine, and then to see whether it appears in an un- changed condition in the thoracic duct. If it does so, the fat must have been absorbed through the intestinal epithelium in an unbroken, unsapon- ified condition, for it is unlikely that, in the resynthesis which occurs in the intestinal epithelium, the fatty-acid molecules would recombine with the glycerine molecules in exactly the same manner as before. There are, however, but very few qualities of neutral fats, apart from those of the fatty acids which compose them, by which they can be char- acterized. The most likely one is that of optical activity. None of the ordinary fats is optically active, although from chemical considerations it is quite conceivable that some should be so. In order to obtain such a fat Bloor49 conducted numerous experiments with the esters of stearic acid.* In a series of experiments Bloor fed isomannid-dilaurate, a syn- thetic fat of dextrorotatory power and as readily absorbed as natural fats, and by examination of the neutral fat present in the chyle flowing from the thoracic duct, found no evidence of the dextrorotatory fat. This result confirms previous work by Frank, who found that the ethyl esters of fatty acids are not absorbed unchanged. The results of both workers emphasize the probability that readily saponifiable fatty-acid esters do not escape saponification under the favorable conditions of the normal intestine. In other words, had the fats been absorbed unchanged, as would be required by the mechanistic theory of fat absorption, they would have appeared in the chyle in optically active conditions. These most important conclusions lead us to inquire as to the reason for the change in fat during its absorption. It can not be for the purpose of preventing the absorption of undesirable fatty substances, such as the petroleum hydrocarbons or the wool fats, because such substances are so rarely present in our food. It is most probable that the breakdown *Bloor prepared an optically active mannitan distearate, but found it to have a very high melt- ing point and to be only half as digestible as the ordinary fats. Its absorption was too slow and unsatisfactory to make it suitable for the above purposes. He, therefore, proceeded to prepare the di-ester of isomannitan with lauric acid, and he found the resulting compounds to be as well-ab- sorbed as ordinary fat, and yet to possess very marked dextrorotatory power, which, of course, they lose on saponification. This fat seemed suitable, therefore, for testing the above question. 694 METABOLISM and resynthesis of neutral fat occurs for the same reason that similar processes occur during the absorption and assimilation of protein. It will be remembered that protein is entirely disintegrated in the intestine into its so-called building stones. These are absorbed separately into the blood, which carries them to the tissues, in which they become re- synthesized to form the body protein. And so it appears to be in the case of fats. The process, in other words, permits of the rearrangement of fatty-acid molecules, as a result of which the newly formed fat is more adaptable for use in the organism. It comes to be more like the char- acteristic fat of the animal. There may be another reason for the proc- ess. It will be remembered that lecithins, which constitute the most important of the fatty substances of the cell itself, are mixed glycerides — that is to say, are compounds containing a variety of fatty acids. The rearrangement of the molecules of neutral fat. which occurs during ab- sorption may be the first step in the transformation of fat into lecithin. In order to throw further light on the question, Bloor has performed a number of interesting experiments in which the chemical properties of fats before and after absorption were compared. The criteria which he took were melting point, iodine value, and mean molecular weight; the melting point representing the solidity of the fat, and the iodine value, its degree of unsaturation — that is, the number of double links in the fatty-acid chain. It was found that during absorption very con- siderable changes occur in these two characteristics; for example, when fat with high melting point and low iodine value was fed, the fat in the thoracic lymph was of distinctly lower melting point and higher iodine value. "When fat with a low melting point and a high iodine value was fed, the reverse change occurred, for the melting point of the thoracic lymph fat was higher and the iodine value lower. These results could be explained as due in the first case to the addition of oleic acid to the fat during its synthesis in the intestinal epithelium, and in the second case to the addition of some saturated fatty acid. When a fat consisting mainly of glyceride and saturated fatty acid, but with a low melting point, was fed, the addition of oleic acid was still found to occur, as judged from the iodine value. This indicates that the change is, not merely in order that the melting point of the absorbed fat may be lowered, but also for some chemical reason. In a fourth series of experiments, a lowering of iodine value occurred after feeding with cod-liver oil, which contains a high percentage of glycerides of highly unsaturated fatty acid. Evidently, then, the intestine possesses the power of modifying the com- position of fat during its absorption, and this modification is apparently of such a nature that it causes a change toward the production of a FAT METABOLISM 695 uniform chyle fat, presumably characteristic of the animal body. The changes are probably greater than could be produced by admixture of the absorbed fat present in the normal fasting chyle, but the source of the oleic acid or of the saturated acid required for this synthesis is at present unknown. CHAPTER LXXVIII FAT METABOLISM (Cont'd) THE FAT OF BLOOD Methods of Determination. — Normally the blood contains only a small percentage of fat, but after a fatty meal it may contain so large an amount that the fat actually rises to the surface of the blood like a cream. By means of the ultramicroscope, examination of the blood in the dark field after a fat-rich meal reveals the presence of glancing particles, the so-called "fat dust." These particles are most abundant about six hours after the meal has been taken, and they gradually disappear by the twelfth hour. They do not appear after a meal when the thoracic duct is ligated. They disappear when oxygen is bubbled through the blood. Fat dust has also been found abundantly present in the blood of em- bryo guinea pigs at full time, but not in the mother's blood. This would indicate that the placenta must have the power of taking the constitu- ents of fat from the mother's blood and building them into fat, which then passes into the blood of the fetus. The placenta under these condi- tions acts like the mammary gland. In this connection it is of interest that there is also much fat present in the blood of pregnant women. The fat content of the placenta is, however, greater in the early stages of pregnancy than later. Although these facts have been known for some time, it has been impossible, either on account of the large quantities of blood required for a chemical examination or because of the difficulty in estimating the amount of fat from the density of the ' ' fat dust, ' ' to follow with any great degree of accuracy the exact chemical changes that take place in the fat of the blood. Recently, however, Bloor has succeeded in elab- orating methods by which the fat content of the blood can be determined with satisfactory accuracy in small quantities of blood, so that a con- tinuous series of observations can be made over a considerable period. The fat is extracted from the blood by an alcohol-ether mixture with moderate heat. An aliquot portion of the nitrate is evaporated in the presence of sodium ethylate, which saponifies the fat. The residue, consisting of soap, is well washed and then treated with hydrochloric acid so as to precipitate the fatty acid. The density of the precipitate 696 FAT METABOLISM 697 thus produced is compared in an optical apparatus, called a nephelometer, with a standard solution of two milligrams of oleic acid treated in the same way. The fatty acids in human blood are mainly oleic and palmitic. The lecithin and cholesterol may also be estimated in the same blood extract. For lecithin the above extract of blood, after the removal of the alcohol and ether, is digested by heating with concentrated HNO3 and H2SO4. This decomposes the lecithin, liberating the phosphorus, a solution of the resulting ash being rendered faintly alkaline to phenol- phthalein and then slowly added to a silver nitrate solution. The density of the pre- cipitate thus produced is compared in the nephelometer with that of a precipitate pro- duced in the same amount of silver nitrate by adding to it a standard phosphoric acid solution. For cholesterol an aliquot portion of the above extract is saponified with sodium ethylate and then saturated with chloroform ; the chloroform extract is mixed with acetic anhydrid and H2SO4 (con.) until the bluish color is fully developed (Liebermann reac- tion), the intensity of which is then compared in a colorimeter with that obtained by similar treatment from a standard cholesterol solution. Variations in Blood Fat. — In the dog the percentage of fat in the blood is remarkably constant under normal conditions. After a fatty meal the increase in fat begins in about an hour, and reaches its maxi- mum in about six. The increase is not found in animals in which the thoracic duct has been li gated. Although this result would seem to contradict the view held by some that part of the fat which can not be accounted for in the thoracic-duct lymph 'is absorbed by way of the portal vein, it does not by itself disprove the hypothesis, for it has been found that the fat content of the portal blood is always higher than that of the jugular. Very interesting results have been obtained following the intravenous injection of emulsions of oil, either the so-called casein emulsion or col- loidal suspensions. Up to a dose of 0.4 gram per kilogram of body weight — which by calculation would suffice to raise the fat content of the blood by 100 per cent — there was no increase in fat content. In or- der to explain this disappearance of fat, it might be imagined that the injected fat particles formed emboli in the smaller Capillaries. Against such a view, however, is the fact that the particles of fat in these emul- sions are one-half to one-seventh the size of a red corpuscle. Although this argument is no doubt of some weight, it should be remembered that the physical condition of these fine fat globules is not the same as that of the red blood corpuscle. Their surface condition may be such that they readily agglutinate so as to form small masses, which may stick at the branching of the smaller arterioles and capillaries. Bloor himself suggests that the injected fat may be stored, possibly in the liver, since the fat in this organ, as we shall see later, increases under similar conditions. When twice the above quantity was fed in the form of egg- 698 METABOLISM yolk fat, some of it persisted in the blood for several hours. This in- crease may have been owing to the flooding of the temporary storehouse with fat, or, more probably, to a retarding influence that lecithin may have on fat assimilation, for lecithin itself persists in the blood for a long time after intravenous injection. During fasting, no increase in blood fat was found unless the animal, by special feeding, had been stuffed with excess of fat prior to the fast- ing period. The lipemia in this case indicates that fat is being trans- ported from one place to another to serve as fuel for the starving tissues. Narcotics were found to produce an increase in blood fat. Ether pro- duced this increase during the narcosis, whereas morphine and chloro- form did not do so until after recovery. The explanation given for the ether effect is that a mixture of blood and ether has higher solvent power for fat than blood alone. The explanation for the chloroform and mor- phine effects is that a certain amount of breakdown of the tissue cells, in which lipins are set free, supervenes upon the action of these narcotics. The blood fat also becomes enormously increased in about forty hours after the administration of phlprhizin, and on the second or third day after the administration of phosphorus. The special significance of these facts we shall consider .later in connection with the relationship of the liver to fat metabolism. By comparison of the fatty acid, lecithin, and cholesterol contents of blood during fat absorption, it has been found that there is a steady but very variable increase in fatty acid, accompanied by no variation in cholesterol, but with an increase in lecithin, which varies from 10 to 35 per cent, but does not run strictly parallel with the fatty-acid increase. It is probable that this increase in lecithin represents that part of the absorbed fat which is intended for immediate use in the tissues (page 705). The more or less independent increase in lecithin is of significance in connection with the fact that in many pathologic conditions of so- called lipemia the increase does not affect the fats of the blood but rather the lipoids (i.e., lecithin and cholesterol). Separate analyses of blood plasma and whole blood show the increase of lecithin to be much more marked in the corpuscles than in the plasma, whereas the fatty-acid increase is -confined to the plasma. To illustrate some of these points the following table will be of value. In it is shown the average distribution of fatty acid, lecithin and choles- terol in' normal individuals and in cases of diabetes, in which disease, as has been known for long, there is marked disturbance of fat metab- olism. FAT METABOLISM 699 BLOOD LIPOIDS IN NORMAL AND IN DIABETIC PERSONS NORMAL PER CENT MILD DIABETES PER CENT MODERATE DIABETES PER CENT SEVERE DIABETES PER CENT Fat by Bloor's | Method | Whole Blood Plasma 0.59 0.62 0.83 0.90 0.91 1.06 1.41 1.80 Total Fatty AcidsJ Whole Blood Plasma Corpuscles 0.37 0.39 0.34 0.59 0.64 0.45 0.65 0.75 0.48 1.01 1.28 0.62 Lecithin Whole Blood Plasma Corpuscles 0.30 0.21 0.42 0.32 0.24 0.42 0.33 0.28 0.40 0.40 0.40 0.40 Cholesterol J Whole Blood Plasma Corpuscles 0.22 0.23 0.20 0.24 0.26 0.21 0.26 0.30 0.20 0.41 0.51 0.24 Glycerides | Plasma Corpuscles 0.10 0 0.38 0.18 0.46 0.23 0.84 0.38 Total Lipoids Plasma 0.68 0.98 1.16 1.98 It will be observed that there is about 0.7 per cent of total fatty sub- stances in normal blood. The fatty acids (palmitic and oleic) amount to about 0.4 per cent, and are equally distributed between plasma and corpuscles; the lecithin, about 0.3 per cent, being twice as abundant in corpuscles as in plasma, and the cholesterol, 0.2 per cent, about equally distributed. In diabetes all of these substances are seen to be increased in proportion to the severity of the disease, the increase being mostly in the plasma. The increase in cholesterol (confined mainly to the plasma) is particularly interesting, since the substance is unaffected in amount by excessive feeding with fat. The Destination of the Fat of the Blood. — In general, it may be said that the blood fat is transported to three places: (1) the depots for fat; (2) the liver; and (3) the tissues. The fat present in each of these places differs from that in the others, as is revealed by chemical examination by the methods described on page 687. The depot fat usually yields about 95 per cent of its total weight as fatty acid. The tissue fat, on the other hand, yields only about 60 per cent of its total weight as fatty acid. This difference indicates that the fatty acid must be combined in the tissues with a much larger molecule than is the case in the fat of the depots. This large molecule is probably that of lecithin or other phos- pholipin, and the smaller molecule in the depots, that of neutral fat. The liver fat takes an intermediate position between depot fat and tissue fat in its yield of fatty acid. When no active metabolism of fat is go- ing on, the liver fat is like that of the tissues ; but when fat metabolism is active, the liver fat occupies an intermediate position between liver fat and depot fat. 700 METABOLISM Another difference among the fats in these three places is Avith regard to the degree of saturation of the fatty-acid radicles. This, it will be remembered, is indicated by the iodine value; the higher the iodine value, the greater the desaturation of fatty acid. In depot fat this value is relatively low — for example, about 30 in the goat and about 65 in man ; depending somewhat on the fat taken in the food, compared with which it is usually a little higher. The fat in the tissues, on the other hand, has a high iodine value, possibly 110 to 130. The iodine value of the fat of the liver is remarkably inconstant, being about the same as that of the tissues when fatty-acid metabolism is not particularly active, but approximating that of the depots when fat mobilization is proceeding. The significance of this fact we shall consider later. The Depot Fat. — The places in the animal body where depot fat is deposited in greatest amount are the subcutaneous and retropsritoneal tissues. These fat depots may sometimes become of enormous size, as in the case of the famous dog of Pfliiger, of whose total body weight 40 per cent was due to fat. Bloor suggests that there may really be two different types of fat storage: one of a purely temporary character, which readily takes up and liberates the fat, but which is of limited capacity and possibly under the control of some quickly acting regulat- ing mechanism, like that of the glycogenic function of the liver; and another of a more permanent nature, into which the fat is slowly taken up, but the capacity of which is very much greater. TAVO questions present themselves concerning this depot fat: (1) Avhere does it come from, and (2) what becomes of it? Regarding the source of tlie depot fat, there is no doubt that it comes partly from the fat and partly from the carbohydrate of the food; in other words, it is either taken ready-made with the food or manufactured in the organism. That some of it comes from the fat of food is now a Avell-established fact, the evidence for AArhich need not detain us long. The best-knOAvn experiment consists in first of all starving an animal until his stores of fat are nearly exhausted and then feeding him Avith some "ear-marked" fat — that is, Avith some fat having a characteristic property AA^hich it Avill not lose during absorption. It Avill be found that the depot fat thereby deposited presents many of the qualities of the fed fat. The "ear- marking" of the fat may be secured by using fats of different melting points, such as mutton fat, Avhich has a high M.P., or olive oil, which has a low M.P. On feeding a previously starved dog with mutton fat, the M.P. of the depot fat approaches that of mutton fat — he becomes a dog in sheep's clothing; Avhereas when olive oil is fed, the subcutaneous fat becomes oily. Or again Ave may "ear-mark" the fat by combining it Avith bromine, when the deposited fat will likeAvise be brominized fat. FAT METABOLISM 701 It must not be imagined, however, that no change takes place in the fat during its absorption and before it becomes deposited in the tissues. -On the contrary, the stamp of individuality is put upon the fat, for, as we have already seen, its iodine value may become altered and its melt- ing point changed during the process of absorption. In other words, although the absorbed fat does not become entirely adapted to conform with the ordinary qualities of the depot fat, yet it tends to change in this direction. That some of the depot fat comes from carbohydrate is wrell known to stock raisers. When, for example, an animal is fed on large quantities of carbohydrate and kept without doing muscular exercise, its tissues become loaded with fat. If we desire strict scientific proof for this, we do not need to go further than the old experiments of Lawes and Gil- bert, who, it will be remembered, showed that the fat deposited in the tissues of a growing pig is greatly in excess of the fat that could have been derived from the fat or protein which was meanwhile metabolized. The experiment was performed on two young pigs from the same litter and of approximately equal weight ; one was killed and the exact amounts of fat and nitrogen in the body determined; the other was fed for several months on a diet the fat and protein contents of which were accurately ascertained. When after four months this pig was killed and the fat determined, it was found that much more had become deposited than could be accounted for by the fat and protein of the food, even suppos- ing that all the available carbon of the protein had become converted into fat. The only conclusion is that the carbohydrate must have been an important source of the extra fat. The Destination of the Depot Fat. — The depot fat becomes mobilized and transported by the blood to the active tissues whenever the energy requirements of the body demand it. During starvation, as we have seen, the depot fat is used to supply 90 per cent of the energy on which the animal maintains its existence. Before the fat is transported, it is probably broken down into fatty acid and glycerine, as which it passes through the cell walls to be again reconstructed into neutral fat in the blood. What agency effects this constant breakdown and resynthesis of fat it is difficult to say. Two ester-splitting enzymes are present in blood, one acting mainly on simple esters, the other on glycerides; but it has been impossible to demonstrate any evident relationship between either of them and the extent of fat mobilization. The Fat in the Liver. — The physiology of the liver fat has been very diligently studied, particularly by Leathes and his pupils.50 The out- come of this work has been to show that the liver occupies an extremely important position in the metabolism of fat, being, as it were, the half- 702 METABOLISM way house in the preparation of the fatty-acid molecule for consumption in the tissues. Fat is a material containing large quantities of poten- tial energy. While in the depots this potential energy is so locked away as to be unavailable for tissue use. To make it available the depot fat is carried to the liver, where the energy becomes unlocked but not actu- ally liberated. The fat is then transported to the tissues, and the libera- tion of the energy occurs. Neutral fat is like wet gunpowder: it con- tains much potential energy, but not in a suitable condition for explo- sion. The liver, as it were, dries this gunpowder, whence it is sent to the tissues to be exploded. The great importance of the liver in fat metabolism is indicated by comparison of the percentages of fat — or better of fatty acid — contained in it under different conditions of nutrition. In the ordinary run of slaughter-house animals the liver contains from 2 to 4 per cent of higher fatty acid, but in about one in every eight animals a much higher per- centage will be found to occur. The same is true in laboratory animals. In the case of the human liver as obtained' from autopsies in certain classes of patients, from 60 to 70 per cent of the dry weight- of the organ, or 23 per cent of the moist weight, may be fatty acid. There is no other organ in the animal body that is ever loaded with fat to this extent. As in the depots, this liver fat might be derived either from fat carried to the organ from elsewhere in the body, or it might represent a surplus of manufactured fat. That transportation of fat to the liver occurs is very readily demon- strable both in the laboratory and in the clinic. About forty hours after giving phlorhizin to a dog, it has been found that enormous quan- tities of fat appear in the liver; a few hours later, hoAvever, this excess of fat may have entirely disappeared. Fatty infiltration of the liver is also observed in phosphorus poisoning, although in this case the fat usually persists till the death of the animal. In man, in delayed chlo- roform poisoning and in cyclical vomiting, enormous quantities of fat may be present in the liver within a very short period of time after the onset of the condition. There can therefore be no doubt that fat is transported to the liver under abnormal conditions, but this can not be taken as evidence that the liver has anything to do with fat metab- olism in the normal animal. Such evidence has been supplied by Coope and Mottram,51 who have been able to show that, at least in rabbits, a similar invasion of the liver with fat occurs in late pregnancy and early lactation. They also found that the fatty acid deposited in the liver in late pregnancy gives an iodine value which lies nearer to that of the mesenteric fatty acid than is the case in normal animals. Mottram con- cludes that "wherever . . . there is abundant fat metabolism, the FAT METABOLISM 703 liver is found to be infiltrated with fats, presumably to be handed on elsewhere when worked up." It is interesting that the fetus is greedy of unsaturated fatty acids. The most likely source of the fat transported to the liver is the fat pres- ent in the depots, unless when digestion is in progress, when it may be the fat from the intestine. That much of it comes from the depots is easily demonstrated. Thus, the more extensive the infiltration of the liver with fat, the more closely will this fat be found to agree with the depot fat in its chemical characteristics. This has been very clearly shown by, first of all, starving an animal so as to clear the depots of fat as much as possible; then feeding it on some "ear-marked" fat (unusual melting-point or a brominized fat) ; and after another day or so of starvation, so as to clear the liver- itself of fat, poisoning the animal with phosphorus or phlorhizin. The liver will be found shortly after- wards to be invaded with fat which has all the ear-marks of that on which the animal had been fed. Evidence of the same character has been furnished in a series of clin- ical cases by observations on the amount of fat and the iodine value of the fatty acid of the liver. This is shown in the accompanying table. FATTY ACIDS OF LIVEK CAUSE OF DEATH HIGHER FATTY ACIDS PER CENT OF DRY WEIGHT IODINE VALUE OF FATTY ACIDS 1. Pernicious anemia 12.1 116.8 Normal 2. Lobar pneumonia 3. Pernicious anemia 13.7 14.25 116.8 116.0 figures 4. Diabetes 14.4 119.6 5. Toxemic jaundice 15.6 109.5 Commencing 6. Accident 17.2 103.5 fatty 7. Empyema 21.5 96.0 change 8. Phthisis 25.4 96.4 9. Broncho-pneumonia 38.4 84.9 10. Appendicitis 44.0 91.1 Marked 11. Carcinoma of bladder 47.2 77.8 fatty 12. Broncho-pneumonia 54.6 71.8 change 1?>. Ulcerative colitis 60.9 80.3 14. Accident 66.3 63.0 15. Dysentery 73.5 69.1 This table clearly shows that the more fat there is in the liver, the nearer this fat approaches in character that stored in the depots. That some of the fat in the liver may come directly from the fat re- cently absorbed from the intestine is also very readily demonstrable. Thus, when cocoanut oil was placed in the intestine of anesthetized an- imals, along with bile salts and glycerine, it was found by Raper52 that 30 per cent of the absorbed oil appeared in the liver. 704 METABOLISM The characteristic feature of cocoanut o.il is that its fatty acids are volatile in steam and are saturated. Some of the fatty acids of the liver are volatile in steam, but they are unsaturated. By distillation in steam of the fatty acids obtained by saponification of the liver, it is possible to determine how much of the cocoanut oil has passed to the liver. Similar results have been obtained when unsaturated fatty acids, such as those contained in cod-liver oil, are fed. In all these cases the rela- tionship of the liver fat to that of the food is even more evident than that between food fat and depot fat, because in the liver the newly absorbed fat is not diluted by that deposited it may be months previously, as is the case in the connective tissues. The question now arises: What happens to the fat during its stay in the liver? An indication of the nature of the change is furnished by observing the iodine value of the fat. This, it will be remembered, in- dicates the degree to which the fatty acid is unsaturated. It does not necessarily indicate the number of unsaturated bonds present in the fatty- acid molecule, for the difference in iodine-absorbing power may depend not on the number of such bonds but on the position in the chain at which a given double bond is inserted. Even with this reservation, how- ever, it is evident that the increase observed in the iodine values shows that the liver has the power of desaturating fat. The advantage of this change depends on the fact that the desaturated fatty acid will be more liable to break up than the saturated fatty acid. In other words, the double linkage" will weaken the chain with the consequence that it is liable to fall apart at this place; such at least is the natural interpreta- tion which the chemist would put on the result. It may not, however, be the correct interpretation, for it has been shown that, although un- saturated fatty acids are more susceptible to chemical change in the laboratory than saturated, yet when fed to animals they appear to be more stable than many saturated acids. It may then be wrong to con- clude that the introduction of a double linkage in fat necessarily means the liability of the fatty-acid chain to break at that point. However this may be, it seems likely that one function of the liver consists in introducing double linkages at places in the fatty-acid chain, as a result of which the chain breaks at these places, and the fragments then undergo further oxidation. Double linkages may be introduced not only in one place in a fatty- acid chain, but in several. For example, it has been found in the liver of the pig, after oxidizing the fatty acids with permanganate, that oxida- tion products are obtained indicating the existence of unsaturated acid with four double links. Permanganate (in alkaline solution) is used for detecting the position of these double bonds, because, when it is allowed FAT METABOLISM 705 to act on unsaturated fatty acids in the cold, it causes hydroxyl groups to be introduced in the position of the double bonds. When .the oxidation is performed at a moderate temperature, the fatty acid falls apart at the hydroxyl groups. A fatty acid with eight hydroxyl groups has been obtained in this way from the liver of the pig. The presence of the hy- droxyl groups has been confirmed by finding that an octobromide is ob- tained by treatment with bromine. An acid of the same formula is said to be present in cod-liver oil. To sum up, we may conclude that there are certain positions, in the chains of carbon atoms which constitute the fatty- acid radicle, where the liver introduces double bonds, and that the weak- ened radicles then circulate to the tissues, where they break up at those positions. But this is probably not the only way in which the liver assists in the metabolism of fat. It may also take part in the building of fatty- acid radicles into the complex molecule of lecithin. The process of de- saturation that we have just considered is probably a preliminary step to this incorporation of the fatty-acid molecule into lecithin, for it is well known that lecithin contains highly unsaturated fatty-acid radi- cles. In support of such a view it is interesting to note that in alcohol- ether extracts from normal and pathologic livers, the lecithins, which are precipitated by acetone, have higher iodine values (i. e., are more unsat- urated) than the neutral fats extracted from the same liver, which also have higher iodine values than the depot fat of the same animal. The desaturation process must, therefore, involve the fatty acids before these become built into the lecithin molecule. The liver is probably not the only place in the animal body where the desaturation of fatty acids is brought about. The relative activity of the different tissues in this regard has been studied by feeding cats with fatty fish and then determining the iodine value of fat from various places in the body. The absorbed fat was more obvious in the liver than in the subcutaneous tissues, because it had not become diluted with fat deposited it may have been months previously, which would be the case in the fat of the fat depots; and it was found that, although the iodine value of the subcutaneous fat was slightly raised, that of the liver was much more so, indicating that the desaturation process had been more active in this organ, but had also occurred to a certain extent in the depots. Before leaving this subject of fat in the liver, it is important to re- call the old observation of Rosenthal, that a more or less reciprocal relationship exists between glycogen and fat in the liver. When much glycogen is present there is little or no fat, and vice versa. It is impor- 706 METABOLISM taut to note that the exact locations of fat and carbohydrate in the he- patic lobule are somewhat different in the two cases. A practical clinical application of the above work is found; in the fact that fats will be more readily utilized by the body when they contain a high percentage of unsaturated fatty acids. It is probably for this reason that Norwegian cod-liver oil is of such undoubted nutritive value. It is much more so than Newfoundland cod-liver oil, because in the prep- aration of this variety oxidation occurs, which makes it no longer unsat- urated. Fish oils in general are more unsaturated than other animal oils, and are for this reason more nutritious. The fat in the tissues differs very materially from that of the liver or the depots. Only 60 per cent of this fat consists of fatty acid, which is present very largely as part of the lecithin molecule, thus accounting for the high iodine value. Some is probably also present as simple glyceride, in a highly unsaturated and therefore very fragile condition. CHAPTER LXXIX FAT METABOLISM (Cont'd) Two very important questions of fatty-acid metabolism may now be considered: namely, (1) how is fatty acid formed from carbohydrate? and (2) what becomes of the fragments into which the fatty-acid molecule is split as the result of the desaturation process? Although these prob- lems involve chemical details of a somewhat complex nature, we must not on this account fail to consider them; for, as we shall see, much of what is known has an important practical application depending on the fact that certain of the intermediary substances may accumulate in the organism and develop a toxic action. The Production of Fatty Acid out of Carbohydrate.— If we place the formulas for glucose and palmitic acid side by side, thus: CH2OH-(CHOH)<-CHO (glucose), and CH3-(CIL,)14-COOH (palmitic acid); we shall see that this transformation must involve: (1) a considerable alteration in the structure of the molecule, (2) the removal of oxygen, and (3) the fusion of several glucose molecules into one molecule of fatty acid. The conversion of carbohydrate to fat therefore involves a process of reduction, and the resulting molecule must be capable of yielding more energy when it is oxidized than the original one of carbohydrate, for obviously the system 02 - CH2 (which corresponds to fat) will develop more energy than that of 02 - CHO (which corresponds to carbohydrate) ; just as a piece of wood when it is burned Avill develop more heat than a piece of charcoal. This explains why one gram of fat yields 9.3 calories of heat, and one gram of carbohydrate, only 4.1 (page 535). Fatty acid therefore contains more potential energy than sugar, and in explain- ing its synthesis from sugar in the animal body we must indicate the source of the extra energy. This is dependent on oxidation of some sugar molecules — which do not themselves become changed to fatty acid — proceeding side by side with the reduction which affects the others and represented in the outcome of the reaction by the combustion products C02 and H,0, thus: 6C6H1206 + 13 02 = 20 C02 + C16H3202 + 20 H20. (glucose) (fatty acid) 707 708 METABOLISM What evidence have we that such a process actually occurs in the body? If we compare the intake of oxygen with the output of carbon dioxide in the respired air, we shall find that usually there is less of the latter; that is to say, the respiratory quotient, as this ratio is called, is usually less than unity. During the extensive conversion of carbohydrate into fat, however, which occurs during the fall months in hibernating animals, the K.Q. has been found to rise as high as 1.4. The great excess of C02 - output over 02 - intake which such a quotient indicates conforms with the above equation. The entire dissimilarity in chemical structure between the molecules of fat and carbohydrate suggests that the primary step in the conversion must be a thorough breakdown of the carbohydrate chain into compara- tively simple molecules, from which the fat molecules are then recon- structed and the unnecessary oxygen set free. The problem is to ascer- tain the chemical structure of these simpler molecules and the manner of their union into fatty acid. Of the several suggestions which have been made, that of Smedleyss seems the most likely. As will be seen from the following equations, the first step is the conversion of glucose to pyruvic acid (page 600, No. 1 in equations). By enzymie action pyruvic acid is converted into acetaldehyde (No. 2), which then condenses with another pyruvic- acid molecule to form a higher ketonic acid (No. 3), from which by the loss of (XX, as in the case of the production of • acetaldehyde from pyruvic acid, an aldehyde is pro- duced (No. 4). This higher aldehyde behaves like acetaldehyde in again combining with pyruvic acid, forming a still higher ketonic acid; and so on until at last a long fatty- acid chain is built up, thus : ( 1 ) C^H^O,, + O, = 2CH3COCOOH + 2H2O ( glucose ) ( pyruvic acid ) ( 2 ) CH3COCOOH — CH3CHO + CO2 (acetaldehyde) ( 3 ) CII3CHO + CH3COCOOH — CH3CH : CHCOCOOH -f H2O (unsaturated ketonic acid) (4) CH3CH : CHCOCOOH i= CH^CH : CHCHO + CO2 ; and so on. (higher aldehyde) (5) From the ketonic aldehyde formed at any stage, an unsaturated fatty acid (with one less C-atom) is readily formed, and this by taking up H may become saturated: CH3CH:CH CO COOH + O = CH3 CH:CH COOH + CO,. During the butyric-acid fermentation of sugar a slightly different process may occur — namely, the lactic acid, which we know is readily produced from glucose, may break down into acetaldehyde (and formic acid), and two such molecules condense to form /3-oxy- butyric aldehyde; and this again condense to form higher fatty acids, thus: ( 1 ) C0H12O6 r= 2CH3CHOHCOOH; (glucose) (lactic acid) ( 2 ) 2CH3CHOHCOOH = 2CH3CHO + H.COOH (acetaldehyde) (3) 2CH3CHO = CH3CHOHCH2CHO ; and so on. (/3-oxybutyric aldehyde) FAT METABOLISM 709 That higher fatty acids, such as caproic (C6H12O2) and caprylic (C8H16O2), have actually been isolated from the products of this fermentation is a very significant fact, and it is of interest to note that Leathes has sometimes found an increase in higher fatty acids to occur during the aseptic incubation of liver pulp. Unfortunately, however, the increase of fatty acid could not be shown to be affected by adding substances to the liver which, according to the above equations, should yield fatty acid. The Method by Which the Fatty Acid is Broken Down. — In the chemi- cal laboratory, ordinary oxidizing agents attack the fatty-acid chain at the C-atom next the carboxyl (COOH) group (the alpha C-atom). But this can not occur in the animal body, because it would leave behind a smaller chain containing an uneven number of C-atoms, and such chains are never found present in the animal fats. On the contrary, the com- moner fats all contain an even number of C-atoms; thus : Butyric, C4H802 ; palmitic, C1CH3202; stearic, C18H3(.02; oleic, C18H3402. The intermediary substances which are produced during the gradual breakdown of the fatty-acid molecule in the normal animal are of a very transitory character so much so indeed that it is impossible for any one of them to accumulate in sufficient amount to permit of isolation, or even detection, by chemical means. How then are we to identify the inter- mediary products? This has been rendered possible by the discovery that, when anything occurs to disturb the normal course of fat metabolism, as, for example, when the tissues are deprived of carbohydrates (as in star- vation or in severe diabetes), the oxidation of the fatty-acid chain stops short when a chain of four C-atoms still remains unbroken. These last four C-atoms seem to form a residue that is more resistant to oxidation than the remainder of the fatty-acid molecule. It is a residue, therefore, which is quite readily further oxidized to C02 and H20 under normal con- ditions, but which, although incapable of becoming completely oxidized Avhen the metabolism is upset, does undergo a partial oxidation, result- ing in the production of various intermediary products. These accumu- late in the body in sufficient amount to overflow into the urine, from which they can be isolated and identified. The fatty acid with 4 C-atoms is 'butyric, CH3CH2CH2COOH, and the first oxidation product formed from it in the body seems to be fi-oxybuiy- ric acid, CH3CHOHCH2COOH. This then becomes oxidized to form a body having the formula CH3COCH2COOH, acetoacetic acid, which, on further oxidation, readily yields CH3COCH3, or acetone. These sub- stances (/?-oxybutyric acid, acetoacetic acid and acetone) appear in the urine during carbohydrate starvation, as in diabetes. It might be objected, however, that a chemical process occurring under abnormal conditions need not also occur in the normal animal. That it probably does, hoAvever, is indicated by the results of the experiments 710 METABOLISM of Knoop and of Embden and his coworkers. Knoop conceived the idea of introducing into the fatty-acid molecule some group which is resistant to oxidation in the body. The phenyl group (C6H5) was found to have this effect. By feeding an animal with the phenyl derivatives of acetic, propionic, butyric, and valeric acids, it was found that the urine con- tained either hippuric (see page 630) or phenaceturic acid. Both of these are compounds of aromatic acids with glycocoll or aminoacetic acid (CH2NH2C'OOH), one of the protein building-stones and always available in the organism to form such compounds, thus: (1) C8H5COOH + CH2NILCOOHz=C6H/X)NHCH2 COOH. (benzoic (glycocoll) (hippuric acid) acid) (2) C6H6CH2COOH + CHJSTH^COOH = C6H5CH2CONHCH2COOH. (phenylacetic (glycocoll) (phenaceturic acid) acid) When either benzoic acid (C6H5COOH) or phenylacetic acid (CGH5CH2- COOH) is formed in the body as a result of the oxidation of phenyl derivatives of the higher fatty acids, the acid combines with glycocoll according to the above equations. From this it folloAvs that if oxidation occurs so that two C-atoms are thrown off at a time (/^-oxidation), fatty acids with an even C-atom chain should yield hippuric acid, and those with an uneven chain, phenaceturic. This was found to be the case, as the accompanying table shows. ACID FED OXIDATION PRODUCT EXCRETED AS Benzoie acid, CCH5.COOH Phonylacotic acid, C6II5 . CH; . COOH « Not oxidized Not oxidized Hippuric acid Phenaceturic acid Phenylwopioiiic acid, C8H, . CH, . CH, . COOH C6H5.COOH Hippuric acid Phenylbutyric acid, CeH5 . CH, . CH2 . CH2 . COOH C6H3 . CH2 . COOH Phenaceturic acid Phenyl valeric acid, C6H5 . CH2 . CH2 . CH . CH2 . COOH C6HC . COOH Hippuric acid (From Dakin.) Embden 's experiments are equally convincing. He studied the forma- tion of acetone in defibrinated blood perfused through the freshly excised liver. Normally only a trace of this substance is formed, but when fatty acids with an even number of carbon atoms were added to the blood, they gave rise to a marked increase in acetone, whereas those with an uneven chain failed to cause any change. The acetone was found to be derived immediately from acetoacetic acid. The following table shows the results. FAT METABOLISM 711 NORMAL FATTY ACID FORMATION OF ACETOACETIC ACID Acetic acid Propionic acid Butyric acid Valeric acid Caproic acid Heptylic acid Octoic acid Nonoic acid Decoic acid CH3.COOH CH3.CH2.COOH CH3.CH2.CH2.COOH CH3 . CH2 . CH2 . CH2 . COOH CH3 . OH2 . CH2 . CH2 . CH, . COOH CH3 . OH, . CH2 . CH2 . CH2 . OH., . COOH CH3 . CH"2 . CH2 . CH2 . CH2 . CH", . CH2 . COOH CH3 . CH2 . CH2 . CH2 . CH2 . CH, . CH2 . CH2 . COOH CH,.CH2.CH,.CH,.CH2.CHj.CH,.CH2.CH,.CH2. + + -f COOH + (From Dakin.) For a long time it was difficult for chemists to understand how such a process of oxidation at the /?-C-atom could occur, since they were unable to bring it about in the laboratory by the use of the ordinary oxidizing agents, but recently Dakin has removed the difficulty by show- ing that hydrogen peroxide (H202) oxidizes fatty acids just exactly in this way. We may sum up the results of these experiments and observations by stating that normal saturated fatty acids and their phenyl derivatives can undergo oxidation, not only in the animal body, but also in vitro, in such a manner that the two (or some multiple thereof} termial G -atoms are removed at each successive step in their decomposition. But we must not be too hasty in concluding from these experiments that the steps in the process are necessarily in the order of first, the produc- tion of a /3-hydroxy acid, and second, the oxidation of this to a ketone group. The mere presence, side by side, of /?-hydroxybutyric acid and of acetone in the' above experiments does not indicate which is the ante- cedent of the other,' and indeed there are several experimental facts that seem to show that the hydroxy acid may be derived from the ketone. For example, when acetoacetic acid is added to minced liver and the mixture incubated, /?-hydroxybutyric acid is formed (a reduction process), although lees usually the reverse action (oxidation) may occur when /?-hydroxy acid is added. A reversible reaction must therefore be capable of occurring between these two substances, thus: reduction CH3.CHOH.CH2.COOH < CH, . CO . CH, . COOH. oxidation (/3-oxy butyric acid) — • • '•> (acetoacetic acid) We know practically nothing as to the conditions determining whether oxidation or reduction shall predominate, but there are two significant facts that one should bear in mind: (1) that a plentiful supply of oxy- gen is necessary for the oxidative process, and (2) that the presence of readily oxidizable material in the liver (e.g., carbohydrates) may deter- mine the direction which the reaction shall take. It is commonly said that fats burn in the fire of carbohydrates, and it may be that the un- 712 METABOLISM doubted diminution in acidosis which occurs in diabetes when carbo- hydrate food is given is dependent upon the directive influence which its combustion in the liver has on the above processes. But we must be cautious not to transfer results obtained by experiments with minced liver in judging of the reactions Avhich occur during life. Provisionally, then, we must assume either that /Miydroxybutyric acid is a necessary stage in the oxidation of butyric acid or that it is formed by reduction of acetoacetic acid, which is really the first step in that process. Of course there is no evidence in the above experiments that the higher fatty acids are also broken down by the removal of two C-atoms at a time, nor has it been possible to detect any ketonic or /3-hydroxy deriv- atives of them in the animal body. We can only reason from analogy that a similar process may occur, although some support is furnished for such a view by the fact that ketonic fatty acids have been found in vegetable organisms. What, then, it may be asked, is the relation of the desaturation of fatty acids which we have seen occurs in the liver (and probably elsewhere) to the ft oxidation? There can be no doubt that both processes can occur in the animal body, indeed in the same organ, e.g., the liver; and it is important to ascertain their relationship to each other. The conclusion Avhich would seem to conform best with the known facts is that the desaturation process occurs (in the liver) so as to break up the long fatty-acid chain into smaller chains, which are then capable of /? oxida- tion (in the tissues) ; desaturation may be the process by which the mole- cule is rough hewn, and /? oxidation that by which the resulting pieces are finally split to their smallest pieces — that is, to molecules of the size of acetic acid, which are finally completely burnt to carbonic acid and water. The increase of iodine value observed by Leathes and his coworkers need not, as has already been pointed out, necessarily indicate that new double linkages have been intro- duced in the fatty-acid chain ; it may merely indicate that structurally isomeric deriva- tives which absorb iodine more readily have been formed. Direct evidence of desatura- tion has, however, been offered by Hartley, who isolated the unsaturated fatty acids (by dissolving the lead soaps in ether) from pig's liver and then proceeded to oxidize them with alkaline permanganate. When the olein of the depot fat is thus treated at a low temperature, two hydroxyl groups become attached where the double linkage existed (forming dioxystearic acid), and when the mixture is now warmed, the molecule splits into two at this place, forming two lower acids (pelargonic and azelaic) : (1) CH,-(CH,),CH:CH(CH2).COOH; (oleic acid) (2) CH,-(CH2)I-CH OH OH CH (CH2)7COOH; (dioxystearic acid) (3) CH, (CHS)7COOH + COOH-(CH2)7COOH. (pelargonic acid) (azelaic acid) FAT METABOLISM 713 We may conclude from this that the double linkage in the oleic acid of the depot fat exists between the ninth and tenth C-atoms. But it is otherwise in the case of the un- saturated acid from the liver (pig's), for under the above process of oxidation this yielded caproic acid, which, since this acid has six C-atoms, would indicate that the double linkage existed between the sixth and seventh C-atoms. Another interesting fact brought to light by the experiments was that a tetraoxystearic acid was formed, which fell apart in such a way as to indicate that the hydroxyl groups occurred between the sixth and seventh and between the ninth and tenth C-atoms. The occurrence of this substance would be satisfactorily explained by the introduction into the molecule of oleic acid of a second double bond— i. e., between the sixth and seventh C-atoms. ' ' The acids found in the pig's liver may be accounted for, in other words, by supposing that desaturation of stearic acid and of the ordinary (depot) oleic acid occurs by the introduction of a double link between the sixth and seventh carbon atoms in each ease" — (Leathes). Still other double links may, however, be introduced into the fatty-acid chain, for acids of the linolie acid series are present in cod-liver oil. Finally, it is of interest to note that caproic acid is a product of the above oxidation process, for it has an even number of C-atoms and therefore will form /3-oxybutyric acid. To go into these chemical problems any further here would be out of place. One other fact, should, however, be borne in mind — namely, that the unsaturated acids may be formed from saturated acids through the intermediate formation of /?-hydroxy and /8-ketonic acids. Their mere presence, in other words, should not be taken as evidence that the oxida- tion of fatty acids is initiated by the introduction of an hydroxyl group at the ft position in the chain. CHAPTER LXXX The classification of animals into two groups — warm-blooded and cold- blooded— according to their ability to maintain the body temperature at a constant level, is more or less arbitrary. Between the two groups an- other exists, represented mainly by hibernating animals, in which at certain times of the year the animal is warm-blooded and at other times cold-blooded. The ability of the higher mammals to maintain a constant body temperature may or may not be present at the time of birth. The heat-regulating mechanism of the human infant for example remains ill developed for some time, so that exposure to cold is liable to lower the body temperature to a dangerous degree. VARIATIONS IN BODY TEMPERATURE In animals in which the heat-regulating mechanism is fully developed, there is not, even during perfect health, entire constancy in temperature in the different parts of the body or in the same part at different periods of the day. The average rectal temperature of man is usually stated as being 37° C. (98.6° F.), but the diurnal variation may amount to 1° C., being highest in the late afternoon and lowest during the night. There are probably several causes for this variation, and they are in part at least dependent upon the greater metabolic activities of the waking hours and upon the taking of food. Apart from these influences, how- ever, others which are less evident appear to operate ; for it has been found that, when the daily program is reversed by night work, the usual diurnal variation, although much less pronounced, still remains evident even although this reversal in habit may have been kept up. for years. It is of interest to note in this connection that nocturnal birds have their maximum temperature at night and their minimum during the day. Regarding the temperature in different parts of the body, that of the rectum is usually about 1° C. higher than that of the mouth, and this again higher than that of the axilla. Of these three the mouth tempera- ture is the most variable, for many conditions, such as mouth breathing, talking, drinking cool liquids and even exposure to cold air, are sufficient to lower markedly the temperature of this region. When the mouth 714 CONTROL OF BODY TEMPERATURE AND FEVER 715 temperature is carefully taken by leaving the bulb of the thermometer under the tongue for a minute or more, it is practically the same as the temperature of the arterial blood of the hand when this is exposed to the ordinary conditions of outside temperature. Greater differences than 1° C. in the temperature of different regions of the body are often ob- served in feeble individuals and in those with some circulatory disturb- ance. FACTORS IN MAINTAINING THE BODY TEMPERATURE The body temperature represents the balance between heat production and heat loss. The production is effected mainly in the muscles by the oxidative processes which are constantly ensuing there. When the activity of the muscles is abolished by paralyzing the terminations of the motor nerves with curare, the temperature of warm-blooded animals immediately falls or rises according to the temperature of the environ- ment. A curarized warm-blooded animal is thus made to behave like a cold-blooded one. Increased muscular activity, on the other hand, promptly raises the body temperature by 1° or 2° C., above which, how- ever, a further rise does not occur, provided nothing has been done to interfere writh the mechanism by which the excess of heat is got rid of from the body. The temperature in such cases adjusts itself at a higher level, at which it remains fairly constant however strenuous the exer- cise. It is possible that a certain amount of heat may also be generated by the chemical processes occurring in the liver and other viscera, but when compared with the muscles this source of heat is undoubtedly in- significant. Many of these chemical processes, as in the liver, instead of producing actually absorb heat, so that the balance between heat- producing and heat-evolving mechanisms may or may not come out in favor of the liberation of heat. The production of heat goes on all the time in muscles on account of the condition of tonic contraction in which they are held (see page 814), and which is also necessary for keeping the joints in the proper degree of flexion or extension. When more heat is required by 4he animal body, the tone of the muscles increases independently of the function which they may be performing in controlling the position of the joints. This increased tone may become so pronounced that it causes visible contrac- tions, which AVC recognize as shivering. Whenever the "insensible hyper- tonicity and the shivering are inadequate to produce a sufficient amount of heat, the animal instinctively moves about in order that the greater contractions may liberate more heat. The heat is produced in the muscles by oxidation of the .foodstuffs that have been assimilated from the blood. Even during the process of as- 716 METABOLISM similation itself a certain amount of heat is generated; this is known as the specific dynamic action of the foodstuff, and is most pronounced with protein and least so with carbohydrate (page 538). Advantage may be taken of this heating power of protein to produce more heat when the cooling conditions are excessive ; in winter, for example, there is an inclination to take more protein food than during summer, and the per capita consumption of such food is "much greater in peoples living in temperate zones than in those living in the tropics. The ultimate amount of heat produced by oxidation is greatest with fat and least with carbo- hydrate. Heat loss in man is effected partly through the lungs, but mainly through the skin. Through the latter pathway heat is lost by the physical processes of heat conduction and radiation and by the evaporation of the sweat. Through the lungs it is lost mainly in the vaporization of the water contained in the expired air (latent heat of vapor). The amount of heat lost from the skin by conduction and radiation depends on the temperature of the skin, which again depends on the rate at which the blood is circulating through the cutaneous vessels. Under ordinary con- ditions of external temperature two or three times as much heat is lost by these methods as by evaporation. The losses by evaporation, under conditions of rest and average external temperature, are about equally divided between the lungs and the skin. From all these facts, it is evident that heat loss occurs mainly by the skin and only to a small degree by the lungs. This means that under average conditions in man the main regulation of heat loss is effected by variations in the skin temperature brought about by peripheral vaso-con- striction and dilatation. The marked sensitivity of the cutaneous blood supply to changes in the temperature of the environment has been very clearly shown by observations made with the hand calorimeter of Stewart described elsewhere (page 281). When the bloodflow through the hand is examined in a person who has been exposed to the outside air, it may be little more than half that which it attains after he has been in a warm room for some time. In the outside air the vessels con- strict to prevent heat loss by conduction and radiation; in the warm room they dilate to facilitate this loss. The afferent impulses which reflexly control the change in the cutaneous blood circulation may be set up by local applications of heat or cold, as can be shown in the hand-calorim- eter experiments by applying a cold pad to the skin of the correspond- ing forearm, when an immediate curtailment of bloodfloAV takes place. Or the reflex may be excited from distant skin areas, as illustrated in the curtailment in bloodflow observed when the opposite hand to that on which the observation is being made is placed in cold water. The CONTROL OF BODY TEMPERATURE AND FEVER 717 magnitude of the change in cutaneous circulation is nevertheless depend- ent upon the extent of the area of the body that is opposed to the change in temperature, as seen in the dilatation of the skin vessels prior to a rise in body temperature when a person is immersed in a warm bath. Although afferent impulses from the skin are therefore of great im- portance in adjusting the cutaneous blood supply according to the amount of surface cooling that has to occur, a further effect is also pro- duced on them by the action on the nerve centers of temperature dif- ferences in the blood itself. Thus, when the temperature of blood going to the brain is raised by placing the carotid arteries on some heating de- vice or when the region of the corpora striata is directly warmed, the skin vessels become dilated as if the animal had been exposed to general warmth. When the loss of heat by radiation and conduction is no longer ade- quate to prevent a rise in body temperature, or when the processes can not operate on account of a high temperature in the environment, the loss of heat from the skin is mainly dependent upon the evaporation of sweat. Under ordinary conditions this evaporation takes place at such a rate that there is no visible perspiration on the surface of the body — the so-called insensible perspiration. "When the heat loss by this channel must become greater, the perspiration is produced in larger amount, so that it collects on the surface of the body ; and, provided the conditions of the environment are such that evaporation can readily take place (low relative humidity), the amount of cooling of the body that can be effected becomes very great. A man may exist without any marked rise in body temperature in a very hot environment even when he is exposed to an out- side temperature that is the same as that of his body, or even greater. To encourage evaporation, however, he should be naked or very lightly clad, and the air should be kept in constant motion so that the layers of air next to the skin, which ordinarily very quickly become saturated with vapor, are transferred and replaced by dryer air. Movement of the air also increases the heat loss by conduction, provided the temperature of the air is not too near that of the body. The importance of the movement of air in the regulation of heat loss has been clearly demonstrated by Leonard Hill,54 F. S. Lee, and others, who have found that a great part of the discomfort experienced by living in stagnant air can be obviated by putting the air in motion by electric fans without doing anything to improve its chemical purity. In one famous experiment a number of young men were placed in an air-tight cabinet at the ordinary temperature of the room. After a time they began to exhibit the symptoms usually attributed to polluted air; they became drowsy and some of them developed headaches, etc. A small electric 718 METABOLISM fan was then started so as to set the air in motion. Immediately all of the men recovered and remained in a perfectly comfortable condition so long as the fan was kept going. The practical application of these facts to the hygienic control of the working conditions in mine shafts, in submarines, in workshops, etc., will be self-evident. The stimulus to increased SAveating seems to be dependent mainly on changes in the temperature of the Wood; for sweating does not im- mediately set in when the body is subjected to heat, as by a warm bath or a hot pack. It usually takes from ten to twenty minutes after the person has been placed in the bath or surrounded by the warm blankets of the pack before SAveating becomes pronounced. It can usually be sho\vn that before it sets in the body temperature has been raised from 0.1 to 0.8 degrees C. (0.2 to 1.4 degrees F.). In this regard, therefore, the response of the SAA'eat glands does not occur so promptly as does the dilatation of the cutaneous vessels. Loss of heat by evaporation of SAA'eat occurs only in certain animals. It is practically absent, for example, in the dog. The degree to which it may occur also varies in different individuals of the same species. The power of Avithstanding high temperatures is proportional in man to the facility AAdth Avhich he perspires. "Where SAveating is interfered Avith by skin diseases, — by ichthyosis, for example, — exposure to heat or in- creased heat production, as by muscular actiAdty, may raise the body temperature to a dangerous degree. Another factor upon Avhich the efficiency of eAraporation of sweat in cooling the body depends is the relative humidity of the air. When this is high, evaporation of Avater into it can not occur, and it is on this account that an increase in body temperature is much more likely to occur in AArarm, humid atmospheres than in those that are dry. At the same temperature people can live in perfect comfort in the dry air of the open plains, but suffer immediately from rise of temperature Avhen they go into the humid air of the river valleys. Similarly, work in hot fac- tories or in mines is quite possible at very high temperatures if the air is kept dry and in motion, but becomes impossible when the air is moist. In judging of the adequacy of air from this point of vieAV, it is there- fore important to take not the ordinary dry-bulb thermometer reading but that of the Avet-bulb.* In animals, like the dog, that do not perspire over the surface of the body, vaporization of the water in the expired air is the most important method of regulation of heat loss. When such an animal is exposed to *The wet-bulb thermometer registers a temperature that is lower than that of the dry-bulb in proportion to the relative humidity of the air. When the air is completely saturated with moisture, the temperature recorded by the two instruments will be the same; when it is perfectly dry, the difference will be maximal. CONTROL OF. BODY TEMPERATURE AND FEVER 719 warmth or when the region of the corpora striata is artificially warmed, the breathing immediately becomes much quicker and deeper, so that pulmonic ventilation is greatly increased and much more water is carried out as vapor with the expired air. To vaporize the water large quanti- ties of heat are required (seen in the latent heat of steam). In man this method is, ordinarily, not of great importance, but it may become so when sweating is interfered with, as in ichthyosis. The more rapid breathing also facilitates cooling by increasing the conduction of heat from the mucous membranes of the tongue, mouth, throat, etc. The im- portance of this method of cooling has been shown by finding that after the introduction of a trachea! cannula a dog can not withstand an in- crease of external temperature nearly so well as a normal animal. There are many other questions concerning the control of heat loss from the human body that might be considered, but it is scarcely nec- essary to do so here. It should be added, however, that the relative humidity of the air in the control of heat loss has a different significance when the temperature is high from that when it is low. High relative humidity at high temperatures, as we have seen, interferes with evapora- tion of sweat, whereas high relative humidity at low temperatures in- creases the heat-conducting power of the air and therefore tends to cool off the surface of the body by greater conduction. It is on this account that it is much more comfortable to live at a low temperature when the air is dry than when it is moist. On the dry plains of the West a tem- perature of many degrees below zero causes less sense of cold to be ex- perienced than in the moist atmosphere at a considerably higher tem- perature along the Great Lakes and in the river valleys. THE CONTROL OF TEMPERATURE In the case of man the body temperature is very largely under volun- tary control, as by the choice of clothing and the artificial heating of the room. Desirable as this voluntary control of heat loss may be, there can be little doubt that it is often managed to the detriment of good health. Living in overheated rooms during the cooler months of the year so diminishes the loss of heat from the body that the tone and heat-produc- ing powers of the muscular system are lowered. Not only does this diminish the resistance to cold, but it causes the food to be incompletely metabolized so that it is stored away as fat. The superficial capillaries also become constricted and the skin bloodless and "pasty." It is not looks alone that suffer, however, but health as well, for by having so little to do the heat-regulating mechanism gets, as- it were, out of gear, 720 METABOLISM so that when it is required to act, as when the person goes outside to the cold air, it may not do so as promptly as, it should, with the result that the body temperature falls somewhat and catarrh, etc., are the result. There can be little doubt that much of the benefit of open-air sleeping is owing to the constant stimulation of the metabolic processes which it causes. As will be inferred from what has been said above, the control between heat production and heat loss is effected through a nerve center located in or near the corpora striata. In most animals, when the spinal cord is cut in the cervical region, the body temperature quickly falls unless artifically maintained. In the case of man, on the other hand, it has usually been observed, after accidental section of the spinal cord in the cervical region, that a rise in temperature occurs. In twenty-four un- complicated cases of spinal-cord injury in man, collected from the rec- ords of Guy's Hospital by Gardiner and Pembrey, it was found that nineteen showed hyperthermia (sometimes amounting to 43.9° C.), and only five, hypothermia (sometimes 27.6° C.). If the patient lived, the ultimate effect of the section, as in the lower animals, would no doubt be the loss of the power of maintaining a constant temperature. The extent to which the animal comes to behave as if cold-blooded after section of the spinal cord varies considerably according to the level of the lesion; if the cord is cut in the upper thoracic region, for example, the regulation against cold, although distinctly less efficient than normal, is far better than when the section is made through the cervical cord. This difference is dependent on the fact that after the lower lesion much larger muscular groups and skin areas are left intact, so as to make regulation possible. Section of the dorsal cord in mice has been found by Pembrey to abolish entirely the increased metabolism which occurs in normal mice when they are exposed to cold. In the light of these experiments it is probable that the difference in the effects produced on body temperature by section of the cervical spinal cord in man and the lower animals depends on the relative im- portance of the heat-producing and heat-dissipating mechanisms. When the control of heat loss is paralyzed in the smaller animals, the cooling of the body becomes excessive in relation to the amount of heat produced in the paralyzed muscles, because the body surface is extensive in com- parison with the body weight (see page 551). In the larger animals such as man, on the other hand, the cooling effect is much less marked, espe- cially when, as is common after such an accident, the patient is kept unusually warm. CONTROL OF BODY TEMPERATURE AND FEVER 721 FEVER The clinical application of a Knowledge of the mechanism of heat regu- lation in the animal body concerns the causes of fever. In the most familiar form fever is produced by infectious processes, but it may also be owing to various other causes, among which may be mentioned the parenteral injection of foreign protein, excessive destruction of protein substances in the body itself, the action of certain drugs, and lastly, injury to the base of the brain or lesions of the upper levels of the spinal cord. Various types of fever are recognized: when the temperature re- mains constantly above the normal, it is known as continuous fever; when oscillations occur but the temperature never falls to the normal level, it is known as remittent; when it attains the normal level at cer- tain periods during the day, it is known as intermittent. Causes of Fever During a sudden rise in temperature -there is, on the one hand, in- creased heat production in the muscles, and on the other, dimin- ished heat loss from the surface of the body. The fever is therefore due to an exaggeration of the processes by which the body normally re- acts to conditions which tend to lower the body temperature. The increased muscular activity thus induced often causes visible contractions, familiar as shivering; and the constriction of the cutaneous blood vessels pro- duces the subjective sensation of chills, and causes the skin to become pale and cold to the touch. The skin muscles also contract, producing "goose skin." During this stage, objective demonstration of the cur- tailment of the skin circulation can be secured by observation of the bloodflow. through the hands and feet (page 283). When the temperature suddenly falls again, the crisis, as it is called, muscles become ' flaccid and produce less heat, and the cutaneous blood vessels dilate, as has been shown by measurements of the bloodflow of the hands and feet. At the same time also, the sweat glands are stimulated and marked per- spiration occurs. Concerning the cause of continuous fever, it must be assumed that the balance between heat production and heat loss has been adjusted at a higher plane than normal. We can not explain the fever on the basis either that heat production is permanently increased or that heat loss is permanently diminished, for in neither of these cases would the tem- perature stand at a permanent level but would steadily rise or fall, ac- cording to which mechanism was disturbed. .While set at this higher plane of fever, the thermogenic nerve centers are still capable of re- sponding in the usual way to the influences which cause the body tern- 722 METABOLISM perature to change in a normal person. For example, when a fever pa- tient is subjected to a hot bath so that his body temperature rises about 0.2 to 0.5 degrees C., sweating occurs just as in a normal individual; or if exercise is taken the. increased amount of heat thereby produced in the muscles is dissipated in the usual way. When, on the other hand, the patient is exposed to cold, the vessels of the skin contract and he shivers. Although fever is not caused by an actual disturbance of balance be- tween heat production and heat loss, neither of these processes is pro- ceeding at its normal level. That there is a distinct increase in the total heat production of the body in acute fevers in well-developed persons has been shown by means of the respiration calorimeter. This increased heat production is not observed in patients who have been brought into a weakened condition and in whom the muscular tissues have become atrophied by long-continued fever. The increased heat production in continuous fever is mainly dependent upon the increase in body tem- perature and is not one of its causes, as is evident from the fact that far larger quantities of heat are frequently produced in normal individuals as a result of muscular exercise or the taking of large quantities of protein-rich food. The heat thus produced is, however, very quickly dissipated, so that only a temporary rise in temperature occurs, (cf. Hewlett.57) Similarly, it can be shown that in continuous fever there is a relative inefficiency in the mechanism of heat dissipation. When the temperature of a normal person is artificially raised through about 1° C., a marked increase in cutaneous bloodflow and profuse perspiration are invariably noted. In a patient with fever of the same degree, on the other hand, there is practically no change in the skin circulation; indeed, it is usually diminished, and there is no unusual perspiration. The heat-regulating mechanism is now fixed on a plane that is higher than the normal, so that although further increase in body temperature, as we have seen, calls forth responses like those in a normal individual, yet at the fever temperature itself there are none of the reactions which a normal individ- ual would exhibit if his temperature were artificially raised to that level.57 The adjustment of the temperature at the higher level is by no means so perfect as it is at the normal level of health, so that a normal subject is more resistant to the effects of cold than is a patient with fever. The degree of response of the fever patient, however, varies considerably from time to time ; a cold bath in typhoid fever, for example, lowers the body temperature much less effectively at an early stage in the disease, when the fever is more or less continuous, than later when it is becoming of the intermittent type. In the third week of the disease the cold bath CONTROL OF BODY TEMPERATURE AND FEVER 723 more readily brings down the temperature and keeps it down for a longer time than during the first or second week. The mechanism for heat loss is also deranged in fever, which explains the rise in temperature that is likely to follow the performance of even moderate muscular exercise or the taking of too hearty a meal in tuberculous and convalescent typhoid patients. Changes in the Body During Fever In seeking for the cause of fever which is evidently of an obscure nature, it is necessary to collect all the information we can regarding the metabolic changes that are then occurring in the animal body. A few of the most significant facts that have so far been collected may be mentioned here. Some of the most important concern the dis- turbance in nitrogenous equilibrium caused by the considerable loss of nitrogen which takes place in fever patients when they are fed on the usual hospital diet prescribed for such cases. This loss of nitro- gen is.no doubt the result of the partial starvation in which the pa- tient is kept; for it has been shown by Shaffer and Coleman-"'5 that patients with typhoid fever may be maintained in nitrogenous equi- librium by feeding them with relatively large amounts of carbohy- drate, which acts by protecting the protein of the body from disintegra- tion (see page 571). Even with a diet excessively rich in carbohydrates that no more than covers the calorie requirements of the patient, nitrog- enous equilibrium has also been attained. The protein minimum to which fever patients can be reduced is nevertheless considerably higher than the minimum in normal individuals. From the above results as a whole, it is probably safe to conclude that there is a specific destruction of protein going on in the body during fever. Further evidence of such a destruction is furnished by the presence in the urine of excessive amounts of creatinin, of purine bases, and, it is said, of incompletely hydrolyzed proteins, such as the albumoses (pro- teoses.) Moreover, when the fever suddenly terminates in crisis, there is a marked increase in the excretion of urea (the epicritical urea in- crease), which indicates that an extensive deamination of protein build- ing stones (ammo acids) is occurring. The so-called "diazo reaction" obtained in the urine during the fever is also believed to depend on the presence of abnormal protein-disintegration products. As to the specific cause of the increased protein disintegration, little is known. Several factors may operate: (1) the partial starvation of the patient, entailing an increased breakdown of protein to meet the calorie requirements; (2) the high temperature, which in itself may stimulate increased protein metabolism, for it has been shown that, when normal 724 METABOLISM animals are artificially warmed, protein metabolism becomes increased ; and (3) toxic protein-decomposition products specifically causing an ex- cessive breakdown of protein. Although there is increased protein breakdown during fever, it must, not be forgotten that only about 20 per cent of the total expenditure of the body is derived from this foodstuff, 80 per. cent coming from non- nitrogenous material, which must be fat, because the available carbo- hydrates are used up at an early stage. Since the general metabolism is increased, the excessive breakdown of the fatty substances, occurring as it does in the presence of a diminished combustion of carbohydrates, interferes with the proper oxidation of the fatty-acid molecules and leads to the appearance of so-called acidosis products in the urine, and consequently to a relative increase in the urinary ammonia (page 616). A tendency to acidosis therefore exists. The acidosis may reach a considerable degree of severity and cause the tension of carbon dioxide in the alveolar air to become diminished. Since a similar degree of acidosis may be produced in partially starved ani- mals by overheating them with moist air, but not so if the animals are liberally fed with carbohydrates, it is probably safe to conclude that abundance of carbohydrate is advisable in the food that is furnished to fever patients. Another interesting metabolic change in fever concerns the salt bal- ance. This is studied by observing the amount of sodium chloride excreted by the urine. As is well known, this becomes markedly diminished until the crisis of the fever, when it suddenly increases. Salt retention is more marked in certain types of fever than in others, and it is essentially dif- ferent in nature from the salt retention that has been observed to occur in nephritis. This difference has been brought to light by examination of the chloride content of the blood. In nephritis, the concentration of chlorides in the blood is considerably increased, whereas in fever it is markedly diminished. The deficiency in salt elimination can not be at- tributed to a deficiency of salt in the food, for it sets in before the diet has been curtailed and, when salt is given to a febrile patient, it is re- tained in the body to a greater degree than is the case in the normal individual. For some reason the tissues in fever have acquired the property of retaining large quantities of salt. • Attempts to study the water balance during fever have frequently been made, but the technical difficulties of such investigations make the re- sults uncertain and of little value. That some retention of water occurs during fever is, however, evidenced by the dilution of the blood. At the crisis this hydremia quickly disappears at the same time as the increased CONTROL OF BODY TEMPERATURE AND FEVER 725 elimination of chlorides is going on. Chlorides and water would there- fore seem to behave in a similar fashion during fever. The Heat-regulating Center In all discussions on the regulation of hody temperature and the causes of fever, it is assumed that a heat-regulating or thermogenic center exists somewhere in the brain. It is believed to be located about the optic thalami or corpora striata, for it has been found in rabbits that destruction of the brain anterior to this region does not cause any change in body temperature, whereas destruction behind it is followed by an entire upset in the heat-regulating mechanism. Fur- thermore, artificial puncture of this part of the brain causes marked elevation in body temperature in rabbits (heat puncture). Most in- teresting experiments have been recorded by Barbour,56 who succeeded in applying heat or cold locally in the region of the centers. By the application of cold, increased muscular metabolism/ on the one hand, and diminished heat loss, on the other, were excited; and conversely, when warmth was applied, an increased heat loss and a diminished heat production were observed. Irritation of this region of the brain in man, as after cerebral hemorrhage, is also accompanied by remarkable dis- turbances in heat regulation. It is believed by many that the essential cause of infectious fever is an action on these centers by toxic substances which develop in the blood. The centers may also be acted on by various drugs, some of which excite them to increase the body temperature, others, to lower the tem- perature when this has already been elevated. When solutions of sodium chloride are injected intravenously or subcutaneously or even sometimes, particularly in children, when administered by mouth, more or less fever may result. This must be a specific action of the Na ion, for, if instead of pure solutions of NaCl. solutions containing calcium and potassium salts as well as those of sodium are injected, no fever is induced. This fact, taken along with the close similarity between puncture diabetes and heat puncture, lends support to the view that in its initial stages experimental fever of this type is the result of an excessive breakdown of glycogen in the liver. It must not be imagined, however, that persist- ent fever can be attributed to such a cause, since the fever remains after the glycogen has all been removed. Other chemical substances produc- ing fever are caffeine, certain other purines, and particularly tetra-hydro- naphthylamin. Belonging to this group of fevers must also be considered the im- portant ones produced by the intravenous injection of certain forms of protein, as those of egg white or those derived from the bodies of bac- 726 METABOLISM teria or from the laked corpuscles of a foreign blood. The fever in these cases is 110 doubt caused by a mechanism closely related to that responsible for anaphylaxis (see page 89). Such injections do not pro- duce fever in animals after division of the cervical spinal cord or ex- cision of the midbrain. It is believed that many cases of so-called asep- tic fever, occurring after severe contusions or other wounds, may be the result of destruction of proteins within the body. Similarly the rise in temperature during infections may be owing to the breakdown protein of the microorganism in the cells. Significance of Fever in the Organism It is impossible at present to state definitely whether fever is a re- action of the organism against some infection and therefore of benefit in assisting the organism to combat it, or whether it is in itself an un- favorable condition. The question can certainly not be answered by observing the behavior of bacteria growing at different temperatures in various media outside the body. That certain bacteria should be found not to thrive at incubator temperatures equal to those found in the body during fever, does not at all prove that this fever is of sig- nificance as a means of combating the growth of the bacteria in the body. It is undoubted that, where the body temperature becomes ex- cessively high, the correct treatment is to keep it down as much as possible. On the other hand, the reduced mortality that has followed the introduction of the cold-bath treatment in typhoid fever may not be due so much to the reduction in body temperature itself as to the favorable effect produced on the nervous system and circulation. We certainly know that in normal animals moderate degrees of hyper- pyrexia produced by exposure to moist heat are well borne for consider- able periods of time, thus indicating that it is the infection and not the hyperthermia that causes the serious damage to the body in infectious fevers. METABOLISM REFERENCES (Monographs and Original Papers) iLusk, Graham: The Elements of the Science of Nutrition, W. B. Saunders Co., ed. 3, 1917. ^Cathcart, E. P.: The Physiology of Protein Metabolism, Monographs on Bio- chemistry, Longmans, Green & Co., 1912. ^Taylor, A. E.: Digestion and Metabolism, Lea & Febiger, New York, 3912. iUnderhill, P. P. : The Physiology of the Amino Acids, Yale Press, New Haven, 1915. sMacleod, J. J. E.: Diabetes, Its Pathological Physiology, E. Arnold, 1913. saFiirth, von: The Problems of Physiological and Pathological Chemistry, etc., J. B. Lippincott Co., 1916. 5b Jones, W.: Nucleic Acids, Monographs in Biochemistry, Longmans, Green & Co., 1914. scMendel, Lafayette B.: Ergebnisse der Physiologic, 1911. CONTROL OF BODY TEMPERATURE AND FEVER 727 sdLeathes, J. B.: The Fats, Monographs in Biochemistry, Longmans, Green & Co. seMathews, A. P.: Physiological Chemistry, Wm. Wood & Co., 1917. s'Dakin, H. K. : Oxidations and Reductions in the Animal Body, Monographs in Bio- chemistry, Longmans, Green & Co., 1912. 5gLeathes, J. B.: Problems in Animal Metabolism, 1906. 6Du Bois, E. F., and collaborators : Clinical Chemistry, Papers 1 to 25, Arch. Int. Med., 1915-17, xvi-xix. 7 Benedict, F. G.: Am. Jour. Physiol., 1916, xli, 275 and 292. sMendel. Lafayette B.: Harvey Lecture, J. B. Lippincott Co., 1914-1915, p. 101. sMcCollum, E. V., and collaborators: Numerous papers in Jour. Biol. Chem., be- ginning 1913. "Hopkins, F. Gowland, and Willcock, E. G.: Jour. Physiol., 1906, xxxv, 88. nBayliss, W. M.: The Physiology of Food and Economy in Diet, Longmans, Green & Co., 1917. i2McCollum, E. V.: Harvey Lecture, Jour. Am. Med. Assn., 1917. "Sweet, J. E., Carson-White, E. P., and Saxon, G. J.: Jour. Biol. Chem., 1913, xv, 181; ibid., 1915, xxi, 309. "Stepp, W.: Biochem. Ztschr., 1909, xxii, 452. isFunk, Casimir: Ergebnisse der Physiologic, 1915. ifiMcKillop, M.: Food Values: What They Are and How to Calculate Them, Rutledge. K'aMcCoy, D. Major: The Protein Element in Nutrition, E. Arnold, London, 1912. "Pembrey, M. S.: Chemistry of Eespiration, in Schafer's Text Book of Physiology, 1898, i. isAllen, F. P.: Glycosuria and Diabetes, Boston, 1913. isJoslin: Diabetes. aoWoodyatt, R. T., Sansum, W. D., and Wilder, R. M.: Jour. Am. Med. Assn., 1915, Ixv, 2067. Also Taylor, A. E., and Hulton, F.: Jour. Biol. Chem., 1916, xxv, 173. ziMacleod, J. J. R., and Fulk, M. E.: Am. Jour. Physiol., 1917, xlii, 193. 22Hamman, L., and Hirschbaum: Arch. Int. Med., 1917, xx, 761-788. 23Cannon, W. B.: Bodily Changes in Pain, Hunger, Fear and Rage, D. Appleton & Co., 1915. 2 latter that is of importance in the present connection. Chromaffin tissue is found in most animals along the front of the aorta, between the renal arteries, where it can usually be recognized by staining the tissue with chromic acid. Sometimes accessory chromaffin tissue is located in distant parts, as in the epididymis of the rat,- for example. It is said that life can be maintained if one-eighth of the total amount of the adrenal substance be present in the body. Attempts to prolong life after adrenalectomy by adrenal transplantation have almost invariably met with negative results, because the graft undergoes a rapid process of necrosis and dis- appears; although it is said that transplantation may sometimes be suc- cessfully accomplished if the grafting is done into the kidney. Adminis- tration of suprarenal extract is also without definite benefit after adrenalectomy. 734 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS Suprarenal Extracts — Preparation Injection, particularly intravenous, of extract of the adrenal gland has furnished us with most of the evidence upon which our knowledge regarding the function of this organ depends. Such an extract is best made by grinding the entire gland with fine sand in a mortar and then extracting with a weak (decinormal) solution of hydrochloric acid. The extract may then be boiled, filtered through muslin and nearly neutral- ized, preferably by means of sodium acetate. If kept in this acid reac- tion, the active principle of the extract does not materially deteriorate with time, but if it be neutralized or considerably diluted, destruction due to oxidation occurs, as evidenced by -a distinct browning of the solution. The active principle of such extracts lias been isolated in a crystalline form (Takamine and Abel). It has been given various names (adrenalin, suprarenin, adrenin, etc.), but the tendency is definitely towards the use of epinephrine. Chemically, epinephrine has been found to be orthodioxyphenylethylolmethylamine. H0_ \ -*CH(OH) - CH,NHCH3. It will be noted that it is closely related to tyrosine (see page 604). It is also closely related to a group of substances (amines) occurring in putrid meat and to which the active principles of ergot belong. It contains an asymmetric carbon atom (asterisked in formula), which indicates that there must be three varieties of epinephrine, differing from one another in the effect which they produce on the plane of polarized light (i.e., a dextro- and a levo-rotatory and a racemic form). Epinephrine can be prepared by synthetic means, the first product of this synthesis being the racemic salt, which can then be split lay appro- priate methods into dextro- and levo- varieties. The levo- variety ap- pears to be identical in its pharmacologic action with the natural product. The dextro- variety on the other hand has only poorly developed physio- logic activities (about seven per cent that of the levo- variety), while the racemic variety comes in between the two in its action. A valuable assay of the amount of epinephrine in tissue extracts can be made by the method of Cannon, Folin and Denis,62 in which an acid extract of the gland is treated with phosphotungstie acid, and the blue color thereby developed compared colorimetrically with a standard blue. Physiologic Action The physiologic effects of the intravenous injection of epinephrine are markedly excitatory and slightly inhibitory in nature. We will consider THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS 735 the excitatory action first. Immediately after the intravenous injection of as small an amount as 0 '00008 milligrams per kilogram of body weight, a distinct rise in arterial blood pressure may be observed. It is quite def- inite with 0 '00008 milligrams per kilogram, and is accompanied by a slowing of the pulse. This slowing is caused by stimulation of the .vagus center, as is evidenced by the fact that if the vagus nerves are cut, or sufficient atropine administered to paralyze them, the same dose of epinephrine produces not a slowing but a quickening of the pulse, and consequently a much greater rise in blood pressure. The vagus action is developed not because of an effect of epinephrine on the vagus center, but secondarily because of the rise in blood pressure. These preliminary experiments indicate that the locus of action of epinephrine, so far as the circulatory system is concerned, is mainly on the small blood vessels, constricting them and thus raising the peripheral resistance. This conclusion can readily be confirmed by applying the epinephrine directly to the blood vessels of the exposed mesentery, or by enclosing a vascular organ such as the kidney in a plethysmo graph during the injection of epinephrine, when a great diminution in volume, accompanying the rise of arterial blood pressure, will be observed. The vasoconstricting effect of epinephrine does not become developed on the large blood vessels near the heart on account of the deficiency in muscu- lar tissue in their walls. Indeed, these vessels may become passively .dilated because of the increased blood pressure. The arterioles of dif- ferent parts of the circulation are not equally sensitive to epinephrine; those of the splanchnic area are most sensitive, whereas those of the heart — the coronary vessels — do not respond at all in most animals (see page 257). The pulmonary and cerebral vessels have a variable reactivity to epinephrine. The effect on the vessels persists after complete destruction, not only of the central nervous system, but also of the vasomotor nerves; epi- nephrine still acts, for example, on vessels the nerve fibers of which have been allowed to degenerate by cutting them several days before the epinephrine is applied. This would seem to indicate that the epinephrine acts directly on the muscular tissue in the Avails of the blood vessels, but this does not appear to be the case, for it has been found that epi- nephrine is incapable of acting on tissues which are devoid of sympathetic nerve fibers, and is also inactive on those tissues in the embryo which have not yet received any nerve supply. In brief, then, although epinephrine acts only on blood vessels that are supplied by the sympathetic nervous system, it is not on the nerve fibers that the epinephrine unfolds its action. "We shall see immediately that this conclusion is in conformity 736 THE ENDOCRINE ORGANS, OR 'DUCTLESS GLANDS with the results of observations made on structures other than the blood vessels. Other muscular structures excited by epinephrine are as follows: (1) the dilator muscle of the pupils, especially after the nerve supply has been destroyed by extirpation of the superior cervical ganglion; (2) the sphincters of the pylorus and of the ileocecal valve; (3) the muscle fibers of the spleen, the vagina, the uterus, the vas deferens, and the retractor penis. Eegarding the action on the uterus, however, it should be noted that a different response may be obtained according to whether the uterus is pregnant or not. The plain muscles of the orbit and globe of the eye are sometimes excited by suprarenal extract, causing the eyes to protrude, the palpebral fissure to become large and the third eyelid to be retracted, changes which are very like those which develop as a result of 'fright. Inhibitory effects of epinephrine on muscle are exhibited by the follow- ing: (1) the muscle of the intestine; (2) the stomach; (3) the esophagus; (4) the gall and urinary bladders. The effect of epinephrine in inhibiting the rhythmic contractions of an isolated portion of the intestine in oxygenated Ringer's solution is a very striking phenomenon, and one which, as we shall see, may be very successfully employed for detecting small quantities of epinephrine. Extremely dilute solutions of epinephrine increase the contractions. The effects of epinephrine on glandular structures are the same as those which would be produced by stimulation of the sympathetic nerve supply of the gland. Thus, the secretions of the lachrymal gland, the salivary gland (in the cat), the mucous glands of the mouth and pharynx, the gastric but not the pancreatic glands, can readily be shown to be excited. From these results as a whole, it is evident that the effect of epineph- rine on muscles and glands is exactly the same as that which would be produced by stimulation of their sympathetic nerve supply. This paral- lelism of action between epinephrine and the sympathetic nervous sys- tem becomes still more evident when we consider certain of the changes in metabolism that follow administration of epinephrine. Injection of epinephrine excites glycogenolyste in the liver so that hyperglycemia and glycosuria become established, results which are also obtained by stimulating the great splanchnic nerve. Epinephrine causes the clotting time of the blood discharged from the liver to be very materially short- ened, an effect also produced by stimulating the splanchnic nerve.63 As in the case of the blood vessels, the above results are obtained even after the sympathetic nerves to the part have been allowed to undergo degeneration, from which it is concluded that the tissues elaborate some THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS 737 substance which reacts with epinephrine. This substance may be pro- duced either at the junction between the nerve and muscle — the myo- neural junction, — or perhaps throughout the protoplasm itself. It is called the receptor substance of Langley, and is believed to react not only with epinephrine, but also with various drugs. The receptor sub- stance seems to increase, if not in amount, at least in sensitivity after the removal of the nerve control. Ergotoxin, which is an amine obtained from ergot and also from cer- tain of the products of histidine, has an action on the receptor substance which is inhibitory and therefore antagonistic to that of epinephrine. The antagonistic action of ergotoxin affects the excitatory but not the inhibitory actions of epinephrine. By using this drug we are en- abled to show that, although the main effect of epinephrine on tissue is excitatory, a less marked inhibitory influence may be simultaneously developed. The inhibitory effect may also sometimes be evoked by doses of epinephrine very much smaller than those used to produce excitatory effects. These facts are well illustrated in the case of the muscle fiber of the blood vessels. With an ordinary dose of epinephrine constriction occurs; after ergotoxin the same dose of epinephrine causes dilatation. Or this latter result may also be obtained by administer- ing to a normal animal quantities of epinephrine that are very much smaller than the usual quantity. The coexistence of inhibitory and ex- citatory influence is also well noted in the case of the uterus. In some animals the effect of epinephrine on this organ is to augment its rhythmic contractions, in others to inhibit them. In the former case, however, if ergotoxin is first of all administered, epinephrine in its usual dosage will invariably produce an inhibitory effect. The ergotoxin no doubt acts on the receptor substance, and similar effects have also been produced with apocodeine. Although it is especially on plain muscular fiber having a sympathetic nerve supply that epinephrine unfolds its action, yet, according to Can- non, it increases the contracting power of voluntary muscle and dimin- ishes the tendency to fatigue.* "For further details of these effects the papers of Iloskins"-' and IIartniaiir'1 should be consulted. CHAPTER LXXXII THE ADEENAL GLANDS (Cont'd) Variations in Physiologic Activity Since it is clearly established that the adrenal glands are indispensable to life and that extracts of them have a v6ry pronounced physiologic action, it remains to consider whether the glands produce this internal secre- tion within the body, and if so, whether it is essential for the well-being of the animal or required only under certain conditions. We must also endeavor to find out upon which of the bodily functions of the intact animal the internal secretion acts. These problems have been attacked by three methods of investigation: (1) by comparing the epinephrine content of similarly prepared extracts of the resting gland and of one removed after a period of supposed increased activity; (2) by collecting the blood as it flows into the vena cava from the adrenal vein and ex- amining it for epinephrine by physiologic tests. These consist in observ- ing the behavior of some tissue that is sensitive to -the action of epineph- rine, such as the intestine or uterus, after applying the .blood or serum to it, or by injecting the blood or serum intravenously into another ani- mal and looking for epinephrine effects; and (3) by allowing the blood of the adrenal vein to be discharged under certain conditions through the vena cava into the blood vessels of the same animal, and observing the effect produced on certain physiologic processes which in one way or another have been sensitized toward the influence of epinephrine. This autoinjection method has recently been used successfully by Stew- art and Rogoff,66 their favorite structure upon which to observe the epinephrine effect being the denervated pupil. Assaying the Epinephrine Content of the Gland With regard to the first mentioned of the methods, either chemical or physiologic means may be employed to assay the strength of the ex- tracts. The best chemical method is that of Cannon, Folin and Denis,ea the principle of which has already been described. The physiologic method yielding most satisfactory results is that of Elliott,67 which con- sists in injecting a portion of the extract intravenously into animals from which the influence of the nerve centers on the heart and blood vessels has been removed by decapitation. The rise in arterial blood 738 THE ADRENAL GLANDS 730 pressure produced by the injection is then a very fair measure of the amount of epinephrine contained in it. It has been shown that the re- sults obtained by the chemical method agree very closely with those obtained by the physiologic, but it should be remarked that it is difficult to see how the physiologic method could be accurate in all cases, vsince it has been shown that with great dilution of epinephrine a reversed effect — a vaso- dilatation — may be obtained. Attempts to assay the strength of an epinephrine solution by investigating the effects which it produces on other preparations, such as isolated loops of intestine or uterus; or the enucleated eyeball of the frog, must prove unsuccessful, since the effects are by no means dependent on the concentration of epinephrine in the extract. When such preparations are used for quantitative purposes, the strength of the extract must be judged by finding the extent to which it can be diluted and still remain active. Quite apart from the foregoing possible sources of error, it must be remembered that the results merely give us an idea of how much epineph- rine may have been contained in the gland at the time of its excision. They can not tell us how much epinephrine the gland was secreting. Prior to excision as much of this hormone might have been undergoing a process of manufacture in the gland as was being discharged from it, so that the assayed amount would represent merely the balance of production and loss of hormone by the gland. We might quite well find that the amount of epinephrine in the excised gland was normal under conditions where there had been an excessive discharge of it into the blood; that is to say, loss and production might have been equal. Where, however, a marked deficiency is found to exist, it probably indicates that exhaustion of the power of producing epinephrine was taking place. The Epinephrine Content of the Blood. — The second method, in which blood from one animal is tested for its epinephrine effect by intravenous injection into another animal or by applying it to some isolated prepara- tion on which epinephrine acts, has yielded important results. Since serum contains all the epinephrine of blood, it can be conveniently used for the tests (Stewart and Rogoff). The isolated physiologic prepara- tions that have been used in testing for epinephrine in the animal fluids are as follows: 1. A segment of the small intestine of a rabbit, suspended in oxygen- ated Locke's solution at body temperature. 2. A segment of the uterus of a nonpregnant rabbit similarly prepared. The apparatus used for observing the contractions of either prepara- tion consists of a small glass chamber furnished below with a hook to which one end of the segment is attached, the other end being connected 740 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS to a muscle lever, so that the regular rhythmic contractions can be regis- tered on a drum (Fig. 190). Epinephrine inhibits the contractions of the intestine but stimulates those of the uterus of most animals, the intestine preparation being the more sensitive (Fig. 191). Indeed, it is said that the inhibition in this case may be obtained with a solution containing 1 part of epinephrine in 20,000,000 of solution. In using this method, however, great care and judgment must be exercised in drawing conclusions, because other sub- stances present in the blood are liable to affect the contractions; thus, Air. vent Metal waterbath 38'c. Harvard muscle warmer with radiated scale n. metal heating rod soldered in wall of water bath Fig. 190. — Arrangement of apparatus for recording contractions of a uterine strip, intestinal strip, or ring, etc. The metal water-bath is made of a cheap metal water-pail with a heating rod soldered through the side at the bottom. A short metal tube is soldered into a 1-inch opening in the bottom to receive a perforated cork for connecting with the Harvard muscle-warmer inside. (From Jackson.) certain substances in blood serum which have been produced by the act of blood clotting may cause augmentation of the beat in both the intes- tinal and the uterine preparations. A certain amount of epinephrine in Locke's solution is consequently more likely to cause inhibition of the intestine than a similar amount added to blood serum, because in the lat- ter case the pressor substance will neutralize the depressor effect of the epinephrine. On the uterine preparation, both the blood serum and the epinephrine have pressor effects. As has been pointed out by G. N. Stewart,68 if both preparations are employed for testing a solution sup- THE ADRENAL GLANDS 741 posed to contain epinephrine, little chance of error is likely to be in- curred; that is, if the solution produces inhibition of the intestine along with augmentation of the uterus, it must contain epinephrine. 3. The fresh carotid artery of the sheep. A ring cut from the artery is suspended in oxygenated Locke's solution and attached below to a Fig. 191. — Tracing showing the effect of epinephrine on the intestinal contractions and on the arterial blood pressure. (The preliminary addition of barium to the nutritive fluid may be disre- garded.) (From Jackson.) small hook and above to a light muscle lever, by which the contraction of the muscle fibers can be observed. Epinephrine causes the muscle to contract, but the test is not so sensitive as the foregoing, especially in the presence of blood serum, because the pressor substances therein con- tained also cause contraction. Blood plasma does not contain the pres- sor substances, so that oxalated plasma should be used in place of serum 742 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS in applying the test. 'To increase the sensitiveness of the muscle, the artery ring should be slightly stretched. by loading the lever. 4. The Wood vessels of a frog. This method depends on the same prin- Funnel, or small pressure bottle Hook through lower jaw Cannula In one aorta I'tg. 192. — Arrangement of apparatus for pcrfusion of the vessels of a brainless frog. (From Jackson.) ciple as in that just described. The fluid supposed to contain epinephrine is added to Locke's solution, which is meanwhile being perfused under constant pressure through the blood vessels and the rate of outflow THE ADRENAL GLANDS 743 noted (Fig. 192). If the fluid added to the inflowing fluid contains epi- nephrine, the outflow will become diminished. This is a very satisfactory method, although it is somewhat limited in scope unless large frogs are procurable, because of the difficulty of getting the necessary cannulas into the vessels (aorta and abdominal vein). 5. The pupil of the enucleated eye of the frog. Extremely small traces of epinephrine are observed to cause a dilatation. 6. The denervated iris. The fluid to be tested is placed in the conjunc- tival sac of an animal from which the superior cervical ganglion of the corresponding side has been removed some days previously. Under such conditions, if epinephrine is present in the fluid, dilatation of the pupil occurs. Both of the preceding reactions we owe to Meltzer.70 It should be emphasized that, although each of these methods is in itself very sensitive for the detection of epinephrine without being al- ways specific, yet the result should not be considered conclusive unless definite effects have been secured by at least two methods that are as far as possible independent of each other. As an outcome of investigations by these methods it has been found that, when blood from the adrenal vein is collected in a pocket of vena cava made by applying clamps above and below the entrance of the adrenal veins, the presence of epinephrine can be revealed, the rate of secretion being from 0.0003 to 0.001 mg. per kilogram of body weight per minute (Stewart and Rogoff). The absolute amount of epinephrine liberated from the gland can be measured only by finding the concen- tration in the adrenal vein blood and the rate of bloodflow. This amount is approximately constant, so that the concentration in the blood which collects in the cava pocket varies inversely with the rate of bloodflow. In asphyxia the bloodflow is decreased so that the concentration of epi- nephrine increases, but there is no change in the absolute amount. Nei- ther anesthesia nor trauma affects the amount. The concentration is likely to rise late in an experiment because of the slowing of bloodflow. Adrenal activity may, however, be excited by massage of the gland, or by stimulation of its nerve supply through the great splanchnic nerve. The presence of epinephrine in blood collected directly from the adrenal veins does not justify us in concluding that, when mixed with the re- mainder of the blood in the body, there would be a sufficient concentra- tion of this substance to develop any of its activities. It has therefore been necessary to devise methods by which this possibility could be tested. The Autoinjection Method. — Such a method was first of all success- fully used by Asher, who employed an animal from which all the abdom- inal viscera had been removed. On stimulation of the great splanchnic 744 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS nerve a rise in arterial blood pressure occurred provided the adrenal veins were open, but not so if the adrenal veins were clamped. By re- moving the viscera, the effect of splanchnic stimulation on the abdom- inal blood vessels themselves is eliminated, and any constriction which occurs in the blood vessels of the rest of the body must obviously be due to the action of epinephrine. The most satisfactory of these methods is that more recently employed by Stewart, Rogoff and Gibson,69 which consists in observing the be- havior of the pupil on the side from which the superior cervical ganglion has been removed about one week previously. Of course the blood pres- sure effect is also observed. Among the most important results secured by this method it may be mentioned that dilatation of the pupil occurs on stimulation of the great splanchnic nerve, provided the vena cava and adrenal vein are unobstructed so that the blood from the adrenal glands can get to the head. If the vena cava is clamped and the splanchnic nerve stimulated, there is no pupil- lary dilatation, but it immediately occurs after the clamp is removed. Epinephrine continues to be discharged for a considerable period of time after stimulating the splanchnic nerve, but the immediate increase which follows the application of the stimulus does not last long, so that more secretion can be obtained by intermittent than by continuous stimula- tion. It does not seem to be' possible to exhaust the adrenal gland of its supply of active material by stimulating the splanchnic — a fact which would seem to throw considerable doubt on the reliability of the con- clusions arrived at by the use of those methods in Avhich extracts of the gland are assayed (see page 739).* Many interesting facts concerning the nature of the innervation of the gland have been secured by one or other of the above methods. After section of the sympathetic chain and the great splanchnic nerves on both sides (in the thorax), no epinephrine is secreted into the blood of the adrenal vein, and when one gland is extirpated and the nerve connec- tions of the other entirely cut, the epinephrine content of the adrenal vein blood sinks to not more than 1/1000 of the normal amount. The animals survive this latter operation and behave in a perfectly normal fashion, indicating that the internal secretion of the adrenals can not have the physiologic significance so often ascribed to it. The splanchnic fibers concerned in the secretion of epinephrine seem to come from a nerve center situated relatively low down in the spinal cord. Section of the cord at the level of the last cervical segment does not affect the spontaneous secretion, but this disappears when the section is made beloAv the third thoracic segment. (Stewart and Rogoff.) "Another great advanlage of tl'e aufoinjection method is that no confusion can be caused by the development of prcssor substances through clotting. THE ADRENAL GLANDS 745 In connection with these observations it is of interest to note that dur- ing stimulation of the splanchnic nerve in a normal animal, the conse- quent rise in blood pressure shows two peaks (see Fig: 29, page 137). The first is no doubt due to direct stimulation of the splanchnic vasoconstric- tors, and the second to the outpouring of epinephrine into the blood, the justification for this conclusion being that the latter rise fails to appear after removal of the adrenal glands. Taking the results as a whole, it is indeed doubtful whether under nor- mal conditions a sufficient amount of epinephrine is discharged into the blood of the vena cava to affect appreciably the tone of the blood vessels, and this conclusion seems all the more justified because of the fact that small quantities of epinephrine have a dilating rather than a constricting influence, at least on certain vessels (Hartman64). It maybe, however, that the maintenance of vascular tone under certain conditions is greatly as- sisted by the presence of epinephrine in the blood. Similarly the sympa- thetic control of other functions may be facilitated by the presence of small amounts. It has been found, for example, that, although stimula- tion of the celiac plexus causes the glycogen stored in the liver to be con- verted into sugar, this result is not as a rule obtained on stimulating plexus shortly after removal of the adrenal glands. The presence of epinephrine in the blood would, therefore, seem to be necessary to bring about functional activity of the sympathetic nerve endings concerned in the glycogenolytic process (see page 637); Adrenalemia. — In the light of these researches it is important to point out that a great part of the work done by clinical observers purporting to show that in such conditions as nephritis and arteriosclerosis there is an increase of epinephrine in the blood, has been found by Stewart and others, using controlled methods, to be entirely unproven.70 Some inves- tigators, however, still hold that temporary conditions, such as transient rises of arterial blood pressure or temporary glycosuria, may sometimes be due to increased adrenal discharge into the blood. Ephinephrine has been thought to be a substance which is secreted into the blood in supernormal amount when certain emergencies arise, the most important of these being fright, or some other extreme emotion. This belief has arisen partly from the similarity between the general behavior of an animal following the intravenous injection of epinephrine and dur- ing states of extreme excitement. Dilatation of the pupils, bristling of the hair, salivation, rise in arterial blood pressure, inhibition of the intes- tinal movements, protrusion of the eyeballs are all symptoms of fear just as they are of epinephrine injection. Impressed by these resemblances Cannon72 undertook an extended research to test the hypothesis that the reaction of an animal to fear and other emotional states is dependent on 746 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS hypersecretion of epinephrine into the blood. The results seemed to con- firm the hypothesis. In the first place, it was found that, whereas the blood drawn from the vena cava opposite the entry of the adrenal veins (by passing a catheter up the femoral vein till its free end lay at this level) in a normal male cat did not give evidence of the presence of epinephrine when tested by means of the intestinal segment method, it did so in a cat that had previously been frightened by allowing a dog to bark at it. Such results were not obtained after removal of the adrenal gland, or in a female cat, which is usually indifferent to such a method of frightening. Cannon also thought that many of the other adaptations which take place in an animal in this condition could be attributed to the presence of an excess of epinephrine in the blood. The three most important of these are: (1) increased discharge of sugar from the liver into the blood; (2) in- creased efficiency of muscular contraction; (3) diminished clotting time of the blood — all of which are adaptations enabling the animal either to con- quer the source of the fear or to be in a better position to recover from any bodily injury involving a loss of blood should he suffer bodily dam- age. Stewart and Rogoff have more recently thrown considerable doubt on these conclusions by finding that cats in which both adrenal glands are entirely removed from the influence of the nervous system, behave like normal animals when frightened, and develop hyperglycemia when as- phyxiated or etherized. It is scarcely necessary to point out that, until it is definitely established by experimental investigation that epinephrine may be discharged in excessive amounts under certain conditions, it is irrational to assume that such may occur in disease. The surgical removal of tlie adrenal gland is certainly not warranted under any circumstances, The Association of the' Adrenal with Other Endocrine Organs We have at present very little accurate and reliable information on the association of the adrenal with other endocrine organs. That epinephrine has an influence on many diverse organs and glands is an undoubted fact, but this is more probably to be attributed to an activating influence on sympathetic nerve endings than to any specific relationship between the adrenal glands and the gland in question. The most important of the results that have been obtained are the folloAving: 1. With the Thyroid and Parathyroid. — Cannon and Cattell, after con- firming Bradford's discovery that an electric current of action is set up in the salivary gland when it is excited to activtiy, proceeded to investigate the occurrence of such a current in the thyroid gland.73 By placing one nonpolarizable electrode on the gland itself and the other on the neigh- boring subcutaneous tissues or on the trachea, a current was found to be set up by stimulation of the sympathetic nerve supply of the thyroid, by intravenous injection of epinephrine, .or by stimulation of the great THE ADRENAL GLANDS 747 splanchnic nerve before it reaches the adrenal gland. This last result, which is the most important in the present connection, Avas, however, not observed when the blood of the inferior vena cava was prevented by the application of a clamp from getting to the heart, but immediately ap- peared, after stimulation, when the clamp was removed. This experiment taken alone does not, however, justify the conclusion that there is any direct relationship between the adrenal glands and the thyroid, because there are in the thyroid gland structures such as the muscle fibers in the blood vessels, which a hypersecretion of epinephrine might affect. Before any direct relationship between the two glands could be claimed to exist, it would be necessary to show that the thyroid action current is obtained with a concentration of epinephrine in the blood lower than that affecting the blood vessels. 2. With the Sexual Glands. — As mentioned above, a very direct rela- tionship exists between the development of the sexual glands and that of the suprarenals,. particularly the cortex of the glands. In addition to the evidence above furnished, it may be mentioned that in hyperplasia of the adrenals changes occur in the testicles, particularly in their interstitial cells. 3. With the Liver. — Of the many functions of this gland that which is most directly associated Avith epinephrine is the production of glucose from glycogen — the glycogenolytic process (see page 669). The injection of epinephrine causes an immediate discharge of such an excess of glucose into the blood that hyperglycemia and glycosuria immediately folloAV. This result is most striking Avhen the injection is made in glycogen-rich animals. In animals from Avhich all the glycogen of the liver has been removed by starvation, the injection of large amounts of epinephrine causes glycogen to accumulate in the liver cells — a result Avhich it is difficult to interpret. In the light of the fact that stimulation of the great splanchnic nerve causes a demonstrable increase of epinephrine in the blood, a natural con- clusion is that the glycosuria and hyperglycemia Avhich are known to re- sult from stimulation of the splanchnic nerve or of its center in the medulla, must be dependent upon a hypersecretion of epinephrine. Evidence supporting this hypothesis seemed to be furnished by the obser- vation that, after the removal of the adrenal glands, stimulation of the splanchnic or of the so-called " diabetic" center in the fourth ventricle no longer produced glycosuria even in a glycogen-rich animal. But it is difficult to see how such an important physiologic process as that of the nerve control of the production of sugar by the liver should be dependent on the hypersecretion of the adrenal gland, especially since the epineph- rine Avould have to be carried by the blood around a considerable part of 748 'THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS the circulation before it arrived at the place on which it is to act. More- over, it has been shown that "stimulation of the previously cut hepatic nerve plexus (around the hepatic pedicle) in a normal animal produces hyperglycogenolysis, in which case there can be no question of a hyper- secretion of epinephrine. No doubt the adrenal glands have some important relationship to the nerve control of the glycogenolytic' process, for, in animals from which the adrenal glands have been removed, stimulation of the hepatic plexus does not produce hyperglycemia. From this result it would appear that the presence of a certain amount of epinephrine in the blood is necessary for the proper transmission of the nerve impulse from the sympathetic nerve fibers to the liver cell. When the nervous system is stimulated in such a way as to excite the glycogenolytic process, two effects both operat- ing in the same direction with regard to the glycogenic function are developed: the one, a hypersecretion of epinephrine, which activates the sympathetic nerve endings, the other, the transmission of the nerve impulse to the liver cell (Macleod arid-R. G. Pearce).74 4. With the Pancreas. — The function of the pancreas here concerned is that of its supposed internal secretion from the Isles of Langerhans. Since epinephrine readily produces glycosuria, and since excision of the pancreas has the same effect, it has been natural to inquire whether any relationship exists between the two glands, and some observers have obtained results which they interpret as indicating that it does. Certain observers even state that glycosuria does not occur after the injection if at the same time extract of pancreas is injected. It is al- most certain, however, that these results are not trustworthy. Thus, removal of the adrenal glands in an animal suffering from pancreatic diabetes does not restore any of the lost power of utilizing glucose during the feAV hours that the animal remains alive.74 That some rela- tionship may, however, exist is indicated by the fact that epinephrine causes dilatation of the pupil when it is dropped into the eye of a per- son suffering from diabetes, whereas it has no such effect in the normal individual. CHAPTER LXXXIII THE THYROID AND PARATHYROID GLANDS Structural Relationships The thyroid and parathyroid glands are intimately associated, anatom- ically, in most animals. The thyroid is present in all the vertebrates, but the parathyroids do not occur below the amphibia. The thyroid exists as two lateral lobes joined over the trachea by the so-called isthmus. The parathyroids are very much smaller, being four in number and located in pairs on the posterior aspect of the thyroid lobes.- The two upper parathyroids are usually more or less embedded in the thyroid tissue, whereas the lower ones are much more loosely attached to the thyroid; indeed, in some animals they are quite separate from it and may be located at a distance, as in the mediastinum. Accessory thyroid and parathyroid glands are sometimes present in the tissues of the neck, or in the anterior mediastinum, accessory parathyroids being common in the rabbit and rat, and parathyroid tissue being present in the thymus in 5 per cent of dogs (Marine75). Before these anatomic relationships were thoroughly worked out, there was much confusion in the interpre- tation of the results following removal of one or the other gland. In their histologic structure and embryologic derivation, the two glands are very different. The parathyroids are developed as an out- growth from the third and fourth branchial pouches, and they are com- posed of masses of epithelial-like cells, sometimes more or less divided up into lobules or trabeculae by bands of connective tissue. The cells contain granules, some of which are of a fatty nature. Sometimes col- loid-like material is found between the cells, or it may be enclosed in small vesicles not unlike those of the thyroid, although usually consider- ably smaller. -The blood vessels are extremely numerous, and form sinus-like capillaries, which come into close relationship Avith the epi- thelial cells of the glands. Nerves also are abundant and pass both to the vessels and to the secreting cells. The blood vessels are derived from the inferior thyroid artery. The thyroid is developed by immediate outgrowth from the entoderm lining the floor of the pharynx, at a level between the first and second branchial pouches. Represented at first by a solid column of cells, there very soon occurs a division at the lower end into two lateral por- 749 7f)0 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS tions, and the original solid column becomes hollowed out. The two lateral branches of the original column divide again and again so as to form a system of hollow tubes lined with epithelium. These afterward become cut up so as to form the closed vesicles characteristic of the gland. Each vesicle is more or less spheroidal in shape, and has no basement membrane, but its walls are formed by a layer of epithelial cells, which may be columnar, cubical, or flattened in shape. Each vesicle is filled with the so-called colloid material, which is peculiar in con- taining iodine, and between the vesicles is a layer of connective tissue often containing small cells, some of which are not unlike those of the parathyroid. The connective tissue also contains the blood vessels, which are very numerous — indeed, the thyroid, in proportion to its size, receives more than five times as much blood as the kidneys, the only tissue that surpasses it in this regard being the medulla of the adrenal gland (see page 211). The nerves arise from both the vagus and the sympathetic systems and have been traced to the secreting epithelial cells. The above description applies to a strictly normal gland. THE THYROID GLAND Condition of the Gland In the crowded communities of the Great Lakes Basin of this conti- nent, it has been found that in most animals the thyroid gland is more or less abnormal. In Cleveland, for example, Marine has found this to be the case in well over 90 per cent of the dogs brought to the laboratory.77 The condition usually goes under the name of simple goiter, which in- cludes all thyroid enlargements except those of exophthalmic goiter. In man the goiter originates usually about the age of adolescence and more frequently in girls than in boys. It may sometimes pass over into the exophthalmic type. The exact pathologic changes in the goitrous gland vary with the species of animal and with the duration of the dis- ease. In man, besides the cystic or colloid goiter an adenomatous type is very common although rare in other animals. From the numerous observations that have been made on the glands of domestic animals, it has been clearly established that the very earliest sign of goiter is a diminution in the iodine content of the gland; fol- lowed by an increase in the epithelial cells and in the blood supply and a decrease in the colloid. Such kyperplasia may be induced in what re- mains after removal of a large part of a normal gland (compensatory hyperplasia), or if a similar operation be performed early in pregnancy, the young when born will be found to have hyperplastic thyroids. A certain degree of hyperplasia exists as an accompaniment of pregnancy, THE THYROID AND PARATHYROID GLANDS 751 and it can be produced in normal animals (particularly rats) by placing them on an excessive meat diet. Important observations bearing on this point have been made by Marine on brook trout, in which it has been found that the so-called carcinoma that develops when the fish kept in hatcheries are fed with unsuitable food and overcrowded, is really a typical hyperplasia. In its second stage this develops into what is known /^*v c. Fig. 193. — Microphotographs of thyroid gland of dog. A, normal hyperplasia; B, active hyper- plasia; C, colloid goiter. (From Marine and L,enhart.) as colloid goiter which is produced by a deposition of colloid material between the rows of cells so as to cause an opening out again of the vesicles (Fig. 193), with a consequent tendency to a reversion to the normal histologic structure, so far as this is possible. The vesicles in such a gland are of enormous size, and the lining epithelium, low cubical, or almost flat in shape. The outstanding characteristic feature of the colloid material is that 752 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS it contains iodine, which exists in combination with a nonprotein nitrog- enous base, and is usually called iodothyrin. In the gland itself the iodothyrin may be in combination with protein, forming iodothyro- globulin. E. C. Kendall79 has recently succeeded in isolating a pure crystalline substance of perfectly constant composition and containing over 60 per cent of iodine. It is called alpha-iodine, and has been identi- fied'as an indole compound. In extremely minute dosage it greatly affects the energy metabolism, and is said to induce symptoms like exophthalmic goiter. Its therapeutic value in cases of thyroid deficiency is remarkable. Kendall believes this substance to be the active constituent of the thyroid and to be associated with the metabolism of amino acids. For one thing, when it is given alone no change occurs in pulse rate, whereas if amino acids are given along with it, there is acceleration. The importance of the relationship between the function of the thyroid and the iodine-containing material is indicated by the changes which occur in the percentage of iodine in the glands under varying condi- tions of activity. Marine observed that the amount of iodine is inversely proportional to the degree of hyperplasia of the gland, and when the hyperplastic condition becomes fully developed, scarcely a trace of iodine is contained in the gland. Later, when the hyperplasia gives place to colloid goiter, the iodine increases again, both absolutely and relatively. Moreover, it has been found that if iodide be administered to an animal suffering from hyperplasia, the hyperplastic condition very quickly disappears (Fig. 192) and the animal becomes normal. Thus, in brook trout, the poor nutritive condition of the fish when hyperplasia has developed can be immediately remedied by placing them in larger quan- tities of running water or by adding small traces of iodide to the water. The administration of small amounts of iodine as in ordinary salt from salt deposits also prevents goiter in farm stock, this having been first noted in the State of Michigan, where prior to the discovery of salt deposits sheep breeding was an entire failure. The importance of admin- istering small doses of iodides to school children living in goitrous dis- tricts has recently been emphasized by Marine and Kimball.78 As small a dose as 0.001 gm. at weekly intervals prevents goiter in puppies sus- ceptible to it. Experimental Thyroidectomy A correct interpretation of the functional changes and symptoms which follow upon partial or complete removal of the thyroid gland, or from its disease, has proved a very difficult problem, partly because sufficient care has not been taken to note how much parathyroid tissue was re- moved along with the thyroid, and partly because the fact has been over- THE THYROID AND PARATHYROID GLANDS 753 looked that the effects produced by thyroidectomy and parathyroid- ectomy are often very different in animals of the same kind at dif- ferent ages. Speaking generally, it may be said that the influence of the parathyroid is focused mainly on the nerve centers and only to a second- ary degree on the metabolic functions, whereas the reverse is the case with the thyroid, its main effect being on metabolism, although it prob- ably also exercises a secondary effect on the nerve centers. More so than in the case of any other endocrine organ, our knowledge concerning the function of the thyroid has been gained by clinical experience, and it is difficult to say whether the cJinical or the experimental method has contributed the greater amount of information. The results of experimental extirpation of the thyroid vary accord- ing to the age of the animal, and frequently they are by no means marked, provided sufficient parathyroid tissue has been undamaged. The symptoms are in general thickening and drying of the skin, with a tendency to adiposity and a loss of tone of the muscle. The body tem- perature is low and the sexual functions become subnormal. Nervous symptoms in the direction of mental dullness and lethargy are also usually present. Surgical removal of the thyroid in man produces the condition known as CQchexia strumipriva. The symptoms may first of all become apparent a few days after the operation, or they may remain latent for years, and then develop so as' to produce the condition known as myxedema. When nervous symptoms are prominent in cachexia strumipriva, it is usually taken as evidence that an excessive amount of parathyroid tissue has been destroyed. Kocher states that after com- plete loss of the thyroid, life is impossible for more than seven years, and that to prevent ultimate ill effects, at least one-fourth of the organ should be left intact. Disease of the Thyroid The symptoms of diseased conditions of the thyroid may be inter- preted as the consequence of increased or diminished functioning of the gland. Sometimes, however, the less active gland is really increased in bulk, this increase being caused by the accumulation in it of very large quantities of colloid material accompanied by an attenuated condition of the vesicular cells (see page 751). When the gland is atrophied at birth, the condition of cretinism soon becomes developed (Fig. 194). The characteristic features of cretinism are: (1) An arrest of growth, espe- cially of the skeleton, accompanied by incomplete ossification of the long bones and failure of the fontanelles of the skull to close properly. (2) Poor development of the muscular system. (3) An unhealthy, dry, swollen condition of the skin, so that it is yellowish in color, the face being pale 754 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS and puffy. (4) An abnormal development of the connective tissues causing a shapeless condition of the surface; the abdomen is always swollen, the hands and feet are shapeless, and the nose depressed. (5) The nervous system also fails to develop properly, so that at the age of puberty or over, the child remains like an infant in his mental behavior, idiotism being common. Indeed, the whole clinical picture is so char- acteristic that once having seen a case no one can fail afterward to Fig. 194. — Cretin, nineteen years old. The treatment with thyroid extract started too late to be ot benefit. (Patient of Dr. S. J. Webster.) recognize the disease. Besides being due to congenital absence of the thyroid (sporadic type), cretinism may also occur as a result of goitrous degeneration of the gland. This forms the so-called endemic variety of the disease, and is more commonly seen in goitrous districts, being not infrequently associated with disease of the parathyroid, in which case the nervous symptoms are very prominent. Atrophy of the thyroid in adults causes the clinical condition known THE THYROID AND PARATHYROID GLANDS 755 as myxedema, and here again the symptoms are very characteristic (Fig. 195). The skin is dry and thick, with a deposition of connective tissue often containing fat in its deeper layers; the hands and feet become unshapely; the lips thick and the tongue somewhat enlarged, so that when the person attempts to speak, it appears as if the tongue were too large for the mouth ; the hair falls out ; there is a low body temperature, and it can be shown that the energy metabolism is greatly depressed, and that a deficiency of oxygen is being consumed. It is said the person can take a larger quantity of sugar than an ordinary individual without the development of glycosuria, but the depression of the metabolic function causes the patient to take sparingly of food, in spite of which, however, the body weight may steadily increase. The sexual function becomes A. B. Fig. 195. — A, Case of myxedema; B, Same after seven months' treatment. (From Tigerstedt.) depressed, and there is involvement of the nervous system as shown by mental dullness and lethargy. Although the thyroid gland is much atrophied in myxedema, symptoms that are very similar may also occur when the gland is enormously en- larged. As already explained, however, this enlargement is due merely to an accumulation of colloidal material and is really an atrophic con- dition. A patient suffering from endemic goiter may at first exhibit symptoms which are usually attributed to a hypersecretion of thyroid material into the blood (the symptoms will be described immediately), but later these give place to symptoms not unlike those of myxedema. It is concluded that the above conditions are due to deficiency of thyroid function, or hypothyroidism, because: (1) the gland is atrophied, 756 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS and (2) similar symptoms to those exhibited by the clinical conditions can be produced experimentally by the removal of the gland in animals. By observations on the effect of administration of thyroid extract to cretinous or myxedematous patients, prompt amelioration of the symp- toms occurs, which certainly suggests that the real cause is the absence of an internal secretion. There is probably nothing more striking in the whole domain of therapeutics than this effect from the administration of thyroid extract or, more so still, of alpha-iodine. If the treatment is started early enough, the cretinous child from being an ill-developed idiot quickly 'catches up with children of his own age and becomes in every respect normal. Even if this treatment is not undertaken until the child is several years of age, it is remarkable how quickly the benefit may show itself. In myxedema and cachexia strumipriva also, the symptoms very quickly disappear and the person becomes perfectly nor- mal by the treatment. In all these conditions, however, the thyroid extract must be administered continuously in order to prevent the reap- pearance of symptoms. Quite distinct from the above described conditions of hypothyroidism are those produced by an excess of thyroid autacoid in the blood, namely, hyperthyroidism. Such a condition can be produced experimentally in normal animals by the administration of thyroid extract or alpha-iodine (Kendall). In man large doses are soon followed by great quickening of the pulse with some irregularity, flushing of the skin, increased per- spiration, tremor in the limbs, emaciation, and marked nervous excita- bility. Along with these symptoms, metabolic investigations have shown that the energy output per square meter of surface is greatly increased, being sometimes nearly doubled; that the nitrogen excretion is exces- sive; and that alimentary glycosuria is very commonly present. The body temperature is not, however, as a rule increased, because although metabolism is excited, yet heat loss is correspondingly increased. Ex- ophthalmos is said to develop very occasionally after such administra- tion, but this is doubtful. Lastly, there are usually digestive disturb- ances, although the appetite is likely to be increased. The pulse is quick- ened after administration of alpha-iodine only when protein food is also taken. This is believed by Kendall to be due to the association between the thyroid hormone and the metabolism of the ammo acids. The symptoms following the injection of the extract are very similar to those of the disease known as exophthalmic goiter. Indeed, the symp- toms are so much alike in the tAvo conditions that it is scarcely neces- sary to describe them specially for the disease except to mention that the exophthalmos is much more likely to be present. Like simple goiter this variety is from three to four times more fre- THE THYROID AND PARATHYROID GLANDS 757 quent in women than in men, a fact of significance when we recall the evidence of association between the thyroid gland and the generative organs. It is said that the disease is usually coupled with persistence of the thymus gland. The thyroid gland in exophthalmic goiter is enlarged, sometimes in one lobe; it is hard and pulpy, and on auscultation a mur- mur is heard. Histologically the gland presents a picture very like that which has been described above as hyperplasia ; that is to say, the vesicles have a deficiency of colloid material ; their epithelium is colum- nar and folded up into the vesicles; and the interstitial tissue between the vesicles is very markedly increased. Exophthalmic goiter is almost universally claimed to be due to hyper- secretion of the thyroid, because: (1) the symptoms of the disease are not unlike those produced by excessive administration of thyroid to a normal individual; and (2) they are in general opposite in character to the symp- toms found in cases where the thyroid gland is atrophied. The blood of a person with exophthalmic goiter when injected into mice increases their resistance to the toxic action of acetonitrile, which is also the case after thyroid extract has been injected. In many cases of exophthalmic goiter partial removal of the gland is said to ameliorate the symptoms. Other clinicians, however, state that if the patient is given proper medical treatment, rest, and diet, equally beneficial results can be obtained. Certain investigators, however, deny that it has yet been conclusively demonstrated that exophthalmic goiter is due to hypersecretion of the thy- roid (Marine). It is pointed out that, if hypersecretion were the cause of the disease, one would expect that the injection into animals of the blood of patients suffering from it would produce symptoms similar to those following the injection of thyroid extract. The results of such experi- ments, however, have been extremely confusing and very indecisive, since it is difficult to recognize in laboratory animals many of the characteristic symptoms, especially those affecting the skin and eyes and the general bodily nutrition. Another difficulty in accepting the hypersecretion hypoth- esis is the fact that an extract of a gland removed from an exophthalmic patient has no different physiologic action on a normal animal from an extract of a normal gland containing the same percentage of iodine. The evidence is by no means conclusive one way or the other, and it may well be that the observed changes in the thyroid gland are not the cause of the symptoms of exophthalmic goiter, but merely, like the other symp- toms of this disease, a result of some condition located elsewhere. The Relationship of the Thyroid with OtheT Endocrine Organs 1. With the Generative Organs. — Evidence of an association between the female generative organs and the thyroid is very strong; thus, the 758 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS thyroid becomes enlarged at puberty, during the menses, and during pregnancy, and in thyroidectomized young animals the sexual glands fail to develop properly. 2. With the Adrenal Glands. — (See page 746.) 3. With the Pituitary Body. — After removal of the thyroid, the pitu- itary becomes greatly altered and enlarged, particularly the pars an- terior, in which it is not uncommon to find that a certain amount of vesicles containing colloid, not unlike those of the thyroid, become devel- oped. This colloid material, however, does not contain iodine. It is said that this increase of the pituitary after thyroidectomy does not occur if thyroid extract be administered. Increased activity of the pars inter- media of the pituitary is also quite plain. These facts Avould at first sight seem to indicate that the pituitary and the thyroid can act vica- riously, but this is very doubtful, for it has not been found that pitu- itary extract has any beneficial effect in the treatment of goiter and myx- edema. Nevertheless the association in function of the two glands must be more or less close, not alone for the above reasons, but also because they are both associated to much the same degree with the sexual organs, and both act on the higher functions of the nervous system in much the same manner. 4. With the Thymus Gland. — The persistence of the thymus in ex- ophthalmic goiter, as well as the anatomic and embryologic relationship between thymus and thyroid, is taken to indicate some close relationship. THE PARATHYROIDS Experimental Parathyroidectomy Experimental parathyroidectomy yields results which vary in dif- ferent groups of animals, undoubtedly because of the fact that in some, such as the rat and rabbit, accessory parathyroids may exist. In . gen- eral, however, it has been found that if more than two of the four parathyroids be removed, very definite and pronounced nervous symp- toms soon supervene and if all four glands be removed, a quickly fatal result is inevitable. The most acute symptoms are exhibited by the carnivora. They may not be apparent for a day or two after the opera- tion, although during the period the animal is in a depressed state, re- fusing food and losing weight rapidly. The muscles are also more or less stiff during this stage. When more definite symptoms appear, they con- sist of a marked abnormality of muscular contraction, leading to the occurrence of fibrillar contractions, or tremors and, later, to cramp-like and clonic contractions. When spontaneous movements are made, a THE THYROID AND PARATHYROID GLANDS 759 peculiar shaking of the foot, like that made by a normal animal to shake water off its pads, is a characteristic symptom. The slightest stimulation of the peripheral nerves is sufficient to induce one of these attacks, which recur with ever increasing frequency, becoming at the same time more pronounced and accompanied by other disturbances, such as diarrhea, profuse salivation, rapid pulse, and dyspnea (in the dog but not in the cat). In cases that are not quickly fatal, the hair tends to be shed, and the teeth to be improperly calcified (in young animals). Where a certain amount of parathyroid tissue has been left — for example, one of the four lobes — the symptoms may not appear except under conditions of special strain to the animal economy, such as pregnancy or improper diet. Thus, in a bitch from which three of the four glands had been removed, no symptoms of tetany occurred until she became pregnant. Under the same conditions it has been found that a diet of flesh is much more apt to bring about the condition than one of vegetables or milk. Tetany, as the above condition is called, may also become developed in man either as the result of surgical removal of the parathyroids or because of their improper development. The symptoms in man are very similar to those observed in laboratory animals, the only difference being that the muscular contractions are more likely to be tonic in. character. Certain symptoms that may develop during pregnancy or in the course of infectious diseases or in newborn infants have also been found to be associated with degeneration of or hemorrhage into the parathyroid (idiopathic tetany), and certain obscure nervous diseases in adults, such as paralysis agitans, may possibly also be associated with changes in this gland. Chorea, epilepsy, and eclampsia have likewise been thought to be associated with it. The parathyroid gland, besides influencing the nerve centers, has also an influence on metabolism. The symptoms produced are: (1) rapid emaciation and failure to grow; (2) a tendency to the production of glycosuria, often detected by finding that the assimilation limit for carbohydrate is lowered (page 652) ; and (3) most definitely of all, an interference with calcium metabolism, as illustrated by the failure of the teeth and bones to calcify properly. This interference with normal metabolism led Kellogg and Voegtlin81 to study the effect produced on parathyroidectomized animals by the administration of calcium. It was found that the symptoms were considerably ameliorated. These authors concluded from their results that the essential cause of tetany is a deficiency of calcium in the blood. It is possible however that the bene- ficial action of calcium salts in this condition is that it decreases the excitability of the nervous system, an action which it is known to possess. 760 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS When the tetany is the result of a complete extirpation of all parathy- roid tissue, the symptoms can be combated by a successful transplan- tation or graft of parathyroid tissue made from an animal of the same species. Indeed, it has been found that the success of a graft of parathy- roid is assured only when the graft is derived from the same kind of animal as that from which the parathyroid has been removed. Implan- tation into the subcutaneous tissue of a tetany patient of parathyroid tissue obtained fresh from the deadhouse has been performed with bene- ficial outcome. Noel Paton, Findlay and Watson80 have recently contributed greatly to our knowledge of the physiologic pathology of tetania thyreopriva, as the above condition is called. The symptoms are not due to any con- dition affecting the muscles themselves, since they disappear after sec- tion of the nerves. Nor are they primarily dependent upon the cere- brum or cerebellum, since ablation of neither abolishes them. This does not imply that secondary involvement of the higher centers never oc- curs; on the contrary, the epileptiform convulsions and disturbances of equilibrium sometimes observed indicate cerebral or cerebellar involve- ment, respectively. This leaves some part of the lower neuron reflex arcs as the site of involvement. It is not the afferent neuron, since the tremors and jerkings persist after section of the posterior roots, leaving the afferent neuron as the affected structure. The foregoing conclusion led Paton and his co-workers to compare the response of muscle and nerve to electric stimulation in normal and parathyroidectomized animals. Although there are considerable varia- tions in the responses of a normal animal, they are very definitely ex- aggerated in tetany when either the motor neuron or the muscle itself is stimulated, the exaggeration in the latter case being dependent upon alterations in the neural structures (nerve endings) in the muscle. The increased electric excitability can not, however, be taken as a measure of the severity of the condition, for it may be no more marked in cases in which there is involvement of the cerebral hemisphere (causing epilep- tiform fits) than in milder cases. As to the cause of the symptoms, many possibilities have to be con- sidered. In the first place, no direct relationship exists between the thyroid and parathyroid in this connection. One cause might be the absence of some substance which checks the activity of the nervous sys- tem, some chalone in Schafer's sense. That such is not the case is shown among other things by the fact that bleeding and transfusing normal saline immediately removes the symptoms for some time. Moreover, the metabolic disturbances go on when the nervous symptoms are slight. It had previously been thought by W. G. Macallum81 that, since symp- THE THYROID AND PARATHYROID GLANDS 761 toms like those of tetany can be induced by deficiency of calcium in the body and the symptoms of parathyroidectomy relieved by administration of this cation, calcium deficiency is the cause of the symptoms. While not denying that these ions may have some relationship to the symptoms, Noel Paton ascribes tliem to intoxication by yiianidine (page 605). The evidence is as follows: (1) Guanidine and methyl guanidine admin- istered to normal animals produce symptoms that are identical with those following parathyroidectomy. (2) There is a marked increase in the amount of these substances in the blood and urine of parathyroidec- tomized dogs and in the urine of children suffering from idiopathic tetany. (3) In certain cases the serum of parathyroidectomized dogs acts upon the muscles of the frog similarly to weak solutions of guani- dine and methyl guanidine. (4) There is a striking similarity in the relative amounts of the nitrogenous metabolites in the urine of parathy- roidectomized dogs and of normal animals injected with guanidine. It is concluded that the parathyroids control the metabolism of guani- dine "by preventing its development in undue amounts. In this way they probably exercise a regulative action upon the tone of the skeletal muscles." Since it is similar with regard to its characters and metabo- lism to the condition following thyroidectomy, it is believed that disease of the parathyroids is the cause of idiopathic tetany. The Relationship of the Parathyroid with Other Endocrine Organs We know very little of the relationship of the parathyroid with other endocrine organs. Vincent and others have stated that after removal of the thyroid itself enlargement of the parathyroid may occur with the formation of colloid material between the rows of cells, but the con- clusion that this represents a vicarious function between the thyroid and parathyroid glands is not generally accepted. The supposed relation- ships among the parathyroid and the pituitary and adrenal glands are also based upon uncertain evidence. CHAPTER LXXXIV THE PITUITARY BODY Structural Relationships Situated at the base of the brain and lying in the sella turcica, the pituitary body in man does not weigh much more than half a gram. It is connected with the brain by a funnel-shaped stalk, the infundibulum. On account of a natural cleft, which runs across the gland in an oblique plane, it is an easy matter to split it into two portions, an anterior, or pars glandularis, and a posterior, or pars nervosa. This cleft in the case of man is usually found to be more or less broken up into isolated cysts containing a colloid-like material, and it represents the remains of the original tubular structure from which the pars glandularis is de- veloped; namely, a pouch growing out from the buccal ectoderm. On histologic examination it will be found that the pars glandularis consists of masses of epithelial cells with large sinus-like blood capil- laries lying between them. These blood vessels are very numerous, so that in an injected gland this portion of the pituitary stands out very prominently. The vessels are derived from about twenty small arterioles that converge toward the pituitary from the circle of Willis, and enter the gland by the infundibulum or stalk by which the gland is connected with the base of the brain. Three types of cell can be differentiated: nonstaining (chromaphobe) and granular (chromaphil), of which latter there are cells with acid-staining and others with base-staining granules, the former being by far the more numerous (Schafer).60 In some animals such as the cat, the cells of the pars anterior are arranged around the blood sinuses in rows as in a columnar epithelium. The cells with acid-staining granules are said to become much increased in number in pregnancy and also in the enlarged gland of acromegaly (see page 772). After thyroidectomy it has been observed that colloid-like masses ac- cumulate in the pars glandularis, the cells sometimes arranging them- selves around these masses as in the thyroid gland. The colloid, how- ever, contains no iodine. The posterior part of the gland, or pars nervosa, is composed almost entirely of neuroglia, cells, and fibers, usually with some hyaline or granular material lying between them, particularly in the neighborhood 762 THE PITUITARY BODY 763 of the infundibulum, into which it may be traced. It is believed that the active principle of the gland is represented by this material. The blood supply of the pars nervosa is relatively scanty. Between the pars nervosa and the intraglandular cleft above referred to is a layer of cells differing from those of either the anterior or the posterior lobe. This layer of cells constitutes the so-called pars inter- media. The cells are somewhat like those of the pars glandularis, except that they are distinctly granular, the granules being of the neutrophile variety, that is to say, they stain with neither basic nor acid dyes. Well- defined vesicles containing an oxyphile colloid material are often found Fig. 196. — Drawing from a photograph of a mesial sagittal section through the pituitary gland of a human fetus (5th month): a, optic chiasma; c, third ventricle; d, pars glandularis; e, infun- dibulum surrounded by epithelial cells; /, pars intermedia; g, intraglandular cleft; h, pars nervosa. (Herring, from Howell's Physiology.') between them. The blood supply is much less abundant than that of the pars glandularis. Although well separated by the cleft from the pars glandularis, the pars intermedia is not well separated from the pars nervosa, because many of its cells extend for some distance into the lat- ter between the neuroglial fibers. Certain of the cells in the pars inter- media may be seen in various stages of conversion into globular hyaline bodies, or a granular mass of material may appear in them. In either case, the cells ultimately break down, setting free the hyaline or granular material, which is believed to be the origin of the similar material al- ready described as existing between the neuroglial fibers of the pars nervosa and therefore ultimately finding its way by the infundibulum 764 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS into the third ventricle of the brain. These hyaline globules are greatly increased after thyroidectomy. It should be mentioned, finally, that at the margin of the intraglandular cleft the intermediary and anterior portions of the pituitary come together, although the cells of each can readily be distinguished on account of their staining properties. This pars glandularis et intermedia also extends as a thin layer over part of the pars nervosa and around the neck of the gland at the infundibulum. These relationships are Avell shown in the accompanying diagram (Fig. 196). Functions Concerning the function of the pituitary, it may be said in general that the anterior lobe has an important relationship to the nutritive con- dition of the body during growth, especially of the skeletal structures, and that the posterior lobe produces a very active autacoid having to do with the physiologic activity of unstriped muscle fiber. The pars inter- media seems to be associated with the posterior lobe in the production of this autacoid. The function of these two parts will therefore be con- sidered together. Function of the Pars Glandularis. — The facts concerning the function of the pars glandularis have been gleaned largely by observing the ef- fects produced by partial or complete removal of the entire pituitary, justification for ascribing to the removal of the anterior, rather than the posterior, lobe the results that are obtained being furnished by control experiments, in which by removal of the posterior lobe alone similar effects are not observed. Complete removal of the pituitary is almost invariably fatal, the con- dition being called apituitarism. Two operative procedures have been employed for the removal of the gland. One of these, elaborated by Gushing and his pupils,82 consists in trephining the skull and elevating the temporal lobe of the cerebrum so as to expose the gland. The other, elaborated by Horsley,83 consists in approaching the gland through the orbital cavity. Although there is some danger of injury to nervous tissues by the intracranial method, its results are more dependable since the gland is actually exposed to view before being removed. Most hypophysectomized animals die within two or three days, unless they are very young. This longer survival of young animals is ascribed to the presence of accessory pituitary material situated in the dura mater lining the sella turcica. The most extensive observations have been made on dogs. On the day following the operation the animal appears about normal, but it gradually becomes less active, refusing food and respond- ing slowly to stimulation. It gradually gets weaker and weaker; muscu- THE PITUITARY BODY 765 lar tremors may appear, the respiration and pulse become slow, the back arched, the temperature subnormal; and, usually within about forty- eight hours, coma develops and the animal dies in this condition. When the symptoms are less acute and death does not occur so early, it is believed by Gushing either that small portions of the gland have been left behind or that some vicarious activity of other organs has developed to replace that of the pituitary. When only a part of the pituitary is removed either unintentionally or intentionally, the symptoms are not nearly so acute, and the condition is known as hypopituitarism. It is by a study of this condition that most facts concerning the function of the anterior lobe have been learned. When the operation is performed on young animals, they fail to grow properly; the milk teeth and the lanugo are retained; the epiphyses do not ankylose; the thyroid and thymus glands are enlarged; and the cortex of the suprarenal and the sexual organs fails to develop. The animal, though small, becomes very fat and may therefore increase in weight. There is distinct evidence of mental dullness. From these results it is concluded that the anterior lobe of the pituitary produces autacoids having to do ivith the development of the skeletal and other structures of the growing animal. That this autacoid is not derived from the posterior lobe is evidenced by the fact that partial injury of this lobe, or indeed its entire removal, is not followed by similar symptoms. Closer examination of the metabolic function in hypophysectomized animals has shown that there is a marked depression in the respiratory exchange of oxygen and carbon dioxide, and that the ability to metabo- lize carbohydrate becomes heightened; that is to say, the animal with- out developing glycosuria can tolerate a larger quantity of sugar than the normal animal. This effect on carbohydrate metabolism may how- ever be associated not so much with the function of the anterior lobe as with that of the posterior, for, as we shall see later, Gushing and his pupils have found that extract of the posterior lobe has a marked effect on the assimilation limit of carbohydrate. Attempts have been made to graft the pituitary, especially the anterior lobe, into various parts of the body. It has been found, however, that within a few days the grafts atrophy and disappear unless there has been complete removal of the pituitary itself, in which case the graft may remain for a month or so and the otherwise fatal outcome of hypophy- sectomy be warded off. Sometimes, where the graft has remained for a longer time, it is said that a temporary increase in the growth of the .animal has been noticed. Other observers have investigated the effects in normal animals of continuous oral administration of pituitary substance or of subcutaneous 766 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS injection of extract. The earlier results were indefinite and confusing, but recently Brailsford Robertson81 has succeeded in isolating from the anterior lobe a substance called tethelin, which accelerates growth in young animals and is thought to have a possible value in hastening the healing process in wounds. Tethelin is precipitated by dry ether from an alcoholic extract of the carefully isolated anterior lobes. It contains 1.4 per cent of phosphorus and nitrogen in the proportion of four atoms for every atom of phos- phorus, two of the nitrogen atoms being present as amino groups and one in an imino group. The effects on growth of mice are in every par- ticular like those of the administration of anterior lobes, and consist in retardation of the first portion of the third growth cycle,* followed by acceleration of the latter portion of this cycle. When fully grown, tethelin-fed mice also differ from normal animals in being smaller in size but of greater weight, with a distinct difference in the condition of the coat. Normal animals at fourteen months of age have "shaggy, staring and discolored coats," whereas in tethelin-fed animals they have the glossy and silky appearance of young animals. During growth, nor- mal animals display a greater variability in weight than tethelin-fed animals. Extraordinary effects have been observed by Clark85 to be produced by feeding laying hens with pituitary gland. Thus, by giving to one- year-old hens, in addition to their usual food, 20 milligrams of fresh pituitary substance for four days, it was found that the average daily number of eggs laid by a batch of 655 hens was raised from 273 during the four days preceding the pituitary feeding to 352 during the four days of the administration, these results being obtained at a time of year when the natural egg-production of the hens was diminishing. It was further observed that not only is the output of eggs greatly increased as a result of the pituitary feeding, but likewise their fertility, for in another experiment in which 35 hens were kept along with two cockerels of the same breed, not only was the output of eggs increased (from 18 up to 33), but the fertility of the eggs was greatly enhanced. Functions of the Posterior Lobe or Pars Nervosa. — As already men- tioned, excision of this part of the pituitary can be tolerably well with- stood by the animal, so much so indeed that from its behavior after the operation we can conclude little as to the function of the lobe. On the other hand, extracts of the posterior lobe injected into normal animals produce effects that are very striking, indicating that the main function *Robertson has contributed valuable and very extensive data on the normal curve of growth of white mice kept under carefully controlled conditions. Three growth cycles are present: the first attains its maximum velocity between seven and fourteen days after birth; the second, between twenty -one and twenty-eight days; and! the third about six weeks, after which the velocity decreases progressively, until further growth ceases between the fiftieth and sixtieth weeks succeeding birth. THE PITUITARY BODY 767 of this lobe is production of an autocoid. The extracts have more or less an epinephrine-like action. Such extracts, rendered protein-free and steril- ized, are obtainable on the market under the various names of pituitrin, hypophysin, etc. From them a crystallizable material has been obtained, but this is probably a mixture of various substances. In discussing the functions of these various extracts, it must be remembered that the inter- mediary part (pars intermedia) is included with the posterior lobe in their preparation. Although the effect of pituitary extract on plain muscle filer (and on glandular tissue) appears, on first sight, to be very like that produced by epinephrine, it has been found on closer examination that the two substances really act in different ways. The rise in blood pressure pro- duced by pituitary autacoid is likely to be more prolonged than that produced by epinephrine. It stimulates increased cardiac activity, but after the vagi have been cut or sufficient atropine administered to para- lyze them, the pituitary autacoid continues to stimulate the strength of the heartbeat without producing the acceleration noted with epinephrine. Whereas epinephrine has little or no action on the coronary vessels or on those of the lungs, pituitary autacoid usually produces constriction of both types of vessel; and on the renal arteries the actions of the two autacoids are entirely different, for epinephrine has a marked constric- ing effect, while the pituitary autacoid produces dilatation. Another striking difference in the extracts from the two glands is re- vealed by repeating the injection after the effect of a previous one has completely passed off. With epinephrine the original effect is repro- duced ; with pituitrin, on the other hand, the effect of the second injec- tion is very often the reverse of that of the first.; that is to say, the blood pressure, instead of rising, may fall, or the rise be very much less marked. Whether this effect of the second dose is caused by the action of an autacoid having a chalonic rather than a hormonic influence, or whether it is due to a reversed effect of the same hormone, it is impos- sible at present to say. The chalonic effect in any case is much more evanescent than the hormonic, and it is not caused by cholin, as some have suggested. The effect of epinephrine, it will be remembered, is abolished by ergotoxin and apocodeine. These drugs, on the other hand, have no influence on the action of pituitrin. The difference in action between the two autacoids is usually explained by assuming that the epinephrine acts on the receptor substance associated in some way with terminations of the sympathetic nerve fibers in involuntary muscle, whereas pituitrin acts directly on the involuntary muscle fibers themselves. Other types of involuntary fiber are also acted on by pituitrin. The uterine contractions for example are stimulated (Fig. 197) ; so are those of 768 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS the intestine (in contrast to the inhibiting effect of epinephrme), and of the bladder-ureter musculature. Dilatation of the pupil of the excised frog Fig. 197. — Tracing showing the action of pituitrin on the uterine contractions and blood pressure in a dog. Made by Harbour's method. (From Jackson.) eye is produced. The effect of pituitrin on the muscle of the bronchioles is shown in Fig. 198. The glands on which the pituitrin has the most pronounced action are THE PITUITARY BODY 769 •the mammary glands and the kidneys. The effect on the kidney is evi- denced by the remarkable increase in the urinary flow following injection of the pituitrin. This diuresis might of course be due merely to the vasodilatation that we have seen such extracts produce — a vasodilatation which is all the more marked because the vessels elsewhere in the body undergo constriction. But pituitrin continues to cause increased urinary outflow in the absence of any demonstrable vascular change; it also acts after the administration of atropine, so that it is considered by most observers to act on the excretory epithelium of the convoluted tubules Fig. 198. — Tracing showing the constricting action of pituitrin on the bronchioles and its effect on blood pressure in a spinal dog. (From Jackson.) in much the same way as certain diuretics, like diuretin. This renal hormonic action of pituitrin would appear to be analogous with that of secretin on the epithelium of the pancreas. Another reason for believ- ing that the secretory hormone is independent of that producing vaso- dilatation of the renal vessels is the fact that a repeated dose of pituitrin, although, as we have seen, it usually has a. depressor action on the blood vessels, still produces a stimulating effect on the excretion of urine. The value of pituitrin as a diuretic in clinical practice is now well recognized. The effect on milk secretion is best demonstrated by placing a cannnla 770 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS in the mammary duets so that the milk may freely flow out. By observ-' ing the rate of outflow during the injection of pituitrin, it will be found that a remarkable increase occurs. After this increased secretion has ceased, however, the injection of more pituitrin has no further effect, indicating that the influence of the first injection must have been, not so much to stimulate the secretion of milk, as to accelerate the outflow of that which previously had been secreted and had collected in the alveoli and ducts. This effect explains why the pituitary galactagogue should have very little if any effect on the total production of milk or on the total amount of fat and other constituents contained in it. Histologic examination of sections of a resting mammary gland and of the same gland after administration of the pituitrin, bears out the above interpre- tation of the action. Alveoli in the resting state will be found largely distended with milk and the epithelium flattened against the basal mem- brane, whereas alveoli from the gland after pituitary activity show small shriveled-up alveoli, containing little milk, and with epithelium that is well marked and stands out prominently from the basal membrane. These facts taken together indicate that pituitrin stimulates the mus- cular fibers of the ducts of the mammary glands, thus squeezing out the milk contained in them. Muscular fibers have been described as existing between the basal membrane and epithelial cells, much in the same way as they do in the case of the sweat glands. At least Schafer has suc- ceeded in demonstrating in this position rod-shaped nuclei which prob- ably belong to muscular fibers.60 By their contraction, the milk in the alveoli is expelled into the ducts. It has also been found that pituitrin stimulates the secretion of cerebrospinal fluid, and that this stimulation is independent of a rise in blood pressure. Pituitrin has a distinct effect on carbohydrate metabolism,. After its intravenous or subcutaneous injection, a marked lowering in the toler- ance for sugar is observed (page 652), usually to such an extent that glycosuria becomes established. Gushing and his pupils have concluded that the posterior lobe contributes an autacoid which stimulates the utili- zation of sugar in the body. Confirmatory evidence for this view is fur- nished by the observation that mechanical stimulation of the posterior lobe, such as is produced by puncturing it with a needle, is followed by a temporary glycosuria, which is said to be as pronounced as that fol- lowing puncture of the diabetic center (page 672), provided glycogen is present in the liver. The production of this carbohydrate autacoid would appear to be under the control of the sympathetic nervous system, for it has been found by Gushing and others that stimulation of the superior cervical ganglion, which has been known for many years to be fre- quently followed by glycosuria, has this effect only provided the posterior THE PITUITARY BODY 771 lobe of the pituitary is intact. Even surgical manipulation of the pitui- tary may excite a hypersecretion of pituitrin, which would account for the glycosuria often observed after experimental excision or partial destruction of the pituitary. A similar irritation may be set up in disease of the gland. The glycosuria which is usually observed after partial hypophysectomy soon passes off, to be followed by a permanent condition of increased tolerance for sugar, because now less pituitrin is being produced. It is said that during the stage of increased tolerance diabetes can not be pro- duced even by excision of the pancreas. The glycosuria produced by irritation of the posterior lobe is accompanied by a marked polyuria (dia- betes insipidus), which may outlast the glycosuria. A. B. Fig. 199. — A, To show the appearance before the onset of acromegalic symptoms; B, The ap- pearance after seventeen years of the disease. (After Campbell Geddes.) Clinical Characteristics Because of their importance from a physiologic standpoint, we shall now proceed to review briefly some of the more important facts that have so far been brought to light by clinical observations. The pathologic condition most frequently observed affecting the pituitary is an adenom- atous growth particularly located in the anterior lobe. Besides pro- ducing general symptoms of pressure, such as diminution of the visual field and, perhaps, headache, a shadow can usually be observed when the patient is examined by means of the x-rays. General symptoms, com- monly ascribed to a hypersecretion of the autacoid of the anterior lobe of the pituitary — hyperpituitarism — begin sooner or later to show them- 772 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS selves. These symptoms are almost exactly opposite in character to those observed in animals after removal of this portion of the gland. Thus, the bones of the extremities and of the face become stimulated to in- creased growth, so that if the patient is young, and the epiphyses there- fore not ossified, remarkable elongation of the long bones occurs, pro- ducing the condition known as gigantism. On the other hand, if the dis- ease does not develop until after ossification is complete, its, effects be- come most marked in the bones of the face, the lower jaw becoming Fig. 200. — Hand of a person affected with acrornegaly. enormously hypertrophied and the supraorbital ridges very prominent. The long bones also become enlarged at their extremities, and there may be some increase in length of the vertebral column, although the stature does not increase because of kyphosis (curvature of the spine). The condition is called acromegaly. Nutritive disturbances of the skin and hairs also become marked, causing the skin to become dry and yellowish, and the hairs to undergo abnormal increase over the body. An early symptom of the condition is a failure of the sexual power (Figs. 199 and 200.) THE PITUITARY BODY 773 After a time the disease begins to affect the pars intermedia et nervosa, and disturbances in carbohydrate metabolism come to be observed, con- sisting usually in a diminished tolerance accompanied by glycosuria, in the early stages of the disease, followed by increased tolerance in the later stages. The glycosuria is usually accompanied by marked polyuria. It should be observed that sometimes tumor of the pituitary has been found to exist postmortem though none of the above symptoms had been recorded during life. In these cases it is probable that the disease from the start had been of such a nature as to produce a tendency to hypo- pituitarism rather than hyperpituitarism, for the symptoms are very like those observed in animals after partial or complete removal of the gland. If the condition commences before adolescence, the body fails to grow, although the child may continue to increase in weight because of the remarkable deposition of fat in the tissues. Sexual development is strik- ingly interfered with, and the secondary sexual characteristics fail to show themselves. In boys, for example, the pubic hairs fail to extend up to the umbilicus; and the hairs on the chin do not develop, whereas the hair of the scalp grows profusely. The bones remain of the female type, and a broad pelvis, rounded limbs, small feet and hands are often ob- served. In these cases there is usually excessive tolerance for carbohy- drates, Avhich may explain the adiposity, sugar being converted into fat. In the light of the experimental results, the effect on carbohydrate metabolism may be explained as due to involvement of the posterior lobe. Mental development is retarded, and psychic derangements are sometimes observed. Where the hypopituitarism does not develop until after adolescence, some of the above symptoms will of course be missed, but many will be observed, such as dryness of the skin, loss of hair, and the tendency in the male to adopt certain of the female characteristics, particularly with regard to the growth of hair. Obesity and increased tolerance for sugar are also evident, and pigmentation of the skin, something like that of Addison's disease, is said often to be a prominent feature. Operative interference in the early stages in many of these cases is of undoubted benefit, as is shown by the brilliant work of Harvey Gushing, to which the reader is referred for further information. The Relationship of the Pituitary Gland with Other Endocrine Organs The relationship of the pituitary gland with other endocrine organs seems to be an intimate one. 1. With the Thyroid and Parathyroid Glands. — That enlargement of the pituitary occurs after thyroidectomy in man has been known for a 774 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS considerable number of years. The enlargement affects more particu- larly the pars anterior, although changes are also described in the pars intermedia et nervosa. Accompanying the enlargement of the anterior lobe, vesicles containing colloid-like material often become developed in it, but even after the hypertrophy has proceeded to a considerable de- gree, this colloid does not contain iodine, nor does an extract have the same physiologic effect as one of the thyroid gland. It can not replace thyroid extract in the treatment of patients with goiter or myxedema, or ameliorate the symptoms produced in animals by the removal of the thyroid gland. Deposition of colloid-like material in the pars anterior also occurs in myxedema. Histologic changes in the pars intermedia et nervosa, although less pronounced than in the pars anterior, are never- theless said to be perfectly distinct following thyroidectomy, and to con- sist in an increase in the hyaline and granular masses which have already been described as present to a certain extent in the normal gland. Less direct evidence of an association in function between the pituitary and the thyroid is furnished by the similarity of the effects produced on the sexual functions and 'on the general development of young animals by the removal of either gland. In both cases the animals fail to grow properly; the sexual organs remain undeveloped; and the mental func- tions are infantile in type. In hypophysial deficiency, however, extreme adiposity is likely to be more marked than is the case in cretinism. 2. With the Sexual Organs. — That the pituitary gland has much to do with the development of the sexual organs has already been shown. Fur- ther evidence of a relationship between the sexual glands and the pitui- tary is furnished by the following observations. After castration en- largement occurs in the pituitary, and on histologic examination the gland is found to contain a large number of oxyphile cells, particularly in the pars anterior. This influence of the sexual glands on the pituitary is believed to depend on the interstitial cells present in them, for it has been found that if the ovary or testis is transplanted into other parts of the body after the castration, the changes in the pituitary do not occur, although, as we shall see, the transplanted gland becomes entirely atrophied except for the interstitial cells. The enlargement of the pitui- tary during pregnancy — an enlargement which often brings it to two or three times its normal weight — is further .evidence of its association with the ovary. 3. With the Suprarenals. — Association of function is suggested in this case by the fact that extracts of suprarenal and pituitary have very much the same effects on involuntary muscular fiber and glandular structures, and it is said that the two extracts mutually facilitate each other's action in this regard. It should be remembered, however, that pituitrin THE PITUITARY BODY 775 and epinephrine do not appear to act on exactly the same peripheral mechanism (see page 767). 4. With the Isles of Langerhans. — Since pituitrin affects carbohydrate metabolism, which is thought to be primarily controlled by the Isles of Langerhans, it is claimed by some observers that a relationship also exists between the pituitary and these structures. Injections of duodenal extracts are also said to cause a hypersecretion of pituitrin into the cerebrospinal fluid. CHAPTER LXXXV THE PINEAL GLAND This peculiar structure lies between the anterior corpora quadrigem- ina, and weighs about two-tenths of a gram. It is largest in the early years of life, and undergoes retrogressive changes after puberty. Micro- scopically it consists of epithelial cells arranged loosely in trabeculse, with large sinus-like capillaries between them; neuroglia and sometimes muscle-fiber cells are also present. Curious globules of calcareous mat- ter (brain-sand) are also found, especially in the pineal gland of man. The gland is developed from an evagination of the third ventricle, and vt is homologous with the so-called median eye of reptiles. The functions of the pineal gland are obscure. In cases where its extirpation has been successfully accomplished (in the fowl), it has been found that the body growth is stimulated and the sexual characteristics developed more quickly. This result would seem to indicate that the clinical observation that tumors of the pineal gland are associated in young boys Avith abnormal growth of the skeleton and with the early development of the secondary sexual characteristics, depends on the fact that a condition of hypopinealism is produced by the growth of a tumor. The immediate effects of the injection of extract of pineal gland are not characteristic, consisting merely of a fall in blood pressure, which is,- however, obtainable when an extract of practically any cellular organ is injected. Prolonged administration of an extract to growing animals is said to accelerate the growth and to bring about a precocious develop- ment of the sexual organs; but this result is somewhat difficult to inter- pret, for, as we have just seen, similar changes occur after experimental removal of the gland. THE GONADS OR THE GENERATIVE ORGANS The Generative Glands of the Male The structures which are responsible for the well-known influence of the testicles on the development of the male sexual characteristics are the so-called interstitial cells of Leydig, which consist of polygonal- shaped epithelial-like cells, with well-marked nuclei and nucleoli. Lipoid 776 THE PINEAL GLAND AND THE GONADS 777 granules, staining black with osmic acid, are also present in the cyto- plasm. The degree of development of the interstitial cells varies in dif- ferent animals, being marked in the cat and man and ill-marked in the rat and rabbit. In animals which show seasonal changes in sexual activ- ity, the cells are most prominent between the periods of sexual activity, when the semeniferous epithelium is less evident. They also become prominent in cases where the semeniferous epithelium is atrophied, either as a result of disease or following ligation of the vas deferens done in such a way that the artery and nerves to the testicles are not included in the ligature. When the testicle or a portion of it is grafted into another part of the body, the semeniferous epithelium degenerates, but the interstitial cells remain alive and become quite prominent. It is believed that the interstitial cells are responsible for the production of an autacoid that has to do with the development of accessory sexual characteristics. Tlie effects of castration are not significant in animals below the verte- brata. In all of these, however, they are very pronounced. The cas- trated male frog fails to show development of the thumb pad, but this development immediately ensues if portions of testis from another frog be placed in the dorsal lymph sac. In birds the results are more pro- nounced; in the castrated male chick the comb, spurs, wattles, etc., fail to develop, but will usually do so if some testis from another bird is trans- planted into its tissues. In mammals the effects are most striking in animals that develop marked male characteristics, such as the growth of antlers in stags. These fail to develop properly and are prematurely shed after castration. In man also, as is well-known from a study of eunuchs, castration has a very profound effect. Hair fails to grow on the face; the larynx remains undeveloped; the epiphyses are a long time in ossifying, so that the stature may become great, but at the same time the limb bones may be more delicate than usual ; the sutures of the skull are slow in closing ; and the whole architecture of a castrated male comes to be very like that of the female. Confirmatory evidence of the influ- ence of the testicles on the development of secondary sexual character- istics is afforded by the observation that malignant tumors of the testes in boys are associated with the premature development of the secondary sexual characteristics, and that these may recede after the removal of the tumor. As a result of castration, interesting changes have also been observed in other ductless glands. Thus, the suprarenal cortex and the thymus become enlarged, whereas the thyroid and pituitary become atrophied. The metabolic functions also become tardy, as is evidenced by a tendency to the deposition of fat. 778 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS When the castration is performed on an adult man, the above changes in the sexual characteristics are of course not so evident, although the prostate, etc., atrophy. The effect on the metabolic functions is, how- ever, very marked, there being a striking tendency to increased forma- tion of fat. It is interesting that accompanying this there should usually occur a lowering of the assimilation limit for carbohydrate, so that glyco- suria is very readily induced. We can not assume, therefore, as Gush- ing has done in the case of hypopituitarism, that the fat deposition is attendant upon an improper combustion of carbohydrate. These remarkable effects of castration have naturally prompted ob- servers to study the influence of injection of testicular extract on the development of sexual characteristics in different animals, but the re- sults have in general *been considered to be negative in character. The Female Generative Organs It is well known that, besides their function in producing ova, the ovaries also produce autacoids that have to do not only with the fixa- tion of the embryo in utero, but also with the changes that occur during pregnancy in the maternal organism. It is however at present uncertain as to where these autacoids are produced in the ovary. The two most likely sources are the stroma cells and the corpus luteum. In the stroma of the ovary of certain animals, groups of cells have been described having a different appearance from those of ordinary stroma cells. They have been called the interstitial cells of the ovary, and are believed to be analogous with the similar structures found in the testicle. It is possible, however, that these interstitial cells are nothing more than cells derived from previous corpora lutea. The latter are formed by proliferation of the follicular epithelium which remains after extrusion of the ovum, and by the ingrowing into the follicle of the so-called theca cells and blood vessels. The fully developed corpus luteum in most animals consists of cells arranged in trabeculae converging toward the. scar which formed at the place where the follicle had burst. The luteal cells, as they are called, are characterized by containing considerable quantities of lipoid material. That the ovary produces some autacoid is evidenced by both clinical and experimental observations. Thus, if both ovaries are removed in a young animal (oophorectomy or spaying), it is well known that not only does the uterus fail to develop properly, but the external changes characteristic of puberty in the female fail to materialize, although act- ually the general effects are not so pronounced as they are in the male after castration. Menstruation does not set in; the mammary glands fail to develop ; and there is a tendency for the hair to grow as in the male. THE PINEAL GLAND AND THE GONADS 779 When the operation is performed in adult life, the changes are not very pronounced, except that menstruation ceases and the uterus and mam- mary glands atrophy. Metabolism also becomes altered, causing a tendency to the deposition of fat, and in the case of the human animal at least, there is frequently evidence of mental disturbance. Attempts to acquire more definite information regarding the physio- logic effects of the ovarian autacoid have recently been made by Schafer and Itagaki.60 Extracts were prepared from the corpus luteum or Graafian follicles or from the hilum ovariae, and observations were made on the effect produced on the behavior of the chief forms of unstriated muscle by adding the extracts to isolated preparations of uterus or intestine or by injecting the extracts into animals. Applied to the isolated prepa- rations, extract of follicular tissue or of liquor folliculi was found to increase the force and rate of the rhythmic contractions of the uterus as well as its tone, whereas inhibition was produced when extract of the hilum was used. Extract of corpus luteum, when injected into the veins, was found to cause the uterus to increase its contraction or if quiescent to begin contracting. It was further noted that extracts of hilum caused a fall in arterial blood pressure, whereas those of corpus luteum had little or no effect. It would appear from these observations that the extracts contain two different autacoids, one having a hormonic and the other a chalonic action on plain muscular fiber. Extract of corpus luteum when intravenously injected also stimulates the outpouring of the milk from the mammary glands, although not so markedly so as extract of pituitary gland. This pituitary-like action is not obtained with extracts of ovary that do not contain corpora lutea. Besides being concerned in the outpouring of milk, corpus luteum has also been shoAvn to be related in some way to the development of the mammary gland during pregnancy. These glands become developed in young virgin rabbits after the continuous administration for a month or so of extract of corpus luteum, and they also develop in unimpreg- nated animals when the corpus luteum is made to develop by artificial means such as puncturing the Graafian follicle. Furthermore, destruc- tion of the corpora lutea in a pregnant rabbit arrests development of the mammary glands. The corpus luteum has also an important func- tion in connection with the formation of the uterine decidua and the fixation of the embryo. Thus, after destruction of. the corpus luteum at an early period in pregnancy, the embryo fails to become adherent to the uterus. 780 THE ENDOCRINE ORGANS, OR DUCTLESS GLANDS DUCTLESS GLANDS REFERENCES* (Monographs) ssVincent, Swale: Internal Secretions and the Ductless Glands, Ed. Arnold, London. 59Biedl: The Internal Secretory Organs, Wm. Wood & Co., 1913. coSchafer, Sir E. A.: The Endocrine Organs, Longmans. Green & Co., New York and London, 1916. (Original Papers) eiFulk, M. E., and Macleod, J. J. E.: Am. Jour. Physiol., 1916, xl, 21. 62Folin, O., Cannon, W. B., and Denis, W. : Jour. Biol* Chem., 1913, xiii, 447. eaCannon, W. B., and Gray, EL: Am. Jour. Physiol., 1914, xxxiv, 232; also with Men- denhall, W. L.: Ibid., 243 and 251. e^Hartman, T. H., and others: Am. Jour. Physiol., 1915, xxxviii, 433; ibid., 1917, xliii, 311; ibid., xliv, 353; ibid., 1918, xlv. esHoskins, E. G. : Am. Jour. Physiol., 1912, xxix, 363 ; Jour. Pharm. and Exp. Therap., 1911, iii, 93 ; Am. Jour. Physiol., 1915, xxxvii, 471 ; ibid., 1916, xli, 513. eeStewart, G. N., and Eogoff, J. M.: Jour. Lab. and Clin. Med., 1918, iii, 209. See full bibliography by Eogoff in this paper. 67Elliott, T. E. : Jour. Physiol., 1912, xliv, 374. ssStewart, G. N.: Jour. E"xp. Med., 1911, xiv, 377; ibid., 1912, xv, 547; ibid., xvi, 502. esStewart, G. N., Eogoff, J. M., and Gibson: Jour. Pharm. and Exper. Therap., 1916, viii, 205. 7oMeltzer, S. J. : Deutsch. med. Wchnschr., 1909, xiii. "Stewart, G. N. : Jour. Exper. Med., 1912, xv, 547. "Cannon, W. B., et al.: Am. Jour. Physiol.j 1911, xxviii, 64; ibid., 1914, xxxiii, 356; also Bodily Changes in 'Hunger, Fear, and Eage, Appleton, 1915. "Cannon, W. B., and Cattell, McKeen: Am. Jour. Physiol., 1916, xli, 74. ''^Macleod, J. J. E., and Pearce, E. G. : Am. Jour. Physiol., 1912, xxix, 419. fs Marine, D. : Personal communication. reMarine, D. : Jour. Exper. Med., 1914, xix, 89. ^ "Marine, D., and Lenhart, C. H.: Jour. Exper. Mea., 1910, xii, 311; ibid., 1911, xiii, 455; also Bull. Johns Hopkins Hosp., 1910, xxi, 95. "Marine, D., and Kimball, O. P.: Jour. Lab. and Clin. Med., 1917, iii, 41. 7oKendall, E. C. : Boston Med. and Surg. Jour., 1916, 175, 557; also Proc. Am. Physiol. Soc., Am. Jour. Physiol., 1918, xliv. sopaton, Noel and Finlay: Quart. Jour. Exp. Physiol., 1917, x, 203. Paton, Noel, Finlay and Watson, A. : Ibid., 233, 243, 315, and 377. siMacCallum, W. G., etc.: Jour. Exper. Med., 1909, xi, 118; ibid., 1913, xviii, 646; Jour. Pharm. and Exper. Therap., 1911, ii, 421. ssCushing, Harvey: The Pituitary Body and Its Disorders, J. B. Lippincott Co., 1912. saHorsley, V. : Brit. Med. Jour., 1885, i, 111. »iEobertson, Brailsford, and Eay, L. A.: Jour. Biol. Chem.. 1916, xxiv, 347, 363, 385. 397, 409. ssClark, L. N. : Jour. Biol. Chem., 1915, xxii, 485. *The numbering is in continuation with that for metabolism. V IX THE CENTRAL NERVOUS SYSTEM CHAPTER LXXXVI THE EVOLUTION OF THE NERVOUS SYSTEM The nervous system of the higher animals consists of the nerve cen- ters, and the nerves with their various interconnecting tracts. The nerve tract and centers are located mainly in the spinal cord and brain, where, by their interlacement, they form an extremely complex struc- ture. The exact position of the centers and the course and connections of the tracts with the centers are problems which, under the title of neurology, have during recent years been contributed to more particu- larly by the anatomist and the pathologist. The information thus gathered tells us the possible tract or tracts of nerve fibers through which the various centers may communicate either with one another or with the structures outside the central nervous system upon which they act. Since each of these centers may, however, be played upon by in- fluences coming from different regions of the body, it is evident that there must remain, as an equally important aspect of the subject, the investi- gation of the means by which the various available centers and tracts are brought into communication and action at the proper time. In other words, we must investigate the functional uses of the available paths. We may compare the central nervous system with a telephone system, the exchanges representing the nerve centers, and the wires the nerve trunks. Any incoming wire may be connected by the operator with any outgoing wire, but a knowledge of how each wire runs does not tell us under what conditions the various wires will be connected for trans- mission of messages. It is the same with the nervous system ; the neurolo- gist can tell us how the tracts and centers run, but not the conditions under which they may act together. This it is the duty of the physiologist to ascertain. Since it is the degree of development of the central nervous system which determines an animal's position in the evolutionary scale, much information concerning the relative importance of the various parts of 781 782 THE CENTRAL NERVOUS SYSTEM it can be gleaned from a survey of the conditions under which the nervous system makes its appearance in the lowest forms of animal life. In the case of unicellular organisms, such as the ameba, the ap- plication of a stimulus to the surface causes a movement, because the protoplasm of the organism possesses, among its other properties, those of excitability, conductivity and contractility. In the case of multicel- lular organisms, on the other hand, some cells are set aside and spe- cialized for the assimilation of food, others for movement, others to receive stimuli from the outside, and yet others to compose the tougher tissues which protect the surface of the animal from injury. This loca- tion of specific function in specialized groups of cells makes it necessary, for the welfare of the organism as a whole, that some means of com- munication should be provided between the distant parts of the animal, for otherwise the cells which are occupied in absorbing food would be unable to move away or be protected from harm when some destructive agency approached them, and indeed the moving (muscle) cells could never know when the welfare of the organism as a whole demanded that they should become active. It is probable that, in some of the lower organisms, the messages trans- mitted from one group of cells to the others are carried by chemical substances present in the circulating fluid — hormones, as they are called (page 729). For the quick adaptation that is necessary in the struggle for existence, however, such hormones are usually too slow in bringing about the response, and very early in the evolutionary scale we find that cer- tain cells become differentiated for this special purpose. The cells thus specialized constitute the nervous system, their differentiation, as would be expected, being, however, antedated by that of. the cells that form the muscular tissues. In the sponges, for example, muscle cells become developed from ameboid epithelium and from a layer underneath the external epithelium. These muscle cells contract slowly so as to cause opening and closing of the small mouths, or oscula, on the surface of the sponge in response to movements in the sea water. They are in- dependent of any nervous structures. In certain Coelenterates the muscle cells respond a little more quickly than in the sponges, and this greater efficiency is found to be dependent upon the appearance of a localized, very primitive nervous system1. This nervous system consists of specially modified epithelial cells, or receptors, sending branches from their inner ends, which either come in con- tact with the muscle cells, or effectors, or become interlaced so as to form a network. In the region between the receptors and the effectors the net- work at first serves merely as a structure whereby the entire muscula- ture of the animal can be brought into harmonious action from a single THE EVOLUTION OF THE NERVOUS SYSTEM 783 point on the surface, as, for example, in the case of the sea anemone (No. 2 of Fig. 201). In the jellyfish, which in contrast to the sea anemone is a free moving animal, we find that the receptors are more highly special- ize and, therefore, much more sensitive, and that the impulses which they receive are transmitted to a more definite nerve network, capable not only of conveying the excitatory process from one part of the animal to another, Sponge 2. Sea anemone Simple form in earthworm Addition of association neurons in earthworm Fig. 201. — Diagram to show gradual evolution of nervous system from an epithelial cell ( Abdomen 26 Outside of thigh 2(5 Shin and sole 28 Back of forearm 33 Loin 48 That the sense of touch is located in spots — touch spots — can best be demonstrated on the calf of the leg. If this is shaved and then carefully explored with a fairly stiff hair, it will be found that there are only some twelve to fifteen spots in an area of a square centimeter at which the hair can be felt. Between these spots there is no sensation of touch. That these spots are composed of specialized receptors can be very clearly shown by pressing a fine needle into one of them, when no pain will be experienced but only a peculiar shotty sense of pressure. Careful examination of the position of the touch spots will further show that they are grouped around hair follicles, particularly on the side from which the hair extends — the windward side, we may call it. This fact explains the well-known experience that an object may be felt more acutely on a hairy surface than after that surface has been shaved. The hairs bend slightly when the object comes in contact with them, thus 794 THE CENTRAL NERVOUS SYSTEM causing pressure to be exerted on the hair follicles, so that the touch corpuscles in the neighborhood of the follicles, or perhaps the fine nerve plexus which surrounds them, becomes excited. The influence of hairs in increasing the touch sensation can be demonstrated by the von Frey method; for example, in one experiment over an area of 9 square mil- limeters of skin with hairs present, 2 milligrams were found to produce the sensation, whereas after the hairs had been removed, it required 36 milli- grams. The frequency of touch corpuscles differs very much in different parts of the body. They are most plentiful on the fingers, relatively infrequent over the skin of the back, and very scarce in the skin directly over bony surfaces. They are entirely absent from the cornea, the conjunctiva of the upper lid, and the glans penis. The adequate stimulus for touch is evidently deformation of the surface. Pressure exerted over all the touch corpuscles of a portion of skin is not felt. This can be demon- strated by dipping the finger into mercury. The pressure of the mercury is felt on the surface but not in the submerged portion of the finger. Touch is the most responsive of all the sensations. Thus, as. has already been noted, a tuning fork can be felt vibrating by the finger when to the ear its note is a continuous one, and the stimuli produced by a re- volving serrated wheel can be felt by the fingers as separate even up to a rate of five or six hundred stimuli per second. Adaptation is also a marked feature of the touch sense, as is the experience of every one who has worn flannel underclothing or a plate of false teeth. Closely related to the tactile sense is the power of discrimination be- tween two points. This is tested by finding at what distance the two points of a pair of calipers stand in order to be distinguished as separate. The result in any given part of the body varies a little according to whether the points rest on touch corpuscles and according to the rela- tionship of the calipers to the hair follicles. On an average, however, we may take the following distances in millimeters as being those at which the two points can be distinguished over different areas of the body: mm. Tip of tongue 1.1 • Volar surface of finger tip 2.3 Dorsal of first phalanx 6.8 Palm of hand 11.3 Back of hand 31.6 Back of neck 64.0 Middle of back, upper arm and side 67.1 It is clear from this list that the power of discrimination tends to dimmish in proportion to the lessening mobility of the part. It is greatest THE PROPERTIES OF EACH PART OF THE REFLEX ARC 795 at the tip of the tongue and the tip of the fingers; it is least on the relatively immobile skin of the back. These distances are much less when the points rest on two touch corpuscles. Under these conditions, for in- stance, the distance for the volar side of the finger tip or even for the palm of the hand may be only one-tenth of a millimeter; and for the arm and back it may become reduced to half a millimeter. Localization of touch is a very accurate process, at least in the most sensitive parts of the skin, but nevertheless it is very probably a mat- ter of education. An evidence of this is the fact that in the much more highly specialized retina the power of localization of objects in the visual field is a process of education and experience. For this reason a person from whom a congenital cataract has been removed, can not locate the objects which he sees until after he has learned by his experience of touch, taste, etc., to associate the portion of the retina stimulated with a certain part of the visual field. If this is true for the retina, it is also probably true for touch. The famous experiment of Aristotle is explicable on the same basis. If the fingers are crossed and a marble placed between the crossed fingers, it will be felt as double, since now it touches two skin surfaces which have not been accustomed to touch the same object, but educated to feel different objects. Experience associates those two skin areas with different objects, not with the same object. The Pain Sense. — It was at one time thought that the sensation of pain was due to overstinralation of any kind of receptor, but it is now known that for this, as for^ other skin sensations, there exist special receptors. Thus, it is found that in certain parts of the body, such as the cornea, and to a certain extent in the glans penis, pain receptors alone are present, and in disease the sense of pain may be entirely abolished, whereas that of touch remains, this condition being called analgesia. Overstimulation of a touch spot does not, as we have seen, cause pain but only a sense of pressure. Although pain is appreciated by special receptors, the charac- ter of the pain is dependent on the other sense receptors simultaneously excited; for example, a throbbing pain is due to the simultaneous pres- sure produced by dilated blood vessels, etc. A sensation of pain accom- panies certain reflexes of a protective nature (nociceptive reflexes, page 825) , and when the reflex is absent the part is likedy to suffer damage. On this account the pain nerves may be regarded as trophic nerves. The sense of pain may also occur in structures which are devoid of ordinary sensibility, such as the intestine and the ureter. CHAPTER LXXXVIII THE NERVE NETWORK In all animals above the Celenterates, no direct protoplasmic contiiiuitj- exists between the various neurons, the transmission of the nerve impulse depending on contiguity rather than continuity of the elements that con- stitute the reflex arc. This transmission may be effected through a syn- apsis coming in contact either with dendrites or with nerve cells. It is extremely difficult to know whether there is really any anatomic con- tinuity between the various fibers which form the network in the gray matter of the central nervous system. We shall not attempt to discuss this vexed question here, but in order that we may learn something of the possible functions of a nerve network, we may consider that present in the walls of the intestine (plexus of Auerbach and Meissner.) This plexus seems to have an important function to- perform in connection with the myenteric reflex (see page 466). At least it has been shown by Meek3 that after transsection of the intestine the muscular and epithelial structures be- come regenerated considerably earlier than the nervous plexus, but that the myenteric reflex, which, it will be remembered, is characterized by a wave of inhibition preceding one of contraction does not occur until after the plexus has been regenerated. NETWORK ON SKIN NERVES A very important type of nerve network, from the medical viewpoint, is that which is produced close to their receptor endings by the branch- ing of the afferent fibers of the skin. Through these branches the vas- cular reactions following the application of an irritant to the sensory surface take place without the intervention of any nerve cells. It used to be thought that such reflex vasodilatation depended upon the trans- mission of an impulse along an afferent neuron to an efferent vaso- dilator neuron, a view strictly in consonance with the neuron hypothesis. That such is not the case, however, is shown by the fact observed 'by Ninian Bruce* that irritants such as mustard oil applied to the skin or cornea continue to produce their usual reaction for some time after 796 THE PROPERTIES OF EACH PART OF THE REFLEX ARC 797 section of the posterior roots of the spinal cord, but fail to do so if the nerve fibers are cut and allowed to degenerate, or if the stimuli are blocked by applying cocaine to the skin. What actually happens is evidently that the impulse set up by the irritant as it travels up the afferent fiber passes on to one of the branches above referred to, along which it then proceeds to the blood vessels, which it causes to dilate. That such vasodilator impulses may be transmitted down the fibers of an afferent nerve has been confirmed by Bayliss, who found that vaso- dilatation occurred in the hind limb when the posterior spinal roots were stimulated (see page 234). Post, roof gang.- Fig. 206. — Diagram to show axon reflex of sensory nerve fiber of skin. A stimulus applied to the skin is transmitted by the sensory fiber (AT), part of it going to the spinal cord (SC), and part of it passing by the collateral (C) to the arteriole (A), which it causes to dilate. In this peripheral branching of the afferent fibers of the skin, we have therefore a sort of neuropile which, like that of certain forms of Celenterates (see page 782), is capable of serving as a pathway for the transmission of a sensory impulse to an effector organ without the in- tervention of nerve cells. Such a reflex is known as an axon reflex, and it is evident that it may occur through any fiber which gives off branches, one traveling to a sensory surface, the other to some effector organ, as occurs in the hypogastric nerves to the bladder (see page 883). THE SYNAPSIS At the point of contact between a branch of one neuron and a nerve cell of the next, we have seen that there exists a structure known as the synapsis. Although this is described by histologists as a tuft-like 798 branching of the end of the axon (Fig. 207), it may really consist of a sort of membrane — the synaptic membrane. It permits the nerve im- pulse to pass in one direction only, from synapsis to cell. Of what this membrane may be composed, we do not know, but there can be no doubt as to its great functional importance in connection with the in- tegration of the central nervous system; for example, failure of an im- pulse to pass between two neurons may be due to retraction of the synaptic membrane from the cell, or to alteration in its permeability to- wards the nerve impulse, perhaps as a consequence of changes in surface Fig. 207. — Arborization of collaterals from the posterior root fibers around the cells of the posterior horn. A, ascending fiber in posterior columns; B, collaterals; C, cells of posterior horn; E, synapsis. (From Ramon y Cajal.) tension. Similar changes may also be brought about by the action of electrolytes or by chloroform, strychnine, and other, drugs. As we shall see when we come to study the reflexes of the higher animals, there can be little doubt that it is in the synaptic membrane that many of the peculiarities reside which characterize conduction in a reflex arc as compared with that in a nerve trunk. The phenomena of summation, of reciprocal inhibition, of facilitation, etc., are undoubtedly depend- ent upon such alterations. The synapsis is also almost certainly the seat of fatigue in the central nervous system, and it is possibly the structure whose physiologic activity becomes upset in surgical shock. THE PROPERTIES OF EACH PART OF THE REFLEX ARC 799 THE NERVE CELL Aside from being a meeting place of fibers coming from various sources, the nerve cell may have other functions, such as that of rein- forcing impulses, just as a relay may reinforce an electric current. It is also responsible for maintaining the nutrition of the axon with which it is connected. In the case of the posterior root fibers of higher ani- Fig. 208. — Normal cell from the anterior horn, stained to show Nissl's granules, a, the axon. (From Howell.) mals, this function is probably the most important which the cell per- forms, for it has been found by separating the ganglia from their blood supply in the frog that, although the cells degenerate in about two weeks, sensory impulses continue to be transmitted through the gan- glia. Similar observations have been made in the case of the crab, in which the cell bodies of the neurons lie on the surface of the ganglion 800 THE CENTRAL NERVOUS SYSTEM mass, from which they can be separated, leaving merely the neuropile, through which, however, the reflex continues to be conveyed. After a time, of course in this case also the reflex disappears, because an axon can not live indefinitely after it has been separated from its nerve cell. These facts regarding the general function of the nerve cell arouse our curiosity as to its morphologic structure. When nerve cells are fixed and stained in various ways they show certain elements in the Fig. 209. — Part of an anterior cornual cell from, the calf's spinal cord, stained to show neurofibrils. ax, axon; a, b, c, dendrites. (From Bethe.) cytoplasm — namely, (1) large granules or masses, which stain deeply with basic dyes .and are called Nissl bodies (Fig. 208), and (2) a fine network of fibrils passing through the cell substance from one process or clendrite to another — neurofibrils (Fig. 209). These appearances in fixed and stained preparations are possibly, however, entirely artificial ; for when nerve cells are preserved in a living state — by being suspended in some of the animal's own lymph or blood serum — it is found, when they are ex- amined by the aid of the ultramicroscope (see page 52), that the cytoplasm THE PROPERTIES OP EACH PART OF THE REFLEX ARC 801 is composed of a viscous fluid full of extremely minute granules, each of which apparently consists of a colloidal solution surrounded by a lipoid envelope (Fig. 210). When the temperature is raised, the granules dis- appear, and when the cells are deprived of oxygen, the cytoplasm and nucleus become swollen. A similar swelling of the cell and nucleus super- venes upon section of the axon; and in stained specimens the Nissl granules disappear and the protoplasm stains diffusely (chromatolysis). In embryonic life the processes of the nerve cells appear to be capa- ble of undergoing a certain amount of ameboid movement, and fibers grow out from them, indicating, therefore, that in the development of the nervous system the nerve cells appear first, and the nerves sub- sequently grow out from them to their proper destination. Prolifera- tion of isolated tissue cells in vitro has been observed for many other Fig. 210. — Living nerve cells (from the dorsal root ganglia of a dog three days old) examined by the ultramicroscope. There are no Nissl bodies or neurofibrils, only fine particles, present in the protoplasm. (From Marinesco.) tissues, such as cardiac muscle, renal epithelium and connective tis- sue. Its occurrence indicates that the therapeutic principle that the aim of treatment should be to give the diseased organ a rest so that by cell regeneration it may recover its lost function, is one which may ap- ply to the nerve tissues of young animals. Whether adult nerve cells may regenerate is as yet not certain. This growing out of nerve fibers from their cells is the essential na- ture of the development of the nervous system in the developing animal. At birth, unlike the cells of other tissues, those of the central nervous system are already provided. No new ones are added during postnatal life. The axons gradually develop from this inherited stock of nerve cells, and by connecting with other neurons serve to bring about the integration which is the important characteristic of the adult nervous 802 THE CENTRAL NERVOUS SYSTEM system. The more complex the integration, the higher the intelligence of the animal. Besides performing these functions the nerve cells serve as store- houses for memory impressions, certain types of them being especially adapted for this function. The differences observed in the relative thick- ness of the cell layers composing the cerebral cortex are more or less associated with the function which it can be shown the different areas of this possess. Nerve cells are extraordinarily sensitive to deficiency in oxygen supply, and yet little evidence of oxygen consumption by the brain can be revealed by the usual methods of investigation (page 396) . THE INTERMEDIATE OR INTERNUNCIAL NEURON It would be profitless at this stage to consider the possible influences that, the intermediate neuron may have on the impulses passing along the reflex arc. Before doing so we must see how the problem can be approached, for it is plain that the neuron in the case of the simpler reflexes is too short to make any investigation of its peculiar functions a possibility. We must study the characteristics of some type of re- flex in which this neuron is drawn out, such as the scratch reflex, in which, as we shall see, it extends from- the shoulder area of the cord to the lumbar region. SPINAL SHOCK Having become familiar with the peculiar properties of each of the structures which go to make up the reflex arc, we may now proceed to consider the function of the arc as a whole. It may be well first of all to consider briefly the experimental method by which such studies may be made. The object aimed at is to simplify the conditions as much as possible, for it will be evident that, in the intact nervous sys- tem, with 'the brain exercising a dominating influence over the great majority of all the reflexes, it would be impossible by applying a given stimulus, to predict exactly what kind of reflex response it might call forth. The reflex will be conditioned upon the accompanying influence Avhich the brain exercises on the reflex involved. In order to render the reflex unconditioned, we must remove the in- fluence of higher centers. This can be done experimentally for the re- flexes of a great part of the body by cutting the spinal cord above the level of the segment in which the reflex under investigation resides. Some of the reflexes elicitable from the cord isolated in this way in- volve only one or two neighboring segments, whereas others spread over several. The reflexes which have been most extensively employed are those which involve the musculature of the hind limbs. Since some of the receptors concerned come from the skin of the flank and shoul- der areas, the section is usually made at the upper end of the thoracic region of the spinal cord. Spinal Shock in Laboratory Animals Immediately after the operation a profound condition of depression sets in, involving all the reflex arcs in the separated portion of cord. This condition is known as spinal shock. It supervenes in all classes of ani- mals having a spinal cord, but is much more profound in the higher than in the lower animals. As a result of this depression, the part of the body below the section exists in a limp and flaccid condition, and the application of even very strong stimuli to the skin will evoke no form of reflex movement. In the case of the lower animals, such as the frog, 803 804 THE CENTRAL NERVOUS SYSTEM the condition begins to pass off in from twenty minutes to half an hour, after which a stimulus applied to the skin of the foot is followed by a typical flexion movement at knee and hip, the so-called flexion reflex. In the rabbit very little reflex response is elicitable for several hours after the operation, but in a few days the reflexes return completely below the level of the section. In the dog, on which a great deal of work has been done, the involved regions of the body are profoundly paralyzed. The skin is in a more or less unhealthy, unnatural condi- tion, the surface cold, the hairs ruffled; and if care is not taken, the slightest abrasion of the surface may result in a nasty ulceration. On account of the paralysis of the centers of micturition and defecation, there is also incontinence of urine and of feces. The Reflexes in the Spinal Animal With reasonable attention, however, the dog makes a wonderful re- covery. After an interval of two weeks the hind limbs, although com- pletely paralyzed so far as voluntary movement is concerned, begin to show considerable signs of improvement. The first reflexes to return are those concerned with the deeper structures, such as the vascular reflexes, thus bringing the skin back to its normal temperature and condition. The reflexes of micturition and defecation also soon return, so that the animal no longer suffers from the continuous discharge of urine and feces. About the same time the knee-jerk becomes elicitable. This reflex is obtained by tapping the tendon which connects the patella with the tibia, the response being a smart contraction of the extensor muscles of the knee joint. The flexion reflex also begins to reappear. This is elicited by applying a pinprick or other hurtful stimulus to the skin of a lower extremity, and when fully developed consists in a flexion of the knee and hip joints. The evident object of this move- ment is that the stimulated parts may be removed from the source of stimulation, and it is plain that all stimuli that produce the flexion reflex are such as would cause in the intact animal a sensation of pain. Such stimuli are thus classified as nocuous, and the reflex is styled a nociceptive reflex. Accompanying flexion of the stimulated limb the op- posite or contralateral limb usually undergoes a definite extension, called the crossed extension reflex. The occurrence of this together with the flexion of the stimulated limb is an important thing to remember in testing the reflexes in man. Malingerers who attempt to make it ap- pear that they have some lesion of the spinal cord may know that if such lesion exists no movement of the leg occurs when the skin is stimulated, but they are unlikely to know that under these conditions the opposite leg also fails to show a simultaneous extension. REFLEXES IN THE SPINAL ANIMAL AND SPINAL SHOCK 805 That the nociceptive reflexes should be among the first to return after spinal transection is of considerable interest as indicating their im- portance in the protection of the animal from injury. They are the essential reflexes of defense, and it is considerably later in the recovery of the animal before reflexes dependent upon stimulation of other tac- tile receptors begin to show themselves. The most important of this latter group of more special reflex movements include the so-called scratch reflex and the extensor thrust. The scratch reflex, as its name implies, is the scratching movement of flexion and extension of the hind limb at a rate of about four contractions per second that occurs when a mechanical stimulus is applied to the flank and shoulder area of the animal. For example, if we gently draw a pencil or the fingers back- ward and forward among the hairs on this region of the spinal animal, the corresponding hind limb will be brought up so that the claws are approximately at the place stimulated, and the limb thus directed will undergo a series of flexions and extensions, designed evidently for the purpose of scratching the area of skin that has been stimulated. If the stimulus is a weak one, only the initial stages of the movement may occur, such as the preliminary flexion of the leg. As we have already stated, the receptive stimulus calling forth this reflex is ve"ry specific in nature. A pinprick or rough friction of the reflex area will not produce it, nor will the application of heat or of a single electric shock. The most adequate stimulus is one simulating as nearly as possible the con- dition which would be produced by the movement on the flank of the animal of some insect. This more or less complicated scratch reflex can of course also be elicited in animals whose spinal cord has not been cut, but we can not predict in such cases whether the reflex will occur. The brain may inhibit the reflex arc and -prevent the movement. In a spinal animal, however, the reflex always occurs provided an adequate stimulus is applied. The great importance of the scratch reflex in the study of the physiology of the spinal cord rests in the fact that a large stretch of cord is involved in the reflex path. The afferent impulses must enter at a much higher level than the efferent impulses leave, and between these two points there must exist a long intraspinal neuron (see- page 813). This permits us to study many conditions influencing reflex action which otherwise in a reflex located in one segment only it would be im- possible to investigate.4* The extensor thrust is elicited by applying pressure to the pad of the paw or the sole of the foot. It consists of a quick extension movement of the corresponding limb usually with a flexion of the opposite limb. After complete recovery from shock, the paralyzed parts of the body are capable of performing even more complex movements than those al- 806 THE CENTRAL NERVOUS SYSTEM ready mentioned. For example, if the animal is held up with the hind legs hanging down, these will often exhibit rhythmic flexion and exten- sion movements, with the two limbs acting alternately, as they would in walking or running. This is sometimes called the mark-time reflex. Another complicated movement may be produced by placing the animal in water, when it may make the movements of swimming, but its swim- ming will not be sufficient to keep it on the surface. These swimming movements are more perfect in the spinal frog. After complete recovery from spinal shock, the hind limbs are more or less in a condition of extension contracture; the vascular and other visceral reflexes are in perfect condition, and a marked rise in blood pressure occurs when one of the sensory nerves of the hind limb is stimulated — an experiment which can be performed in such animals without the administration of any anesthetic, since the animal feels no pain. In female spinal animals impregnation may occur and preg- nancy proceed in normal fashion accompanied by the usual secretion of milk. The significance" of this fact will be dwelt upon later. SPINAL SHOCK IN MAN As we ascend the animal scale we find that recovery from spinal shock takes longer and longer to occur and becomes less and less perfect. In the case of man, recovery is never complete, for a permanent condition, M-hich has been called "isolation dystrophy," supervenes before the symptoms of shock have been recovered from. The tendon jerks are permanently abolished in complete lesions of the cord in man, and even when the lesion involves only one lateral half of the cord, this reflex is either entirely absent or very feeble on the corresponding side, though normal on the other (Holmes5). Severe lesions above the lower dorsal region practically always leave the legs in a permanently flaccid con- dition, with accompanying atrophy, but sometimes automatic movements of flexion and extension, like those of the mark-time reflex, may set in. When the injury of the cord is less severe, the limb musculature is flaccid and toneless for some time, the tendon jerk and the abdominal and cremasteric skin reflexes being entirely absent. After some time, however — it may be as early as ten days — the muscles begin to reac- quire some tone, and a little later the tendon jerk becomes elicitable. Regarding the behavior of the flexion reflex after spinal injuries in man, it has been found that the part of it known as the Bakinski reflex is not elicitable after severe lesions, but in those that are less severe a flexion of the great toe may occur on stimulation of the sole. Later this movement may be accompanied by contraction of the hamstrings, and later still, in favorable cases, by flexion at knee and hip. In these REFLEXES IN THE SPINAL ANIMAL AND SPINAL SHOCK 807 cases also the Babinski reflex changes from a flexion to an extension of the great toe. It is important to note in connection with the above association of movements, that the sensory area of the sole is connected with the same segment of the spinal cord that furnishes the motor fibers to the flexors of the toes and the hamstrings (first sacral.) The recovery after shock therefore sets in earlier for unisegmental reflex areas than for those involving several segments. The Cause of Spinal Shock The relationship of the profundity of spinal shock to the phylogenetic position of the animal indicates that the shock must be due to the isolation of the lower spinal segments from the higher centers (Pike6). It has been suggested that the spinal section in the higher but not in the lower animals breaks a nervous pathway in which normally the reflex impulses travel. According to this view, the afferent impulse, when it enters the spinal cord in the lower animals, chooses the shortest possible route to the effector neuron of the same or closely adjacent seg- ments bv the collateral branches springing from the sensory neuron. In the higher animals, however, it would appear that, although this local spinal pathway is present and mav be taken, yet it is usually passed bv and the impulse travels up to the higher centers, from which it is then transmitted bv the pyramidal tracts to the motor neurons con- cerned. This would appear to be the pathway for nervous reflex im- pulses in higher animals — the beaten track. When the spinal cord is severed, therefore, the condition of shock supervenes because impulses have not yet learned -that they may find a shorter road to the motor neuron by the collateral than by the pathway which they usually travel. Thev learn this only after some time, which explains the slow re- covery of the reflexes from shock. It is obviously a difficult matter to supply direct proof in support of the above hypothesis of the cause of spinal shock, but besides the in- direct evidence furnished by observations on the degree to which this condition supervenes in different groups of anmials, the hypothesis also conforms well with all the other facts which we know regarding the condition. For example, it is well known that the portion of the body above the transection of the spinal cord in no way suffers from the shock. Sherrington has described a monkey the cord of which was cut below the cervical region, and which immediately after the opera- tion amused itself by catching flies with the anterior extremities, whereas the posterior extremities were in a condition of the profoundest shock. Such experiments further indicate that the shock can not be dependent 808 THE CENTRAL NERVOUS SYSTEM upon the lowering of arterial blood pressure which a section of the cord higher than the mid-dorsal region must entail. The poor nutritive condition of the skin which we have seen to exist in the hind limbs in shock, shows that the blood vessels in them are profoundly dilated, but evidently the fall in blood pressure has nothing to do with the faulty conduction through the spinal cord, for such a fall would affect the centers for the fore limbs as well as those for the hind, and yet the former show no symptoms of shock. Exactly similar shock is obtained by any section of the spinal cord as high up as the medulla. Of course as the section is made higher and higher up, the resulting paralysis becomes more and more marked, and may reach such a degree of severity that recovery of the animal be- comes an impossiblity. When we come to consider the functions of the various parts of the brain, we shall have occasion to study the effects following section at higher levels of the cerebrospinal axis. Meanwhile, however, it is im- portant to note that when a section is made across the crura cerebri, so that the cerebral hemispheres alone are isolated from the rest of the nervous system, a condition of contracture of all of the extensor muscles occurs. This condition is known as decerebrate rigidity. CHAPTER XC PHYSIOLOGICAL PROPERTIES OF THE SIMPLE REFLEX ARC We may now proceed to study the properties of reflex action occur- ring through the isolated spinal centers of a spinal animal. There are two aspects of the question to be considered: (1) the properties of a single reflex arc, and (2) the action or influence of one reflex arc on another. The importance of the latter will be evident when it is re- membered that complicated muscular movements depend for their proper coordination entirely on the interaction between the various reflex arcs which compose the nervous system. This interaction, as already ex- plained, has been called by Sherrington the integration of the nervous system. Probably the simplest way to study the physiologic properties of the simple reflex is to compare the mode of conduction of a nerve impulse through it with conduction along a simple nerve trunk. By comparing the two modes of conduction we shall be better able' to appreciate the modifications to which the impulse is subjected by con- duction through the reflex arc. The important points are these: 1. The Latent Period. — The latent period, or period which intervenes be- tween the moment of application of the stimulus and the response, is very short in the case of a nerve trunk, and under normal conditions always the same, but is quite variable and sometimes very long in the case of a reflex arc. Thus, in the case of the conjunctival reflex, which is produced by applying a stimulus to the corneal conjunctiva (causing a closing of the eyelids), the reflex time is very short and invariable, whereas in the case of the scratch reflex it may vary from two and a half to three and a half seconds, according to the strength of the stimu- lus. The seat of delay in the reflex arc is probably in the synapse, but its cause is obscure. 2. Grading of Intensity. — In a nerve trunk the intensity of the im- pulse is more or less proportional to the strength of the stimulus. This can be judged by observing either the action current in the nerve by means of a galvanometer or the response of the end organ; e. g., muscle, attached to the nerve. In the case of a reflex arc, on the other hand, there is by no means so evident a parallelism between stimulus and response. Reflexes, however, vary considerably in this regard. The conjunctival reflex and the extensor thrust behave according to the so-called "all or nothing principle;" i. e., the intensity of the response is more or less independent of the strength of the stimulus. In other reflexes, such as the flexion reflex and the scratch reflex, the intensity of the response 809 810 THE CENTRAL NERVOUS SYSTEM is much more nearly proportional to the strength of the stimulus. Thus, a feeble stimulus applied to the flank calls forth only a slight flexion of the hind limb of the same side, whereas a stronger stimulus sets going a typical scratching movement. 3. After-effect. — When a stimulus is removed from a nerve, the effect which it produces, as judged, for example, by the action current, im- mediately disappears. There is no after-response. In reflex arcs, how- ever, such a phenomenon is usually well marked. Particularly is this the case in the flexion and scratch reflexes of the spinal dog. A mo- mentary stimulus of optimal strength applied to the scratch skin-area may produce no immediate response, but after its removal a violent scratching movement may set in. This after-discharge, in cases in which the stimulus is strong, may indeed, as in the flexion reflex, be more marked than the response during the time of application of the stimulus. In this particular reflex, the after-discharge often takes the form of a clonus, with a rate of contraction of from seven and a half to twelve per second. The crossed extension reflex also has a very pronounced after-discharge, which may outlast the stimulus for from ten to fifteen seconds. Regarding the phenomenon of after-discharge, Sherrington has stated that there is "no feature of the conduction of a reflex arc which distinguishes its mechanism more universally from that of a* nerve fiber, tract or trunk than lengthy after-discharge." 4. Summation. — When a subliminal stimulus — that is, one that has in itself no visible effect — is frequently repeated in the case of a nerve, no response occurs. In the case of a reflex arc, however, such repeti- tion of subliminal stimuli ultimately calls forth response. This sum- mation is very evident in the case of the scratch reflex; e. g., one or two electrical stimuli applied to the scratch field-area call forth, as a rule, no movement of the corresponding hind leg, but if these same stimuli are frequently repeated, the typical reflex scratching movement will occur. Evidently, then, in a reflex arc there is a considerable amount of resistance towards a single stimulus, which resistance is overcome by a succession of stimuli. In other words, the threshold of the excitability of the reflex mechanism becomes lowered as a result of its previous stimulation. Each stimulus excites the sensory surface so that it responds more easily to the succeeding stimulus. 5. IrreveTsibility of the Direction of Conduction. — This is well illus- trated in the so-called Bell-Ma gendie law of conduction in the spinal nerve roots. A motor impulse travels out of the cord by the anterior roots, while a sensory impulse travels in by the posterior. This direc- tive influence can not depend on the nerve trunks or the nerve cells, for nerve trunks conduct equally in both directions, and so also must the nerve cell. The irreversibility must therefore depend on the synaptic PHYSIOLOGICAL PROPERTIES OF THE SIMPLE REFLEX ARC 811 connections. It can be demonstrated by observing the action cur- rent produced in the spinal cord by stimulating the anterior or posterior spinal roots. In the former case no action current is observed, but it is very evident in the latter case. 6. The Refractory Period. — This has been well denned by Sherrington as being "a state during which apart from fatigue the mechanism shows less than its full excitability." "We are already familiar with the re- fractory period in the cases of the heart muscle and the musculature of the esophagus and intestine. For example, the application of a stimu- lus to the quiescent frog heart while it is contracting in response to an im- mediately preceding stimulus fails to produce any further effect. The re- fractory period is extremely brief (one thousandth of a second) in a nerve trunk, but is much longer in a reflex arc, being probably longest in the case of the scratch reflex, in which it is demonstrated by the fact that, however frequently we apply suitable stimuli to the sensory surface, the rhythm of response of the contracting limb is always the same. After each stimulus, therefore, a refractory period must become developed during which a repetition of the stimulus has no effect. It is evident that the existence of the refractory period is the factor responsible for the rhythm of the movements. It is interesting to consider the exact structure of the reflex arc that is responsible for the existence of the refractory phase. It obviously can not be a function of the motor neuron, for through the same motor neuron may be discharged, at one time, impulses which bring about the scratching movement and, at another, those causing a tonic flexion of the same muscles. Nor can the seat of the refractory period be in the sensory area of the skin or the afferent neuron, for if a scratch move- ment is elicited by stimulation at a point A in the proper skin area, the rhythm of response which it calls forth will not in any way be altered by the application of a second stimulus applied at B at some distance from A and having a different frequency (Fig. 211). There is evidently, therefore, some part of the reflex arc that is common to impulses starting both at A and at B, for if in each of these spots a refrac- tory phase occurred, then there would be interference before the two im- pulses had reached the centers of the spinal cord. By exclusion, there- fore, "the seat of the refractory phase seems to lie somewhere central to the receptive neuron in the afferent arc "-—(Sherrington18). Many other types of reflex activity illustrate rhythm due to the re- fractory phase. Two laboratory examples may be given: (1) When the central end of an afferent root is stimulated in the lumbar region of the spinal cord, the movement produced is distinctly rhythmic .in character. (2) Upon stimulating the central end of the sciatic nerve in a frog whose spinal cord has been cut some days previously, a clonic 812 THE CENTRAL NERVOUS SYSTEM action of the contralateral foot occurs, and the rate of the rhythm, is not affected by variation in the frequency of the stimulus. In all the above cases the refractory period may be held responsible for the rhythmic nature of the contraction. In other reflexes it exists for another purpose. In the case of the extensor thrust, which it will be remembered is elicited by pressure applied to the pads of the plantar aspect of the foot, the momentary extension of the leg lasts only for a little less than two-tenths of a second, but is followed by a refractory Fig. 211. — Tracing from the hind limb of a spinal dog during the scratching movements pro- duced by applying stimuli at two skin points (A and B), the application of the stimuli beJng in- dicated by the signals. Not only were the stimuli applied at different points, but at B they were of much greater frequency than at A. Although there is a slight change in "local sign," it will be observed that there is no alteration in rhythm, indicating that this property must be a function of the final common path. (From Sherrington.) period lasting nearly a whole second, during which a second stimulus elicits no response. The object of this long refractory period is no doubt that opportunity may be given for the flexor muscles to perform the contraction that would naturally ensue during the normal occurrence of the extensor thrust, as in the act of walking. When the animal places his foot on the ground, the sudden pressure exerted on the pad of the foot immediately calls forth the extensor thrust, by means of PHYSIOLOGICAL PROPERTIES OF THE SIMPLE REFLEX ARC 813 which, the weight of the body is temporarily removed from the ground, and the muscles perform the contractions necessary to produce flexion of the limb. Although the refractory period is unaffected by the strength of the stimulus it is very dependent upon the internal condition of the nerve reflex arc, such as that caused by changes in blood supply or by narcosis. Reflex conduction is much less resistant than nerve conduction to various conditions affecting the nutritive condition of the conducting pathway. For example, deprivation of oxygen causes but slight interference with the conduction along a nerve trunk, but very soon abolishes the spinal reflexes. Even in the frog, reflex movements entirely disappear in thirty to forty-five minutes after the centers have been rendered completely anemic, and in mammals they disappear in a few minutes. In the case of drugs such as chloroform, 0.3 per cent of the drug may be required to abolish conduction in a nerve, whereas a much lower percentage is suffi- cient to abolish it in a reflex arc. Prom the above differences in conduction in a nerve trunk and a re- flex arc, we learn many facts concerning the importance of the latter, and we further see that the differences are due very largely to the synaptic connection. SUCCESSIVE DEGENERATION Before concluding the subject, it may be of interest to consider briefly the method of successive degeneration, by which Sherrington succeeded in demonstrating the exact tracts in the white matter of the spinal cord along which the intraspinal neurons travel from one segment to another. This was worked out in the case of the scratch reflex in the following manner: The spinal cord was first of all cut in the upper thoracic region, so that degeneration occurred in all the descending tracts below the level of the section. In about a year's time these degenerated tracts had entirely disappeared, and the debris of the degenerated fibers had been replaced by cicatricial tissue, so that a section of the cord revealed noth- ing but healthy nervous tissue with cicatrices where the degenerated tracts had existed. When at this stage a second cut was made across the cord a little lower than the first one, further degeneration occurred involving those fibers whose centers were located between the two cuts — that is, the fibers coming from the intraspinal neurons, with the cells of which the afferent nerve fibers coming from the skin of the scratch re- flex area were connected. A section of the cord, stained appropriately for degenerated fibers, at this time demonstrated these fibers to exist in the lateral column of white matter, those that travel a short distance— i. e., between neighboring segments — being near the gray matter, and those traveling greater distances, towards the outside. CHAPTER XCI RECIPROCAL INNERVATION Reciprocal Inhibition. — It might appear that to bend a joint or to move the eyeball the only muscular action required would be contrac- tion of the muscles which flex the joint or rotate the eyeball, and that the antagonistic muscles would merely become passively elongated. When we remember, however, that all the muscles of the body are or- dinarily in a condition of slight contraction, or tone, and that this tends to become increased. when the muscles are passively stretched, then we see that for efficient movement there must be inhibition of the tone of the muscles which oppose those that are contracting. This reciprocal inhibition, as it is called, is a very widespread function throughout the animal body. Sometimes it is purely peripheral in origin, as in the claw of the crayfish, where stimulation of the nerve causes an opening of the claw due to the contraction of one set of muscles and the simultaneous inhibition of their antagonists. Instances of peripheral reciprocal in- hibition in the higher animals are not so common, but are illustrated in the case of the myenteric reflex, where it will be remembered a contraction of the intestine over a bolus of food is accompanied by inhibition in front of the bolus. The reciprocal action in this case is probably dependent on the myenteric plexus. On the other hand, reciprocal inhibition of central origin is very com- mon in the higher mammalia. Thus, in the case of the lateral movement of the eyes, if we cut the third and fourth nerves to one eye, say, the left, the external rectus of that eye will alone be under the control of the nervous system, through the sixth nerve; nevertheless, if we after- ward cause the animal to look toward the right, as by holding some ob- ject in that direction, it will be found that the left eye as well as the right follows the object. Obviously there must be an inhibition of the external rectus muscle of the left eye, an inhibition which is pronounced enough to bring about a movement of the eyeball, and which exactly cor- responds in point of time with the contraction of the external rectus of the right eye. This movement, due to the atonicity of the external rec- tus, does not however succeed in causing the eye to rotate beyond the midline of the field of vision. This is an instance of a willed reciprocal inhibition ; i. e., a reciprocal inhibition brought about by stimuli coming 814 RECIPROCAL INNERVATION 815 from the volitional center in the cerebrum. The same result may be obtained by electric stimulation of the center for eye movements on the cerebral cortex. The most important details concerning the mechanism of reciprocal innervation have been obtained by studying the flexion reflex in a spinal animal which has completely recovered from shock. In such an animal the tonus of the extensor muscles of the knees is well marked. This tonus is maintained by afferent impulses transmitted to the spinal cord from receptors situated in the muscles, and its degree of intensity can be estimated by the briskness of the knee-jerk, which, it will be remem- Fig. 212. — Record from myograph connected with the extensor muscle of the knee. During the time marked by the lower signal, the skin of the opposite foot was stimulated, thus causing the crossed extension reflex. While still maintaining this stimulation, faradic shocks were ap- plied to the skin of the foot of the same side (as indicated by the upper signal), with the result that immediate inhibition of the contracted extensor occurred. (From Sherrington.) bered, is elicited by tapping the patellar tendon, and consists of a sud- den extension movement at the knee joint. By observing the briskness of the knee-jerk we are therefore enabled to form an estimate of the tonicity of the extensor muscles; and if after doing so we throw the flexors which are their antagonists into activity by eliciting the flexion reflex, the knee-jerk will be found much less active. If we prevent the flexors from acting on the knee joint and the leg is held in an extended position, irritation of the skin of the leg will cause the flexion of the- 816 THE CENTRAL NERVOUS SYSTEM disconnected hamstring muscles simultaneously with, a visible relaxation of the extensors (Fig. 212). If the leg is held properly, this relaxation may be marked enough to cause a slight flexion at the joint; and in any case, if the knee-jerk is regularly elicited by equal taps applied to the patellar tendon, it will be found that, while the flexion is being produced, the knee-jerk is very much less than normal, if not entirely absent, thus in- dicating that the tone of the extensor muscles is diminished. This ex- periment is very striking when performed on a decerebrate animal, in which, as we shall see, the extensor muscles of the limb are in a per- manent state of hypertonicity (Fig. 213). Before it is permissible to conclude that this reciprocal inhibition is a necessary event in the movement of a joint, we must however show that it occurs at exactly the same time as the flexion of the antagonist. Sher- rington has succeeded in doing this in a considerable variety of experi- ,Solar ganglion A/) Fig. 230. — Diagram (after I.angley) showing the manner of connection of the fibers compos- ing the great splanchnic nerve. The left-hand diagram represents the usual arrangement, the preganglionic fibers (black) passing through the ganglia of the sympathetic chain and having their cell stations in the solar ganglion, from which the postganglionic fibers (red) then emerge to run to their destination along the blood vessels. The right-hand diagram shows a possible exceptional arrangement. THE AUTONOMIC NERVOUS SYSTEM 879 composed entirely of fibers belonging to voluntary effector neurons, but also of connector fibers of the involuntary system. These ' are distin- guished from the voluntary fibers by being much smaller in diameter; indeed it was by this characteristic that Gaskell succeeded in tracing the course of the involuntary fibers. In brief, therefore, the fibers of the internuncial neurons of the volun- tary nervous system are confined within the central nervous system, where they are contained mainly in the white columns of the spinal cord, the pyramidal tracts, for example, being composed of internuncial fibers from the cerebral neurons; the corresponding fibers of the involuntary nervous system (connector), on the other hand, run out of the- cord with the anterior roots to effector neurons situated either in the ganglia of the sympathetic chain or in peripheral localities. Just as the voluntary internuncial fibers give off many collaterals, so do those of the involun- tary system, so that an impulse transmitted by one internuncial neuron may excite a broad field of effectors. We shall see later that it is through these collaterals that reflex responses can apparently often be excited by the stimulation of the central ends of nerves — such as the hypogastric to the bladder — after all connections with the central nervous system have been severed. (Fig. 230.) To elucidate the further course of the involuntary fibers, and deter- mine the location of the effector neuron nerve cells, it becomes necessary to supplement anatomic with physiologic methods of investigation. The various functions of the innervated parts — vascular changes, muscular movements, glandular activity — are observed by the usual methods of the physiologist, and the nerve roots or nerves believed to contain the involuntary fibers either cut or stimulated. If a change is observed in the functions, it indicates that part at least of the involuntary nerve supply is contained in the nerve structure that has been cut or stimu- lated. Such a result does not, however, inform us as to whether the fiber is that of the connector or effector neuron — whether it is pre- ganglionic or postganglionic. This may be determined in many cases by observing whether nerve degeneration occurs as a result of cutting the fibers, but the most useful method for answering the question is that discovered by Langley by the use of nicotine, which in certain con- centrations specifically paralyzes the synaptic connections between the connector and the effector neurons. If a weak (1 per cent) solution of this alkaloid is painted on a ganglion or peripheral nerve plexus in which the connector neuron finds its effector nerve-cell, it will break the nerve path, so that physiologic responses produced by stimulating the preganglionic fibers become no longer elicitable. When the involun- tary connector fibers run through several ganglia, as in the sympathetic 880 THE CENTRAL NERVOUS SYSTEM chain, it becomes possible, by systematically painting the ganglia with nicotine, to tell exactly in which of them the fiber finds its effector nerve cell. The course and functions of the effector neurons of the three outflows — bulbar, thoracicolumbar, and sacral — vary in many details and must be considered separately. THE THORACICOLUMBAR OUTFLOW, OR SYMPATHETIC SYSTEM PROPER The connector fibers are sharply confined in their outflow from the cord between the first thoracic and the fourth lumbar segments, and they run by the white rami communicantes to the sympathetic chain, where some of them connect with effector nerve cells in its ganglia, while others run beyond the chain to find their effector cells in collateral gan- glia represented by the semilunar, superior and inferior mesenteric and the renal in the abdomen. The fibers of the effector cells, often called postganglionic, are distinguished from the connector or preganglionic fibers by being nonmedullated. Those derived from cells in -the lateral sympathetic ganglia proceed to their destination either by way of the gray rami communicantes to the segmental nerves after the fusion of the anterior and posterior spinal roots, or by the outer walls of the blood vessels. (Fig. 231.) The effector neurons supply the following structures: 1. The blood vessels and heart. 2. The musculature of the sweat glands. 3. The musculature of the hair follicles and other muscles lying under the skin. 4. The musculature of the so-called segmental duct, which is repre- sented in the adult by the uterus, Fallopian tubes, ureters, etc. 5. The sphincters of the intestine. Regarding the innervation of the Hood vessels, the exact situation of the ganglia in which the effector neurons are situated and of the nerve roots which contain the connector fibers, is shown in the accompanying table (page 881). It is clear that the innervation of the blood vessels is practically con- tinuous, the effector neurons being situated both in the lateral and in the collateral chain of ganglia. Those of the former run to the vessels of structures innervated by the cranial and spinal segmental nerves, while those of the latter supply the vessels of the abdominal and pelvic viscera. The connector fibers to the sweat glands are also strictly confined to the thoracicolumbar system, the cell station being found in the ganglion Post, root Ant. root — Preganglfonic fiber '---Sympathetic ganglion rami Postganglionic fiber Fig. 231. — Diagram (after Langley) to show the manner in which a preganglionic fiber, emanating from the spinal nerve by the white ramus communicans, connects in a ganglion of the sympathetic chain with a nerve cell (red), the axon of which then proceeds as the postganglionic fiber (red) by way of the gray ramus communicans back to the spinal nerve, along which it travels to the periphery. It will be observed that the preganglionic fiber does not form its synapsis in the first ganglion it encounters. THE AUTONOMIC NERVOUS SYSTEM 881 stellatum for the fore limb, and the lower lumbar and upper sacral ganglia for the hind limb. When they are stimulated, the muscular fibers surrounding the sweat glands contract and squeeze out the sweat. SITUATION OF BLOOD SITUATION OP MOTOR ROOTS CONTAINING CONNECTOR VESSELS GANGLION CELLS NERVES Head and neck. Superior cervical ganglion. 1, 2, 3, 4, 5, thoracic; 2, 3, 4, give maximum effect. Heart. Ganglion, stellatum and in- .1, 2, 3, 4, 5, thoracic; 2, 3, give ferior cervical ganglion. maximum. Anterior extremity. Ganglion stellatum. 4, 5, 6, 7, 8, 9, thoracic, and 10 slightly. Posterior extremity. 6th lumbar, 7th lumbar, and 11, 12, 13. thoracic; 1, 2, lumbar 1st sacral ganglion. and 3 slightly. Kidney. Eenal ganglion. 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, thoracic ; 1, 2, 3, 4, lumbar. Spleen. Semilunar ganglion. 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, thoracic; 1, 2, 3, lumbar. Abdominal viscera. Superior mesenteric ganglion 6, 7, 8, 9, 10, 11, 13, thoracic ; 1, 2, and semilunar ganglion. 3, lumbar. Pelvic viscera. Inferior mesenterie ganglion. 1, 2, 3, 4, lumbar. (Gaskell) The ganglia for the pilomotor fibers are more widespread (extending from the fourth thoracic to the coccygeal ganglia).; but the connector fibers are again strictly confined to the thoracicolumbar region. Stimu- lation of these fibers causes movement of the hairs, or on hairless animals, the condition called "goose skin." The Motor Nerves of the Muscles Surrounding the Segmental Duct. — It will be observed that the connector fibers to the abdominal and pel- vic viscera are collected into two special nerve trunks, the greater and the lesser (or lumbar) splanchnics. The collateral ganglia (semilunar and superior and inferior mesenteries) with which these connect, have nothing to do with the segmental nerves, but their nerve cells send fibers (postganglionic) which supply the various viscera not only with vaso- motor fibers but also with the "sympathetic" fibers, which we have seen exercise such an important control over their glandular and muscular functions. All of the fibers contained in the lumbar splanchnics do not, however, have their cell stations in the inferior mesenteric ganglia, but run through it and proceed in the hypogastric nerves to find their effector cells on the musculature of the various structures that are developed from the Miillerian and Wolffian ducts — i. e., of the ureters, uterus, Fal- 882 THE CENTRAL NERVOUS SYSTEM lopian tubes and vas deferens. Many of the fibers of the hypogastric nerves are therefore those of involuntary internuncial neurons. The ileocolic and internal anal sphincter muscles of the intestines and internal vesical sphincter receive their nerve supply from effector neurons situated in the superior and inferior mesenteric ganglia, the internuncial fibers arising from the thoracicolumbar region. It is possible that the other sphincters of the intestinal canal — viz., the cardiac and pyloric sphincters of the stomach — are similarly innervated. (Fig. 232.) Great aid in working out these nerve connections is received by study- ing the effect of epinephrine, which acts specifically on those tissues that are supplied by the sympathetic nervous system.* Epinephrine has no effect on tissues innervated by the bulbar or sacral outflows, and it develops its action peripherally, being indeed more potent on a dener- vated organ even after all its nerves have been allowed to degenerate. Advantage of this action of epinephrine has been taken in the investi- gation of doubtful cases of sympathetic innervation, such as in the cere- bral, coronary, and pulmonary blood vessels. The outcome of these in- vestigations has been discussed elsewhere. THE BULBOSACRAL OUTFLOW OR THE PARASYMPATHETIC SYSTEM From the medulla oblongata arise involuntary connector neurons, which are carried mainly by the vagus nerves but partly by the seventh, ninth and eleventh cranial nerves to effector nerve cells situated periph- erally on the structures to which the nerves run (Fig. 233). These include in a general way the muscles and glands of the alimentary canal and its derivatives as far as the end of the small intestine. In the small intestine itself the cells of these motor neurons are those of Auerbach's plexus found between the two muscular coats. In the diverticula, which include the lungs and the gall bladder, the nerve cells to which the vagus fibers run are also situated peripherally. The sacral outflow occurs through the second and third sacral nerves, the fibers joining to form a single nerve (the pelvic nerve or nervus erigens) on each side. This passes directly to the bladder, where it connects with a plexus, often called the hypogastric, which extends over the bladder and neighboring portion of the rectum. The branches run to connect either with the nerve cells of the ganglia of the plexus itself, or with nerve cells situated on the walls of the large intestine and blad- der. The pelvic nerve makes its connections with the periphery in the •Its action is always the same as that which is produced by stimulation of the sympathetic nerve supply, whether this effect is one ot stimulation or inhibition. Splanchnic (Poafgan Head e NecK (Postqangij Vagus— -Medulla ===Key — PreGangiionic Sympathetic PostGanglionic Sympathetic Pre Ganglionic Bulbo-Sacral (Para Sympathetic) PostGanglionicBulbo-Sacral Arm 1 (Post Heads- Neck (Pregan'gj Sympathetic Ganglia Jhoracic Splanchnic -v nerve -Coellac Plexus & ^-\ 5gp.Mes.Gang. .Leg (PreyangJ Arm . (Pregang) Lumbar .Splanchnic nerve Abdominal Viscera (Precjang.) -Inf. Mes.Gang. .Leg (Post gang.) illeck. Pelvic visceral nerve Fig. 232. — Diagram showing the main parts of the autonomic nervous system to be used along with Fig. 233. For the sake of clarity several of the preganglionic fibers of the sympathetic autonomies are omitted, but the position of their egress from the cord is indicated in the side notes. The diagram shows clearly the distribution of the bulbosacral autonomic system by way of the vagus and the first, second and third sacral nerves. THE AUTONOMIC NERVOUS SYSTEM 883 same manner as the vagus. Taken together these two nerves supply the musculature of the gastrointestinal tract, including the cloaca, the vagus as far as the end of the small intestine, and the pelvic nerve from this point on. It must of course be remembered that certain muscles— namely, the sphincters of the small and large intestine — receive their nerve supply from the sympathetic (page 882). Just as structures innervated by the sympathetic are peculiarly susceptible to the action of epinephrine, it has been discovered that those innervated by the bulbosacral system are very susceptible to the action of acetyl- choline, which is present in ergot. They are not acted on by epinephrine, nor are the structures upon which this acts affected by acetyl-choline. AXON REFLEXES At this place it is convenient to consider for a moment the phenome- non which has already been referred to as an axon reflex. It was dis- covered that when one of the hypogastric nerves was cut and the central end stimulated there was a reflex contraction of the bladder and the in- ternal anal sphincter, along with vasoconstriction in the region of the rec- tum and that this occurred, even after disconnecting the inferior mesen- teric ganglion from the spinal cord by cutting the lumbar splanchnic nerves. Injection of nicotine immediately abolished the response. It looked as if reflex action was possible through the ganglion; which would justify the name "sympathetic" originally given to the involuntary nervous sys- tem in the belief that the ganglia were centers for local reflex actions. Further investigation showed, however, that this reflex is not similar to those occurring in the voluntary system, but is dependent upon the presence of a collateral on internuncial fibers that run through the in- ferior mesenteric ganglia to nerve cells situated peripherally on the walls of the bladder and rectum. The collaterals terminate by synapsis around nerve cells in the ganglion, the axons of which, as we have seen, run to the bladder, the rectal blood vessels, and the internal sphincter ani. The evidence for this explanation depends on the observation that the axon reflex is no longer possible after the lumbar splanchnics have been cut and time allowed for their fibers to become completely degen- erated. Similar reflexes depending on collaterals have been found in the lateral chain, and there can be little doubt that they are of frequent occurrence throughout the whole involuntary nervous system, just as they are, within the spinal cord, in the voluntary. It is because of these collaterals and the fact that nerve fibers transmit impulses in both directions that a stimulus transmitted through one or a limited number of connector 884 THE CENTRAL NERVOUS SYSTEM neurons may excite a broad field of effectors and cause a widespread effect. FUNCTIONS OF AUTONOMIC NERVES The functions of the autonomic nerve fibers have been discussed in connection with the structures which they supply, and we shall require in this place only to review them in a general way. Two opposed effects may be obtained: stimulatory (augmentory) and inhibitory; and these may be produced through one nerve by its being stimulatory for one set of muscle fibers and inhibitory for another set in the same viscus. The branches running from the inferior mesenteric ganglion to the colon,, for example, are augmentory (constrictor) for the blood vessels and inhibitory for the muscular walls of the colon. The greatest interest centers on the inhibitory impulses. They are best known in connection with the vagus nerve to the heart, the sympa- thetic to the small intestine, and the hypogastric to the musculature of the bladder. It is interesting to compare the nature of inhibition in the involuntary and voluntary systems. In the latter, it will be remembered, inhibition can occur only through the internuncial neurons and the ef- fector nerve cell, stimulation of the effector nerve fiber never having any other than an augmentor effect. It is quite otherwise in the involun- tary nervous system, for stimulation of the effector nerve fiber, after complete destruction of the effector nerve cell, -is still followed by a typical inhibition. This, it will be remembered, may be demonstrated on the frog heart by applying electric stimulation to the white crescentic line after paralyzing the effector nerve cells by nicotine. The same may also be shown in the case of the chorda tympani, where stimulation of the post- ganglionic fibers in the hilus of the gland causes dilatation of the blood vessels after paralysis of the ganglion by nicotine, vasodilatation being of course a phenomenon of inhibition. It is a difficult matter to designate precisely' Which fibers in any part of the involuntary nervous system are inhibitory and which augmentory. Indeed, as mentioned above, one fiber may perform both functions. In cases where the existence of inhibitory fibers is doubtful, great aid is afforded by the use of ergotoxine, an alkaloid of ergot, which possesses the remarkable property of specifically paralyzing the augmentor nerves of the sympathetic system (but not of .the parasympathetic) ; that is, the same fibers as are stimulated by epinephrine. When, a particular structure is supplied with augmentory and inhibitory fibers by a com- bined sympathetic nerve, electric stimulation or the application of epi- nephrine usually gives only augmentory effects; after the injection of Lac/iryma/q/and Nasal mucosa — .;y-:— - Sublingual /f gland I '/, W Submaxillary gland SubmaxillarytSublingbal) ganglion Iliocecal sphincter* B/a Ve5/ca/ sphincter-Jr Urethral sphincter//; <-r". ^= Cran/a/ and 5acra/ nervej mofor = red inhibitory = Wue Thoraclco-lumbar nerves motor =.ye//ow inhibitory*qreeh 'Postganglion'tc fibers are dotted, thus — N.XI Jup. cervical ganglion Thyroid gland Inf.cervical ganglion Ansa subclavia •Stellate (/L* Thoracic) ganglion fibers Pilo motor muscle Celiac (Semilunar) ganglion(Solarplex) Splanchnic nerves Sympathetic chain .1 [ \~Lumbarsplanchnics Sup. Mesenteric ganglion Inf. ttesenteric ganglion hypogastric nerves T> P HailecK PelvidHypogastricjnterilidc) < sphincter \ ^ plexus. (YesicalSt rectal portions) Pelvic nerves (Nervus erigens) Fig. 233. — Schematic representation of the involuntary nervous system. (From Jackson.) THE AUTONOMTO NERVOUS SYSTEM 885 ergotoxine, however, a reversed effect is observed; namely, inhibition instead of augmentation. By taking advantage of this fact, Dale has been able to demonstrate in the hypogastric nerves inhibitory fibers to the uterus, and Elliott has demonstrated the inhibitory action of epi- nephrine on the muscles of the ureter in the dog. Inhibitory 'fibers have also been discovered by these methods in the great splanchnic nerves, in the nerve roots supplying the kidney, and in the cervical sympathetic supplying the blood vessels of the mucous membrane of the mouth, etc.; that is, in nerve trunks which previously were believed to contain only augmentory fibers. The accompanying diagram from Jackson will give an idea of the currently accepted views concerning the distribution of augmentory and inhibitory fibers. (Fig. 233.) THE AFFERENT FIBERS OF THE AUTONOMIC SYSTEM It has long been known that the exposed viscera are remarkably insen- sitive. This experience is in accord with the observation that the supply of afferent fibers to the viscera is relatively very small. In the hypo- gastric and probably in the great splanchnic nerve, Langley computes that only about one-tenth of the medulla ted fibers are afferent. At the two ends of the alimentary canal, where cooperative reflexes between the somatic musculature and the viscera are of importance, a greater number of afferent fibers are found in the autonomic nerves; for ex- ample, in the pelvic nerve about one-third of the fibers are afferent, and, as we have frequently seen, the vagi contain large numbers of them coming from the lungs, stomach, and no doubt from other viscera. The afferent visceral fibers, as above stated, arise like those of the voluntary system, from the ganglion cells of the posterior roots. They travel in company with the connector fibers through the white ramus communicans, so that the stimulation of the central end of one of these may cause reflex rise in blood pressure and other movements. It is found that, after opening the abdominal cavity under local anesthesia, cutting and suturing of the viscera may be continued without causing any pain. When the viscera are inflamed, however, and under certain conditions of stimulation, such as the distention of the bile ducts with biliary calculi, or the violent contraction of the intestines, excruci- ating pain may be evoked. This pain is frequently not localized hi the viscera, but is referred to certain parts of the surface of the body, and it has been shown by Mackenzie and by Head that it is referred to the area of skin which is supplied with sensory nerves by the same segment as that to which the afferent autonomic fibers of the particular viscus run. It has further been shown that vascular disease may cause sensi- 886 THE CENTRAL NERVOUS SYSTEM tivity of the corresponding cutaneous areas, so that clinical methods are available for localizing the site of the disease by studying the exact position and extent of the referred pain or skin tenderness. NERVOUS SYSTEM REFERENCES (Monographs and Original Papers) iParker, G. H.: Proc. Am. Philos. Soc., 1911, i, 217-225. ^Head, H., and Rivers, W. H. R.: Brain, 1908, xxxi, 323-450. sMeek, W. J. : Am. Jour. Physiol., 1911, xxviii, 356-360. *Bruce, A. Ninian: Arch. f. exper. Path. u. Phannakol., 1910, Ixiii, 426-433. 4aSherrington, C. S. : Numerous papers on reciprocal innervation of antagonistic muscles, Proc. Roy. Soc., Vol. B, 66; also in Jour. Physiol., xxii, xxxiv, xxxviii, xliii, and Quart. Jour. Exper. Physiol., ii. sHolmes, Gordon: Brit. Med. Jour., 1915, ii, Nov. 27, Dec. 4 and 11. ePike, F. H. : Am. Jour. Physiol., 1909, xxiv, 124-152. 7Jolly, W. A.: Quart. Jour. Exper. Physiol., 1910, iv, 67-87. sLombard, W. P.: Jour. Physiol., 1889, x, 122-148. sCollier, J.: Lancet, April 1, 1916, 711. icRanson, S. W., and von Hess, C. L. : Am. Jour. Physiol., 1915, xxxviii, 128. "Head, H., and Thompson: Brain, 1906, xxix, 537. i2Sherrington, C. S., and Brown, T. Graham: Jour. Physiol., 1913, xlvi, Proc. Physiol. Soc., p. xxii. isBrown, T. Graham, and Sherrington, C. S.: Proc. Roy. Soc., 1912, 85, B, 250-277. "Gushing, Harvey: Proc. Am. Physiol. Soc., Am. Jour. Physiol., 1909. isLuciani, L.: Kleinhirn, Ergebnisse der Physiol., 1904, I. isBlack, Davidson: Cerebellar Localization in the Light of Recent Research, Jour. Lab. and Clin. Med., 1916, i, 467. irGaskell, W. H. : The Involuntary Nervous System, Monographs on Physiology, ed. by E. H. Starling, Longmans, Green & Co., 1916. Other Monographs not Specifically Referred to in the Context • 18Sherrington, C. S. : (1) The Integrative Action of the Nervous System, Silliman Lec- tures, Yale University. Scribner's Sons, New York. (2) Shafer's Textbook of Physiology, II. Young J. Pentland, London, 1899. J. S.: Recent Researches on Cortical Localization and on The Function of the Cerebrum in Further Advances in Physiology, ed. by Leonard Hill, London, E. Arnold, 1909. INDEX Abdominal respiration, 307 Abnormal pulses, 276 Absorption, in general, 13 from stomach, 456 of fats, 691 Acapnia, 292 Accessory food factors, 584 Acetaldehyde, 708 Acetoacetic acid, 683, 709 Acetone, 683, 709 Acid: buffer action, 36 excretion of, by kidneys, 46 number of fats, 687 total concentration of, 32 Acidity, actual degree of, 23 Acidosis : ammonia-urea ratio during, 616 compensated, 39 in diabetes, 683 in nephritis, 683 in starvation, 569 relationship to alveolar CO2, 354 relationship to breathing, 354 theory of, 38 uncompensated, 39 Acids, of urine, 524 Actual degree of acidity and alkalinity, 23 Adenine, 635 Adenosine, 638 Adjusters, 783 Adrenal glands and diabetes, 673 Adrenaline (see Epinephrine) Adsorption, 65 compounds, 70 conditions influenced by, 67 effect of chemical forces on, 68 effect of electric changes on, 67 everyday reactions depending on, 66 of gases, 66 Afferent fibers of autonomic system, 885 Afferent spinal pathways, 830 Age, 584 effect on creatinine excretion, 624 Alanine, 600, 603, 606, 649, 666 Albolene absorption, 692 Albuminuria, 519 Alkali retention, determination of, 48 Alkaline buffer, 36 Alkaline reserve, 38 measurement of, 41 Allantoin, 636, 639, 645 Allied reflexes, simultaneous integration of, 823 successive integration of, 823 Alloxan, 635 Alveolar air: clinical investigation of, 347 estimation of gases in, 344 Fridericia method, 340 Haldane method, 340 Pearce method, 345 tension of CO,, 46, 339, 356 during breathing in confined space, 357 tension of oxygen, 339 Ambard's equation, 527 in acid excretion, 48 Amboceptor, 96 Amino acids, 597 and energy output, 541 in blood, 606 chemistry of, 598 determination of, 599 fate of, 610 groups, 598 in growth, 576 in tissues, 607 in urine, 530, 620 structure of, 602 > 603 Aminoacetic acid (see Glycocoll) Aminopropionic acid (see Alanine) Ammonia : ammonia-urea ratio: influence of acidosis on, 616 in disease, 620 influence of liver on, 617 as reserve alkali, 616 excretion of, 615 excretion of acid in combination with, 46 of urine, 530 Ammonium carbamate, 616 Ammonium carbonate, 616 Amoeba, 782 Amylases, 81, 90, 491 Amylolysis, 491 in stomach, 454 Amylopsin, 491, 656 Anacrotie wave, pulse, 203 Analysis (psychic), 858 Anaphylactie reaction, 595, 601 Anaphylaxis, 89 Anarthria, 862 Anastomosis, intestinal, 470 887 SS8 INDEX Anemia, 93 bloodflow in, 283 Anesthesia, 831 Aneurism, bloodflow in, 284 pulse in, 143, 200 Angina pectoris, fibrillation in, 196 Animal calorimeter, 536 Anions, 16, 59 Antagonistic muscles, 818 Antagonistic reflexes, 824 Anterior roots, 787 Anticoagulants, 99 Antidromie impulses, 234 Antiferments in blood, 89 Antithrombin, 104, 112 Antitoxins, 69 Antitrypsin, 90 Aortic regurgitation, pulse in, 131 Apesthesia, 838, 851 Apex beat, tracing of, 275 Aphasia, motor, 860, 862 sensory, 862 subcortical, 862 Apnea, nervous element in, 332, 362, 365 Apparatus for measuring respiratory ex- change, 554 Appetite juice, nature of, 440 Arc, reflex, 784 Arginase, 81, 616 Arginine, 605, 616, 627 Aromatic sulphates, 632 Arrhythmia of sinus, 266, 277 Arterial pressure, 122 Arteries, bloodflow in, 198 Arteriosclerosis, diastolic pressure in, 143 Aspartic acid, 605, 666 Asphyxia, 311 Assimilation limit, 652 Association areas, cerebral, 852, 861 neurons, 783, 785 Astasia, cerebellar, 8C7 Asthenia, 867 Asthma, dead space in, 311 Ataxy, cerebellar, 866 Atonia, cerebellar, 867 Atophan, 651 Atropine, effect on glands, 422 Auditory center, 851 Auricle, pressure in, 148 propagation of beat in, 191 Auricular curve, contour of, 153 Auricular fibrillation, 196, 269, 280 Auricular flutter, 196, 269, 279 Auriculoventricular orifice, 148 bundle, 183 node, 183 Ausculatory method (of blood pressure), 130 Autocatalysis, 77 Autonomic nerves, cerebral, 423 sympathetic, 423 Autonomic nervous system, 877 afferent fibers of, 885 Autonomic nervous system — font M axon reflexes in, 883 bulbosacral outflow, 882 connector fibers of, 878 functions of, 884 general plan of construction, 878 parasympathetic, 882 thoracicolumbar outflow, 880 internal vesical sphincter, 882 Axon, 784 reflexes, 797, 883 Azelaic acid, 712 Bacillus coli communis, 500 Bacteria, in intestine, 499, 657 in stomach, 482 Bacterial digestion, 499 Balance, energy, 535 . material, 543 sheet of body, 543 Banting cure, 571 Basal heat production, 538 Basal ration, 5.76 Basophile cells, 96 "Bends" in caisson workers, 402 Benzoic acid, 630, 710 Benzoyl chloride, 631 Beriberi, 584 Beta-hydroxy butyric acid, 709 Bile, 442 and fat digestion, 690 chemistry of, 494 constituents of, 492 from gall bladder, 492 functions of, 493 pigments of, 495 salts, 494 Bilirubin, 495 Biliverdin, 495 B-imidazolylethylamine, effect on blood vessels, 397 Birds, removal of liver from, 618 Blood : absorption into, 13 amino acids in, 606 amount in body, 135 antiferments of, 89 circulation of, 122 dissociation curve of, 383 fat of, estimation, 696, 697 variations in, 697 ferments of, 89 gases of, transportation, 379 general properties of, 85 mass movement of, 281 means by which gases are carried, 390 oxidation in, 396 proteases of, 89 proteins of, 87 origin, 88 INDEX SS9 Blood— Cont'd quantity of, in body, 85 refractive index of, 88 specific gravity of, 86 sugar level of, 657 regulation, 671 transfusion of, 93, 135, 139 viscosity of, 140 volume of, 136 water content of, 86 Blood cell, red, fate of, 93 origin of, 92 regeneration of, 93 stroma of, 91 white, 96 Blood clotting, 98 in diseases, 111 in physiologic conditions, 110 influence of calcium on, 103 influence of tissues on, 104 intravascular, 107 methods of retarding, in drawn blood, • 99 negative phase of, 108 theories of, 106 time of, 100, 108 visible changes during, 98 Blood corpuscles in mountain sickness, 401 Bloodflow : clinical conditions affecting anemia, 283 cardiovascular diseases, 284 fever, 284 diseases of nervous system, 285 mass movement of, 208 movement in veins, 214 variations in, 282 velocity of, 206 visceral, 212 Blood gas manometer, 381 Blood platelets, 97 Blood pressure, 122 diastolic, 127 effect of hemorrhage on, 135 effect of pleural pressure on, 306 factors maintaining, 134 H-ion of blood on, 237 mean arterial, 123 in shock, 290 systolic, 127 tracing, 125 Blood vessels, 880 elasticity of, 142 tone of, 236 Body fluids, reactions of, 35 Body weight and energy production, 539 Botulism, 503 Bowman, capsule of, 507 Bradycardia, 193 Brain : circulation in, 247 vasomotor nerves, 252 volume of,' 250 Breathing, in compressed air, 399 in rarefied air, 360 periodic, 363, 371, 376 Brownian movement, colloids, 57 Bruits, 158 Buffer action of blood, 374 Buffer substances, 36 Building stones of protein, 597 Bulbosacral outflow, 882 Butyric acid, 709 C Cadaverine, 629 Caffeine, 635 Caisson disease, 402 cause of, 403 decompression of workers, 406 prevention, 404 symptoms, 402 working conditions in, 408 Calcium ion, influence on clotting, 103 influence on heart, 166 Calcium rigor, 166 Calomel electrode, 30 Calorie, 535 Calorimeter, 535 animal, 536 Benedict, 537 bomb, 537 hand, 281 respiration, 536 Eussel-Sage, 537 Calorimetry, direct, 546 indirect, 546, 554 Canals, semicircular, 873 removal of, 874 Cannabin, 577 ' Capillary analysis of colloids, 56 Carbamino reaction, 599 Carbohydrates, absorption of, 657 assimilation limits, 652 digestion of, 656 and growth, 583 metabolism of, 652 production from protein, 665 saturation limit, 652 Carbon balance, 547 Carbon dioxide, combining power, 42 effect on respiratory center, 352 estimation in blood, 390 output, 550 volume percentage in blood, 391 Carbon dioxide tension, 337 in alveolar air, after exercise, 367 estimation of, 339, 344 in mountain sickness, 361 in periodic breathing, 375 in arterial blood, 337 in venous blood, 342 Carbonic acid (see Carbon dioxide) Carboxyl group, 598 Cardiac decompensation, 311 890 INDEX Cardiac depressor nerve, 239 Cardiac muscle, physiologic characteristics of, 176 Cardiac pouch (stomach), 453 Cardiac sphincter, 448 Cardiorenal disease, bloodflow in, 284 energy output in, 542 Cardiograms, 275 Cardiovascular disease, bloodflow in, 284 Casein, 488, 576 Caseinogen, 488 Catalase, 90 Cations, 16 Catalysts, 72 Catalytic power, 23 Celenterates, nervous system of, 782 Cellulose, digestion of, 500 Centers : association, 852, 855 diabetic, 672 motor, 843 sense, auditory, 851 visual, 851 sensory, 850 word centers, 862 Cephalin, 689 Cereals and growth, 581 Cerebellar ataxy, 866 Cerebellum : ablation of, 869 clinical observations, 870 extirpation of, 869 functions of, 865 lobes of, 868 localization of function of, 867 Cerebral circulation, 247 Cerebral compression, 253 Cerebral cortex, stimulation of, 844 structure of, 852 Cerebral localization, 843 clinical observations, 849 hemispheres, removal of, 840 Cerebral vessels, ligation of, 247 Cerebrospinal fluid, 248 Cerebrum, higher functions of, 860 CH, method of expressing, 27 Cheyne-Stokes breathing, 371, 377 Chlorides, urine, 513 Cholesterol, 494, 688 estimation of, 697 Choline, 689 Chorda tympani, 231, 396, 423 Chromatolysis, 801 Chromatine, 638 Chromosones, 638 Chyme, 456, 482 Circle of Willis, 247 Circulation of blood: control of, 216 influence of gravity on, 244 mass movement of blood, 208 through the heart, 257 Circulation of blood — Cont'd through the liver, 255 through the lungs, 253 time of, 213 Circulation time, 206 Clinical application, circulation, 259 nervous system, 828, 849, 862 respiration, 310, 399 Clotting of blood (see Blood clotting) Coagulative ferments, 82 Cod-liver oil, nutritive value, 706 Coefficient of oxidation, 393 Coefficient of solubility of gases, 337 Cold spots, 792 Collaterals, 784 Colloids : Brownian movement, 57 capillary analysis, 56 characteristic properties of, 50 diffusibility of, 51 dispersion means, 54 dispersoid, 54 electric properties of, 55 osmotic pressure, 57 electrophoresis, 56 external phase, 54 gelatinization, 61 heterogeneous, 51 homogeneous, 51 imbibition, 62 internal phase, 54 isoelectrie point, 64 lyophobe, 60 mutual precipitation of colloids, 56 osmotic pressure of, 141 size of colloid particles, 53 suspensions, 53 suspensoids and. emulsoids, action of electrolytes on, 63 Tyndall phenomenon, 51 Compensated acidoses, 39 Complemental air, 300 Compressed air sickness, 399 cause of symptoms, 403 prevention of, 404 treatment of, 406 Concentration cell, 30 Concentration point, auricles, 185 Conception, 861 Concept, 861 Conditioned reflexes, 481, 856 Conductivity, determination of, 17 equivalent, 19 molecular, 19 specific, 17 Conductivity cell, 18 Conglutin, 538 Construction of autonomie nervous sys- tem, 877 Contracture, extension, 806 Cooking, 593 Coronary circulation, 257 Coronary vessels, vasomotor nerves, 268 INDEX 891 Corpora quadrigemina, 840 section behind, 840 section in front of, 840 Corpuscles of blood, red, 91 white, 96 Cortex, removal of, 843 Coughing, 300, 412 Cranial cavity, pressure in, 251 Creatine, 606, 613 chemistry of, 622 estimation of, 623 in disease, 626 metabolism of, 624 origin of, 626 Creatinine, 613 chemistry of, 622 coefficient, 624 estimation, 623 in urine, 529 metabolism, 624 of blood in disease, 651 origin of, 626 Crista acustica, 873 Critical concentration, 8 Crossed extension reflex, 804 Cuorin, 689 Current of action of heart, 187 Cyanosis, 360, 400 Cysteine, 603 Cystine, 577, 592, 604 Cystosine, 637 Cytases, 463 D Dalmatian dog, purine metabolism of, 646 Dalton's law, 336 Dead space, 302, 310 Deafness, 864 Deamidization, deaminization, 501 Deaminizing enzyne, 639 Decerebrate rigidity, 808 Decerebration, 843 Decolorization of liquids by charcoal, 66 Decompression, 406 Defecation, 470 blood pressure during, 412 Defibrinated blood, 101 Degeneration, successive, 813 Deglutition, 445 Delayed conduction, 270, 276 Delirium cordis, 195 Dendrites, 784 Depression of freezing point, 10 of urine, 523 Depressor nerve, 238, 239, 240 Depressor substances, 397 Dessert, physiologic value of, 437 Detoxication compounds, 629 Detoxication process, 501 Dextrins, 491, 656 Dextrose (see Glucose) Diabetes : acidosis in, 684 and the ductless glands, 678 assimilation limits in, 652 blood examination in, 659 blood fat in, 699 center, diabetic, 672 early diagnoses of, 652 energy output in, 542 experimental, 672 fat metabolism in, 683 ketosis, 683 pancreatic, 678 nervous, in man, 674 permanent, 676 phlorhizin, 665 postprandial hyperglycemia, 659 renal, 661 starvation treatment in, 684 treatment of, 653 Diabetic acidosis, 684 Diabetic center, 672 Diabetic gangrene, 258 Dialuric acid, 645 Dialysate, 52 Dialysis, 12 method, colloids, 51 Diaphragm, action of, 320, 321 physiology of, 324 Diastolic filling of heart, 153 Diastolic pressure, 127, 132 measurement of, in man, 128 Dicrotic notch, 202 wave, 203 Diet at different ages, 590 of different communities, 589 Dietetics, 588 Differential manometer, 381 Diffusion, 12 Digestibility of foods, 593 Digestion, by pancreatic juice, 489 in intestine, 489 in stomach, 481 mechanism of, 444 Digestive glands: control of, hormone, 425 nervous, 423 general physiology of, 418 microscopic changes during activity, 418 Dispersion medium, colloids, 54 Dispersoid, colloids, 54 Dissociation, 16, 17 JMssociation constant, 19, 388 Dissociation curve: of blood, 383 of hemoglobin, 383 influence of salts on, 385 influence of H-ion concentration on, 386 influence of temperature on, 386 Dissociation hypothesis, applications of, 21 Dissociation, rate of, 380 892 INDEX Diuresis, 578 Diuretics, 578 Diver 's palsy, 402 Douglas method, 544, 558 Dropped beat, 271 Dtt Bois formula, 541 Ductless glands, 729 in diabetes, 678 Dyspnea, 314, 349 Dystrophy, isolation, 808 E Earth worm, nervous system of, 783 Eck fistula, 617 Eclampsia, 620 Edema, 62, 120 Edestin and growth, 577 Effectors, 782 Elastin, digestion of, 486 Electric conductivity, 16 Electric currents, development of, 29 Electric properties of colloids, 55 Electrocardiograms, 158, 259 normal, 261 standardization of, 260 ventricular complex, 262 waves of, 261 P-wave, 189, 261 T-wave, 220, 263 Electrocardiograph, 260 Electrocution, cause of death in, 195 Electrolytes, 16 action of, on colloids, 63 Electrolytic solution pressure, 29 Electrophoresis of colloids, 56 Electrostatic attraction, 29 Emboli, 107 Emetics, 450 Emotional glycosuria, 675 Emphysema, 311, 314, 324 Empyema, 324 Emulsions, 688 Emulsoids, colloids, 60 Endocrine organs, 729 • Endoenzyme, 71 Endogenous metabolism, 615 of purines, 647 Energy balance, 535 Energy output, and age, 541 and body weight, 549 and disease, 542 and muscular work, 549 and sex, 541 and surface area, 540 in starvation, 568 Enterokinase, 443, 489 Enzymes, 71 action of temperature on, 74 amylases, 81 and catalysis, 72 antienzymes, 81 arginase, 81 E nzymcs — Cont 'd coagulative ferments, 82 conditions of activity, 82 endoenzymes, 71 glyoxylase, 82 invertases, 81 lipases, 81 nature of, 72 oxidases, 82 peculiarities of, 80 peroxidases, 82 properties of, 73 proteases, 80 reversibility of action of, 25, 77 specific action of, 73 types of, 79 urease,' 82 velocity constant, 74 Epieritie receptors, 790 Epilepsy, Jacksonian, 849 Epinephrine, 236, 502 and diabetes, 673 Equilibrium, nitrogen, 571 Equivalent, conductivity, 19 Erepsin, 490, 601 Ergastoplasm, 420 Ergot, 502 Ergotoxine, 209 Erythroeytes, 91 fate of, 94 regeneration of, 93 Escapement, 218 Esophagus, during swallowing, 446 inhibition of, 447 peristaltic wave in, 447 Esters, 686 Ester value, 687 Ethereal sulphates, 501, 632 Excelsin, 577 Exogenous metabolism, 615 Exophthalmic goiter, 756 energy output in, 542 Excretion of acid combined with ammonia, 46 Excretion of urine, 507 Extension contracture, 45, 806 Extensor thrust, 57 reflex, 805 Exteroceptors, 788, 822 Extrasystole, 266 Eyes, movements of, 847 F Factor safety, in diet, 592 Fatigue of reflexes, 825 Fats: absorption of, 691 chemical theory, 693 mechanistic theory, 692 and growth, 583 blood, 696, 697 destination of, 699 INDEX 893 Fats, blood— Cont 'd determination, 696 during absorption, 698 during fasting, 698 variations in, 697 chemistry of, 686 depot fat, 699, 700 destination of, 701 desaturation of, 705, 712 digestion of, 690 fat dust, 696 liver fat, 699, 701 metabolism of, 696 tissue fat, 699, 706 transportation to liver, 702 Fatty acids, 686 acid number, 687 breakdown of, 709 ester value, 687 formation from carbohydrates, 701, 707 • in liver in disease, 703 iodine value, 688 melting point, 687 Eeichert-Meissl value, 688 saponification value, 687 Feces, 499, 521 Ferments (see Enzymes) Ferments in blood, 89 Fever, bloodflow in, 284 cold-bath treatment, 284 purine excretion during, 648 Fibers, anterior root, 100 connector, 193 internuncial, 802 Fibrillation, auricular, 196, 269 ventricular, 195 Fibrin, 99 fibrin needles, 99 source of, 101 Fibrin ferment (see Thrombin), 102 Fibrinogen, 87, 101, 103, 111 Filtration, 13 Final common path, 787, 821, 824 Fistula, biliary, 492 gastric, 434 salivary, 430 Flexion-reflex, 804, 821 Flutter, auricular, 269 Food: accessory factors of, 593 cooking, importance of, 593 effect of, on circulation, 243 effect on creatinine excretion, 624 laxative qualities, 594 palatability, 593 Food factors, accessory, 584 Foodstuffs, rate of leaving stomach, 458 Forced breathing, 324 Formaldehyde titration, amino acids, 487 Formation of solid surface films, 66 Freezing point constant, 10 Freezing point, depression of, 10 Fridericia's method for alveolar air, 340 Frontal visual center, 851 Fructose, 666 Functions of autonomic nerves, 884 Fundus of stomach, 452 G Gallstones, 494 Galvanometer, string, 187, 259 Ganglia, 784 Gas in stomach, 462 Gas laws, 3, 336 Gases, adsorption of, 66 coefficient of solubility, 337 estimation of, 344 partial pressure of, 336 solution of, 336 tension of, 336 transportation in blood, 390 Gaskell's clamp, 175 Gastric contents, regurgitation of, 449 Gastric digestion, 481 rate of, 487 Gastric fistula, 434 Gastric juice, quantity secreted, 440 strength of, 441 Gastric secretion, 432 hormone control of, 437 local stimulation of, 438 nervous control of, 434 Gastric tube, 453 Gastrin, 439, 456 Gastroenterostomy, 460 Gastrointestinal contents, reaction of, 505 Gelatin, 578 Gelatinization, 61 Glands, changes during activity, 422 electric changes, 422 normal conditions of activity, 430 oxygen consumption of, 396, 421 respiration of, 396 Globulin, 577 Gliadin, 576 Glomerulus, 507 Gluconeogenesis, 662, 677, 680 direct method, 663 indirect method, 664 in normal animals, 667 Glucose, 708 fate of absorbed, 662 glucose to nitrogen ratio, 664 injections, intravenous, 653 subcutaneous, 656 parenteral assimilation, 656 tolerance for, 653 utilization of, in tissues, 677 Glutarnic acid (see Glutaminic acid) Glutaminie acid, 605, 667 Glutein, 577 Glutelin, 577 Glyceric aldehyde, 665 Glycerol, 665 894 INDEX Glycocholic acid, 494, 631 Glyeine, 494, 603 Glycinin, 577 Glycocoll, 601, 630, 667, 710 Glyeogen, 662 fate of, 669 sources of, 662 Glycogenase, 662 Glycogenolysis, 669 hormone, 676 nervous, 672 postmortem, 670 Glycolaldehyde, 665 Glycolysis, 677 Glyconeogenesis (see Gluconeogenesis) Glycosuria, alimentary, 659 emotional, 675 postprandial, 659 relation to sugar of blood, 660 renal, 661 Glycuronates, 630 Glycuronic acid, 630, 631, 632 Glyoxal, 631 Glyoxylase, 82, 666 Glyoxylic acid, 631 G-N-ratio, 664 Goiter, exophthalmic, 542 Gout, 648, 650 etiology of, 650 guanine, 640 uric acid excretion in, 648 Grading of intensity of reflex action, 809 Gram molecule, 3, 5 Gram molecular solution, 22 Gravity, on circulation, 244 compensation for, 245 Growth, 574 accessory factors, 585 basal ration, 576 carbohydrates and, 583 curves of, 576 curves of inhibition, 579 fats and, 583 inorganic salts and, 586 lysine and, 578 proteins and, 574 trypanophane and, 578 vitamines, 584 Guanidine, 605, 622 Guanine, 635 gout, 640 Guanosine, 638, 639 Giinsberg reagent, 487 H Haldane-Barcroft apparatus, 45 Haldane gas apparatus, 559 Haldane's method for alveolar air, 340 Heart : action of, 144 auricular curve, 153 diastole of, 145 Heart— Cont'd isometric period in, 149 muscle, properties, 176 nutrition of, 161 opening and closing of valves, 154 oxygen requirements- of, 396 oxygen supply of, 164 perfusion of outside body, 161 postsphygmic period, 150 presphygmic period, 149 pressure in, 146 pumping action of, 134, 144 resuscitation in situ, 164 rhythmic power in, 170, 174 sounds of, 157 systole of, 145 utilization of glucose in, 681 vagus control of, cold blooded, 217 vagus control of, mammalian, 220 vagus terminations in, 225 ventricular curve, 146 work of, 212 Heart beat : • arrhythmia of, 266 myogenie hypothesis of, 171 neurogenie hypothesis of, 170, 172 origin of, in cold-blooded animals, 170 origin of, in mammalian, 182, 189 pace maker of, 174. propagation of, 224 sympathetic control of, 223, 227 ultimum moriens, 185 vagus control of, 217, 220 Heart block, 174, 270, 276 effect of vagus on, 219 Heart disease, vital capacity of lungs in, 314 Heart-lung preparation, 158 Heat production and age and sex, 541 and body weight, 539 surface, 540 disease, 542 Heat spots, 792 Heat value of foods, 535 Hematocrit, 7 Hematoporphyrin, 496 Hemiplegia, 258 Hemodromograph, 200 Hemoglobin, 91 dissociation constant, 388 dissociation, curve of, 380, 382, 383 estimation of, 92 rate of dissociation, 386 relationship to bile pigments, 496 specific oxygen capacity of, 379 transportation of O2 by, 390 Hemolysis, 7, 95 Hemolytie jaundice, 93 Hemophilia, 112 Hemopoietic activities of bone marrow, 93 Hemorrhage, 59 immediate effects of, 137 recovery from, 138 INDEX 895 Hemorrhagic diseases, 112 Henle, loop of, 507 Hepatic artery, flow in, 255 Heterocyclic compounds, 604 Hexoses, 652 Hibernating animal, metabolism of, 549 Hibernation, breathing during, 374 Higher functions of cerebriim, 860 H ion or hydrogen ion, 168 H-ion concentration, 22 after hemorrhage, 142 catalytic power of, 23 determination of, 31 of intestinal contents, 505 law of mass action and, 26 method of expressing, 27 method of measurement: electric method, 29 indicator method, 32 standard solutions for, 34 H-ion concentration in blood: effect on dissociation curve, 386, 389 effect on respiratory center, 335 Hippuric acid, 530, 630, 710 Hirudin, 100 Histamine, 397, 502 Histidine, 606, 623 Homogentisic acid, 502, 531 Hordein, 578 Hormones, 3, 729 in control of circulation, 216 respiratory, 349 Howell theory (blood clotting), 106 Hunger, 471 Hunger contractions : alcoholic beverages and, 478 control of, 476 during starvation, 475 in esophagus, 474 inhibition of, 477 in stomach, 471 nerve centers and, 479 remote effects of, 474 rhythmic, 471 splanchnic nerve and, 477 vagus nerve and, 477 Hiirthle manometer, 126, 146 • Hydrocephalus, 249, 253 Hydrochloric acid, amount of, 482 and emptying of stomach, 460 functions of, 482 source of, 483 Hydrogen ion (see H ion) Hyperacidity, 461 Hyperesthesia, 831 Hyperglycemia, in pancreatic diabetes, 680 postprandial, 659 splanchnic, 673 Hyperpnea, 349, 359 Hyperthyroidism, 756 Hypertonie solution, 6 Hypertonicity, 63 Hypogastrie nerves, 797 Hypothyroidism, 755 Hypotonic solution, 6 Hypoxanthine, 635 Ignition juice, 438 Ileocolic muscles, 882 Ileocolie sphincter, 467, 469 Imbibition, 62 Imidazole and growth, 586, 604, 623 Imidazole ring, 623 Imidazolylethylamine, 397, 426, 502 Immediate induction, 823 Impulses, nature of, 830 Index test, 870 Indican, 632 Indicator method, list of indicators, 33 Indole, 501, 604, 632 Indoxyl sulphate of potassium, 632 Induction, immediate, 823 successive, 824 Inhibition, reciprocal, 814 Inhibitory effects of autonomic nerves, 884 Inner vation, reciprocal, 814 Inorganic constituents of urine, 531 Inorganic salts and growth, 586 Inosine, 639 Inosinic acid, 637 Inspiration, negative pressure during, 305 Integration of allied reflexes, 822 Integration of nervous system, 809 Intercostal muscles, 319 Internal anal sphincter muscles, 882 Internal vesical sphincter, 882 Internal respiration, 378 Intestinal bacteria, 657 Intestinal juice, control of, 442 Intestinal obstruction, 470, 504 Intestinal secretions, 441 Intestine : absorption from, 13 anastomosis of, 470 bacterial digestion in, 499 digestion in, 489 law of, 466 movements of: large, 468 clinical conditions effecting, 470 small, 463 nature of, 466 nervous control of, 467 Intracardiae pressure curves, 146, 151 Intracranial pressure, 253 Intragastric pressure, 454 Intrapleural pressure, 304 Intrapulmonie pressure, 299 Intra vitam anticoagulants, 100 Intravascular clotting, 107 Inulin, 664 Invertase, 81, 492, 657 Invertebrates, segmented, 783 Inverting enzymes, 657 896 INDEX Involuntary fibers, course of, 879 Iodine value of fats, 688 lonization, 16 Irradiation in nervous system, 826 Irreversibility in reflexes, 810 Isoeleetric point, 64 Isoleucine, 604 Isomaltose, 79 Isometric period, 149 Isotonic solution, 6 Jacksonian epilepsy, 849 Jugular pulse tracing, 274 Juice, gastric, 434 intestinal, 442 pancreatic, 441 K Keith and Flack, conducting tissue in heart, 185 Kent, bundle of, 185 Ketonic acid, 708 Ketosis, 683 Kidney, oxygen requirements of, 396 removal of, 621 structure of, 507 Knee-jerk, 804, 815, 828 reinforcement of, 829 Lactalbumin, 577 Lactam, 649 Lactase, 491, 657 Lactic acid, 397, 603, 676, 665, 708 effect on respiratory center, 376 in mountain sickness, 362 produced by exercise, 367, 413 Lactim, 649 Language, 860 Latent period, 809 Laws of gases, 336 of mass action, 23 applied measurement of H-ion concen- tration, 26 Lead poisoning, 650 Lecithin, 689 estimation of, 697 in bile, 498 in blood, 696, 699 Leech extract, 100 . Legumelin, 578 Legumin, 578 Lesions of nervous system, 835 Leucine, 604, 666 Leucocytes, 96 sensitizing of, 70 transitorial, 97 Leucocythemia, 648 Levulose, 656 Levy and Rowntree method, 41 Limulus, heartbeat of, 172 Lipase, 25, 90, 491, 687 Lipemia, 699 Lipoids of blood, 699 Lissauer-tract, 831 List of indicators, 33 Litten's diaphragm phenomenon, 321 Liver : circulation through, 255 disease of, 620 glyeogen in, 662 metabolism of fats in, 701 perfusion of, 618 removal of, 617 urea formation in, 617 Local irritants, 243 Localization, cerebral, 843 Locke solution, 168 Loven reflex, 244 Lungs, circulation through, 253 mode of expansion of, 325 Lymph: absorption into, 13 electric conductivity, 16 filtration in, 118 formation and circulation, 115 formation of, 15 Lymph spaces, 115 Lymphagogues, 119 Lymphatics, 115 Lymphocytes, 96 Lyophobe colloids, 60 Lysine, 592, 605 Lysine and growth, 576 M Maculae acusticse, 873 Maintenance, diets for, 579 Malingerers, 42 Maltase, 491, 657 Maltose, 491 Manometer : blood-gas differential, 382 Hiirthle, 124, 146 mercury, 123 . optical, 146 spring, 126 valved mercury, 152 Mark-time reflex, 806 Mass action, 23 Mass action and H-ion concentration, 26 Mass movements of blood, 281 Mastication, 444 Mechanics of respiration, 299 Medulla, section above, 839 Megacaryocytes, 103 Melting point, fats, 687 Membrane synaptic, 798 Memory, 786 Mercury manometer, 123 INDKX 897 Metabolism: calculations, 544 endogenous, 615 exogenous, 615 general, 534 in starvation, 566 normal, 570 of carbohydrates, 652 of fats, 686 of proteins, 595 of purines, 637 special, 534 Methyl glyoxal, 665 Methyl group, 598 Methyl purines, 635 Methylation, 627 Methylglyoxal, 665, 666 Mett's method, 487 Microcytes, 94 Microtonometer, 339 Mid-capacity of lungs, 311 Milk, clotting of, 488 Miniature stomach, 433 Minimal air, 300 Mononuclear leucocytes, 96 Morawitz theory, blood clotting, 107 Motor areas, ablation of, 843 stimulation of, 844, 846, 848 Motor nerves of segmental duct muscles, 881 Mountain sickness, 360, 399 adaptation to, 400 alveolar CO2 in, 360 blood corpuscles in, 401 Movements, of intestine, 463 of stomach, 452 Municipal food statistics, 591 Muscarine, action on heart, 226 Muscle, cardiac, properties of, 176 refractory period, 178 respiration in, 395 staircase phenomenon (treppe), 177 skeletal, 177 respiration in, 394 Muscles, antagonistic, 818 Muscular exercise, 243, 539 circulatory changes during, 410 effect on metabolism, 551 effect on respiration, 366 H-ion during, 413 purines during, 647 redistribution of blood during, 415 respiratory changes during, 410 temperature of blood during, 415 Mutual precipitation of colloids, 56 My enteric reflex, 796 Myogeuie hypothesis of heartbeat, 171 Myxedema, 755 energy output in, 542 N '. Narcotics and blood fat, 698 Necrosis of liver, 620 Negative pressure in ventricle, 152 Nephelometer, 697 Nephrectomy, 621 Nephritis, 650 acidosis in, 683 urea retention in, 528 Nerves : of skin, 796 network, 4, 29 regeneration of, 36 segmental distribution of, 837 specific properties of, 789 vasodilator, 797 Nerve cells, 33, 799 Nervi erigentes, 231 Nervous control: of gastric secretion, 434 of ileocolic sphincter, 468 of intestinal glands, 442 of intestinal movements, 467 of pancreas, 427 of salivary glands, 423 of stomach movements, 458 Nervous diabetes, 672 in man, 674 Nervous system: autonomic, 877 bulbar fibers, 882 functions of, 884 sacral fibers, 882 thoracicolumbar fibers, 880 effect of section at various levels of: anterior root, 99, 835 just behind medulla, 839 just behind post. corp. quad., 840 just in front of ant. corp. quad., 840 posterior roots, 836 spinal cord, 839 evolution of, 718 influence on excretion of urine, 519 integration of, 786, 809 Network, nerve, 796 . neurofibrils, 800 neuropile, 784, 797 Neurogenie hypothesis, of heart, 172 Neurons, 784 association, 783, 785 intermediate, 802 internuncial, 802 Neutrality, regulation of, 36 - Nicotine, 233 action on vagus, 226 Nissl bodies, 800 Nitrogen : excretion of,' premortal rise, 566- in starvation, 566 undetermined, urine, 613 Nitrogen balance, 570 Nitrogenous constituents of urine, 523 Nitrogenous equilibrium, 571 Nitrogenous metabolites, in starvation, 568 898 INDEX Nociceptive, 795, 804 impulses, 832 reflex, 825 Noeud vital, 327 Nonelectrolytes, 16 Nonthreshold substances, 512 Normal acid, 22 Normoblasts, 93 Nuclease, 638 Nucleic acid, 637, 689 Nuclein ferments, 90 Nucleins, 637 Nucleoside, 638 Nucleotide, 638 Nystagmus, 871, 875 O Obesity, Banting cure for, 571 Oleic acid, 868 Olein, 868 Opsonins, 70 Organs, loss of weight during starvation, 568 perfusion of, 618 Ornithine, 616, 631 Ornithuric acid, 631 Orthopnea, 313, 318 Oscillatory method of blood pressure, 130 Osmometer, 5, 230 Osmosis, 4 Osmotic pressure, 4, 10 and formation of lymph, 13 and hemolysis, 7 and plasmolysis, 8 measurement by depression of freez- ing point, 11 in physiologic mechanisms, 13 in production of urine by kidneys, 14 of transfusates, 141 Ovalbumin, as food, 577 Ovovitellin, as food, 577 Oxidases, 82 Oxidation of blood, 387 Oxybutyric acid, 616, 683, 709 Oxygen : coefficient of oxidation, 393 determination of, 562 estimation in blood, 390 requirements of tissues, 393 tension in alveolar air, 340, 344 tension in arterial blood, 337 transportation by blood, 379 volume percentage in blood, 390 Oxygen insufficiency, and periodic breath- ing, 373 effect of, on respiration, 350, 359 Oxygen supply of heart, 164 Oxyproteic acid, 629 Pacchionian body, 249 Pain: sensation of, 832 Pain, sensation of — Coiit'd transmission in eorji, 830 sense, 795 Palatability, 593 Palmitic acid, 686, 707 Pancreas : hormone control of, 420 histologic changes of, 429 oxygen requirements, 396 nervous control of, 427 sugar metabolism and, 678 Pancreatic diabetes, 678 Pancreatic digestion, 489 Pancreatic juice, 441 and fat digestion, 690 secretion of, 420, 426 Pancreatin, 490 Parasympathetic system, 882 Paroxysmal tachycardia, 269, 278 Partial dissociation, 271 Partial pressure of gases, 336 Pathways, sensory, in spinal cord, 830 Pelvic ganglionic group, 878 Pentose, 664 Pepsin, action of, 485 products of, 486 Pepsinogen, 485 Peptides, 601 Peptone, 105, 486 Perception, 861 Perfusion, of kidney, 631 of liver, 618 Perfusion fluid, of heart, 165 Perfusion of heart, 161 Periodic breathing, causes of, 372 types of, 371 Peripheral resistance, 134, 229 Peristalsis : in esophagus, 446 in large intestine, 468 in small intestine, 465 in stomach, 453, 456 Peristaltic rush, 466, 470 Peristaltic wave, 465 Pernicious anemia, energy output in, 542 Peroxidases, 82 PH, 27 Phagocytes, 97 Phenaceturic acid, 710 Phenol, 501 Phenolacetie acid, 502 Phenolphthalein, 482, 525 Phenylacetic acid, 631, 710 Phenylalanine, 604 Phenyl group, 604 Phlorhizin, 664, 665 Phosphates, excretion of, 47 Phosphate solutions for H-ion, 34 Phosphates of urine, 532 Phospholipins, 689 in bile, 498 Phrenic center, 328 Physicoehemical basis, 1 INDEX 899 Physiologic processes depending on ad- sorption, 69 Pigments, absorption of, 117 Piloearpine, action on heart, 226 Pilomotor fibers, 880 Pitot's tubes, 201 Plasma, 99 Plasmolysis, 8 Platelets, of blood, 97, 106 Plethora, 86 Plethysmograph, 209, 230, 273, 303 Pleurisy, 324 Plexus of Auerbach and Meissner, 466, 796 Pneumothorax, 305 Poikilocytes, 94 Polygraph, 273 Polyneuritis, 584 Polynuclear cells, 96 Polypeptides, 487, 601 Polyphosphoric acid, 637 Polysaccharides, 489 Polysphygmograms, 273 Portal vein, bloodflow in, 255 Postdicrotic wave, pulse, 203 Postprandial hyperglycemia, 659 Postcentral convolutions, 850, 854 Posterior roots, 787, 836 Postsphygmic period, 150 Postural reflexes, 826 Potassium, microchemical test for, 421 Potassium ions, on heart, 167 Potential acidity of urine, 524 Precentral convolutions, 843, 854 Precipitins, 595- Predicrotic wave, pulse, 203 Prefrontal region, 854 Premature beats, 277 Premortal rise, 566 Presphygmie period, 149 'Pressor impulses, 238, 239, 240 Pressure : intrapleural, 304 effect of, in blood pressure, 306 intrapulmonic, 299 negative, 305 osmotic, 10 Pressure pulse, 127 Principle of Willard Gibbs, 66 Proline, 604 Proprioceptive impulses, 865 Proprioceptors, 822 Prosecretin, 426 Proteases, 89 Protein sparers, 571 Proteinases, 80 Proteins : as colloids, 63 bacterial digestion of, 501 chemistry of, 597 metabolism of, 595, 613 end products, 613 minimum requirement, 572, 592 of blood, 88 Proteins — -Gont 'd. relative value of, for growth, 611 salting out of, 60 Proteose, 486 Protopathic impulses, 831 Protopathic receptors, 790 Protothrombin, 103, 106, 111 Psychopathology, 862 Ptomaines, 502, 629 Ptyalin, 491, 656 Pulmonary circulation, 253 Pulmonary ventilation, 350 Pulses, 198 abnormal, 276 alternans, 181 bigeminus, 181 contour of wave, 200 length of wave, 199 palpable, 201 pressure, 127 pulse curves, 202 pulse waves, 189, 200, 203 rate of transmission, 198 velocity, 200 venous, central, 205, 274 venous, peripheral, 205 Purkinje fibers, 184 Purine bodies (see Purines) Purines : chemistry of, 529, 613, 634 endogenous, 641, 643 exogenous, 641 metabolism of, 637 in starvation, 569 synthesis of, 646 Putrefaction, intestinal, 501, 530 Putrescine, 629 Pyloric . canal, 452 Pyloric sphincter, control of, 456 Pyloric vestibule, 453 Pyramidal cell lamina, 854 Pyrimidine bases, 636, 637 Pyruvic acid, 600, 708 B Runii communicantes, 233 Eaynaud's disease, bloodflow in, 258 Eeaction deviation, 871 Eeaction of urine, 524 Eeactions depending on adsorption, 66 Eeactions of body fluids, 35 Eeceptors, 782, 788 distance, 785 epicritic, 790 external, 788, 822 internal, 788 of skin, 790 projicient, 785, 788 proprio, 822 protopathic, 790 temperature, 791 touch, 793 900 INDEX Reciprocal inhibition, 814 action of strychnine on, 819 Eeciprocal innervation of blood vessels, 241, 814 Bed blood corpuscles, origin of, 92 Eeduction of blood, 387 Referred pain, 885 Reflex, conditioned, 431 unconditioned, 431 Reflex arc, 784 after effect, 810 grading of intensity, 809 irreversibility of conduction, 810 latent period, 809 oxygen deprivation, 813 properties of, 13, 29, 49 refractory period, 811 summation, 810 Reflex conduction, resistance of, 813 Reflexes : allied, simultaneous integration of, 823 antagonistic, 824 axon, 797 Babinski, 807 conditioned, 856 cremasterie, 856 crossed extension, 804 extensor thrust, 805 fatigue of, 825 flexion, 804, 821 integration of allied, 821 interaction between, 821 irradiation of, 826 mark-time, 806 myenteric, 796 nature of, 825 nociceptive, 825 postural, 826 unconditioned, 431, 856 Refractive index, blood, 88 Refractory period, 811 Refractrometric methods, 88 Regeneration of erythrocytes, 93 Regulation of neutrality, 36 Regurgitation of gastric contents, 449 Reichart-Meissl value of fats, 688 Reinforcement of knee-jerk, 829 Renal diabetes, 661 Renal function, theories of, 511 Rennin, 488 Reserve alkalinity, measurements of, indi- rect methods, 42, 46 measurement of, titration methods, 41 Residual air, 300, 311 Respiration : abdominal, 307 beyond the lungs, 378 during muscular exercise, 410 in compressed air, 402 in rarefied air, 399 mechanics of, 299 movements of diaphragm in, 321 movements of ribs in, 319 Respiration calorimeter, 536 Respiratory center, 327 afferent impulses to, 331, 332 automaticity of, 329 hormone control of, 335, 349 reflex control of, 331 sensitivity to alveolar CO,, 357 stimulation by OCX, 352 subsidiary, 328 Respiratory changes in muscular exercise, 410 Respiratory exchange : according to body weight, 550 and body temperature, 551 clinical method for determining, 554 in diabetes, 678 and muscular exercise, 551 and temperature of environment, 551 in tissues, 393, 397 Respiratory hormone, nature of, 349 Respiratory movements, 315 Respiratory passages, pressure of air in, 299 Respiratory quotient, 545 in diabetes, 678 influence of diet on, 547 influence of metabolism on, 549 influence of muscular activity on, 370 Respiratory tracings, 303 Respiratory valves, Pearce's, 554 Reticulated erythroblasts, 93 Reversible action of enzymes, 25 Ribs, movements of, 315 musculature of, 319 undulatory movements of, 317 Right lateral connection, heart, 185 Rigidity, decerebrate, 808 Rolandic fissure, 855 Roots, 787 anterior, 787, 835 posterior, 787, 836 Rhythmic segmentation, 464 Sacral outflow, 882 Salicylates, 648, 657 Saline injection, effect on blood pressure, 139 Saliva, control of secretion, nervous, 423 psychic, 431, 856 normal secretion, 431 Salt, dietetic value, 586 -~ Salted blood, 100 Salting of proteins, 60 Saponification, 687 Sarcosine, 623 Saturation limits, 652, 654 Scratch reflex, 805, 821 Scurvy, 585 Sea anemone, nervous system of, 783 • Second wind, 415 Secretory fibers, varieties of, 424 INDEX 901 Secretin, 425 chemical nature of, 426 mechanism of action of, 420 Secretion (see under various glands) general considerations, 418 Segmental distribution of nerves, 837 Segmentation movements, 463 Segmented invertebrates, nervous system, 783 Semicircular canals, 873 eye movements and, 875 removal of, 874 Semilunar valves, 150, 155 Semipermeable membrane, 4 Sense, temperature, 791 touch, 793 pain, 795 Sensory centers, 850, 851 Serine, 603 Serum albumin, 87 Serum globulin, 87 Sex, effect on creatinine excretion, 624 effect on energy output, 541 Sham feeding, 435 Shell shock, 287 Shock, 287 anesthetic, 288 blood pressure in, 290 experimental investigations, 289 gravity, 287 hemorrhagic, 288 nervous, 289 recovery from, 805 secondary symptoms of, 295 spinal, 288, 803 surgical, 289 treatment of, 295 vasomotor control in, 290 Sinoauricular node, 185, 266 Sinus arrhythmia, 266, 277 Sinus bradycardia, 266, 277 Skatole, 501, 632 in urine, 531 Skeletal muscle, respiration in, 394 Skin, receptors of, 790 Soap, 686 Sodium ions, 166 Solution of gases, 336 Solutions : gas laws and, 3 gram molecular, 5, 22 hypertonic, hypotonic, and isotonic, 6 nature of, 3 Sorensen method for estimating amino groups, 599 .Sounds, cardiac, 157 recording of, 158 Specific conductivity, 17 Specific dynamic action,, 538 Specific gravity of urine, 522 Sphingomyelin, 689 Sphygmic period, 273 Sphygmograph, Dudgeon's, 201 Spinal animal, 804 Spinal column, 786 Spinal cord : ablation of, 839 in laboratory animals, 803 in man, 806 hcmisection of, 831 sensory pathways in, 830 successive degeneration in, 813 Spinal pathways, afferent, 830 Spinal shock, 803, 807 Spirometer, 556 Splanchnic circulation in shock, 292, 294 Splanchnic nerve, 233, 672 Sponges, nervous system of, 782 Stalagmometer, 65 Standard of neutrality, 26 Standard solutions, preparation of, 34 Stannius' ligature, 176 Starvation, 566 acidosis during, 569 cause of death, 570 effect of creatinine excretion, 625 „ energy output during, 568 excretion of nitrogen, 566 loss of weight, 568 nitrogenous metabolism, 568 purines during, 569 secretion of gastric juice during, 476 sensations during, 475 sulphur during, 569 treatment of diabetes, 684 Statistical method, in diet control, 589 Stearic acid, 687 Stokes-Adams syndrome, 193 Stomach : arrangement of food in, 455 digestion in, 481 emptying of, 456 effect of pathologic conditions on, 460 rate of, 458 gas in, 462 miniature, 433 movements of, 451 effect on food, 454 Stroma of red cell, 91 Stromuhr, 207 Strychnine, action on reciprocal inhibition, 819 Subarachnoid space, 116, 248 Subcostal angle, 321 Subcostal borders, 321 Subdural space, 116 Submicrons, 54 Substantia-gelatinosa, 831 Successive induction, 824 Successive regeneration, 813 Sugar, storage of, 662 Sugar level in blood, 657 Sugar metabolism (see Carbohydrates), 652 relation of pancreas to, 678 902 INDEX Sulphates, ethereal, 632 Sulphates, of urine, 532 Sulphur, excretion of, 614 in starvation, 569 Summation in reflexes, 810 Superior laryngeal nerve, influence on res- piration, 334 Supplemental air, 300 Surface area, and energy output, 540 Surface tension, measurement of, 64 Surgical shock, 289 Survival period, 580 Suspensions, 51 Suspensoids, colloids, 60 Swallowing, 445 center, 447 of liquid food, 448 nervous control of, 447 sounds produced by, 449 x-ray during, 449 Sweat glands, 880 Sympathetic control of heart, 227 afferent, 223 Sympathetic nerve, 424 Sympathetic system, 878, 880 Synapsis, 784, 797, 819 Synaptic fatigue, 296 Synaptic membrane, 798 Syntonin, 486 Systolic index, 128 Systolic pressure, 127 measurement of, in man, 128 Tabes dorsalis, 286 Tachycardia, paroxysmal, 269 Tactile impulses, 833 transmission in cord, 833 Taurine, 494 Taurocholic acid, 494 Temperature : after-effect, 792 effect on metabolism, 551 sensation of, 792, 832 transmission in cord, 832 Temporary association, 857 Tendon jerks, 828 Tension of CO, in venous blood, 342 of gases in alveolar air, 46, 339 Tetanus, in stomach, 471 Tetanus toxin, action on reciprocal inhi- bition, 819 Theine, 635 Theobromine, 635 Thermoesthesiometer, 791 Thoracic operculum, 316 Thoracicolumbar outflow, 880 Thrombin, 102 Thrombogen, 106 Thrombokinases, 106 Thromboplastin, 106, 111 Thrombosis, 107 Thrombus formation, 113 Thymic acid, 649 Thymine, 637 Tidal air, 300 Tissot method, 544, 556 Tissue fluid, 116 Tissue juice, 117 Tissues : amino acids in, 607 influence of, on clotting, 104 oxygen requirements of, 393, 397 utilization of glucose by, 681 Titrable acidity and alkalinity, 22 Tonometer, 338, 381 Tonus rhythm, of stomach, 471 Torcular'herophili, 250 Touch, discrimination, 794 localization, 37, 795 sense, 793 Toxins, 69 Transfusion of blood, 135, 139 Trephining, 253 Treppe, 178 Trichlorlactamide, 635 Trimethylamine, 629 True colloidal solutions, 51 Trypsin, 426, 428, 601 action of, 489 Trypsinogen, 426, 428 Tryptophane, 592, 596, 604, 632 and growth, 576, 578 Tubules, uriniferous, function of, 517 Tumors and diet, 582 Turbidity of colloids, 51 Turck's method, 115, 842 Tyndall phenomenon, colloids, 51 Tyrodes solution, 168 Tyrosine, 502, 604, 632, 666 Tryptic digestion, products of, 490 U Ultramicroscope, 800 Uncompensated acidosis, 39 Unconditioned reflex, 431, 856 Undetermined nitrogen, 613, 629 Undulatory movement of ribs, 317 Urea, 527, 608 in blood, 610 during disease, 651 excretion of, 615 Urease, 82, 610 Uric acid, 529, 531, 614, 618 amount of, 522 bases of, 531 chemical nature of, 634 endogenous excretion, 647 in disease, 651 metabolism of, 643 of blood, 648 synthesis of, 644 under drugs, 648 Uric acid diathesis, 634 INDEX 903 Uricase, 640 Uricemia, 650 Uricolytic index, 641 Urine : acids of, 531 amino acid, 530 aromatic oxyacids of, 530 chlorides of, 531 composition, 521 creatinine of, 529 depression of freezing point of, 523 hippuric acid, 530 homogentisic acid, 531 inorganic constituents of, 531 nitrogenous constituents of, 523 normal organic salts of, 523 phosphates, 532 physical processes involved in produc- tion of, 14 purine bodies of, 529 rate of excretion, 643 reaction of, 524 skatole, 531 solid constituents of, 525 specific gravity of, 522 sulphates of, 532 total potential acidity of, 524 urea of, 527 Uriniferous tubule, 507 Urobilinogen, 496 Utilization limit, 653 Vagus, 878 control of heart, 217 impulses, afferent, 222 Vagus center, effect of nicotine on, 226 location of, 222 tonieity of, 221 Vagus nerve, influence on respiration, 332 Valine, 604, 606 Valves, cardiac, mechanism of, 154 auriculoventricular, 154 semilunar, 155 Van Slyke method for acidosis, 42, 44 Van Slyke method for amino groups, 600 Vascular reflex, 283 Varicose veins, 214 Vasoconstriction, 229 Vasoconstrictor fibers, 229 methods of detecting, 229 of extremities, 233 of head, 233 of viscera, 233 origin of, 232 Vasodilator fibers, 234 methods for detecting, 229 origin of, 234 Vasodilator nerves, 797 Vasomotor center: afferent impulses, 238, 239 chief center, 235 effect of H-ion of blood on, 238 Vasomotor center — Cont 'd hormone control of, 237 subsidiary centers, 235 Vasomotor fibers, 231 origin of, 232 Vasotonic impulses, 240 Veins, disappearance of pulse in, 205 Velocity constant, enzymes, 75 Velocity, mean lineal, 206 pulse, 200 Venous blood, tension of CO, in, 342 Venous outflow, 230 Venous pulse tracing, 273 Venous return to heart, 292 Venous sinus, 248 Ventilation of lungs, 350 Ventricle, curves of pressure in, 146, 148, 151 Ventricles : conductivity tissue of, 184 fibrillation, 195 spread of beat in, 192, 194 Vignin, 578 Viscera, blood supply of, 247 Visceral bloodflow, 212 Visceral sensitiveness, 885 Viscosity of blood, 140 Visual center, 851 Visual psychic areas, 854 Visual sensory area, 854 Vital activity, 14 Vital capacity, 300, 313 in disease, 314 Vital theory of urine excretion, 572 Vitamines, 584 Vividiffusion, 606 Volition, 786 Vomiting, 449 W Water content of blood, 86 Water hammer, in blood pressure measurement, 133 Wheatstone bridge, 18 White crescentic line, 226 Wiggers manometer, 146 Willard Gibbs, principle of, 66 Word blindness, 863 Word centers, 862 Word deafness, 863 Xanthine, 635 Xanthine oxidase, 639 Xanthosine, 639 X-rays, in study of stomach, 433 movements of stomach seen by aid of, 451 Zein, inadequacy for growth, 578 Zymogen granules, 420, 421, 429 UNIVERSITY OF CALIFORNIA Medical Center Library THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW Books not returned on time are subject to fines according to the Library Lending Code. Books not in demand may be renewed if application is made before expiration of loan period. 10m-7,'59 (A3819s4)4128