7 I rt y A MANUAL OF PHYSIOLOGY Witb practical BY °N'.' STEW ART, M A ^ D Sc ^ M D EDIN ^ D 517 7. Urea ------ - 518 8. Ammonia in urine - - - - - '521 9. Total nitrogen in urine - - - 521 10. Uric acid ... . 523 11. Creatinin .... . 523 12. Hippuric acid - - - - - 524 13. Proteins in urine ... . 524 14. Sugar in urine - - 525 15. Pentoses in urine - - 528 16. Acetone in urine - - 529 17. Determination of the freezing-point of urine - - 529 1 8. Examination of urine - - 531 19. Urinary sediments - - 531 CHAPTERS X., XI., AND XII. 1. Glycogen . . 7j5 2. Catheterism - - - - 3. Experimental glycosuria . - (1) Injection of sugar into the blood » - - (2) Phlorhizin glycoauria . (3) Alimentary glycosuria - - - (4) Estimation of the sugar in blood - - - 4. Milk - 5. Cheese - - 719 6. Flour - - *. 719 7. Bread - ••• ... 720 8. Excretion of urea (and total nitrogen) and proteins in food - 720 9. Action of epinephrin - _ 720 10. Measurement of the heat given off in respiration - - 721 CONTENTS xxiii CHAPTERS XIII. AND XIV. PAGE 1. Difference of make and break induction shocks - - 807 2. Stimulation by the voltaic current - - 810 3. Ciliary motion - 811 4. Direct excitability of muscle — curara - - - 811 5. Graphic record of ' twitch ' - - 811 6. Influence of temperature on the muscle-curve - - 813 7. Influence of load on the muscle-curve - - 813 8. Influence of fatigue on the muscle-curve - 813 9. Seat of exhaustion in fatigue of the muscle-nerve preparation 813 10. Influence of veratrine on muscular contraction - 814 11. Measurement of the latent period of muscular contraction - 815 12. Summation of stimuli - - - 816 13. Superposition of contractions - - 816 1 1. Composition of tetanus - 816 15 Contraction of smooth muscles - - - 817 1 6 Velocity of the nerve-impulse - - - - 818 17. Chemistry of muscle - - - 819 1 8. Reaction of muscle in rest, activity, and rigor • - 820 CHAPTER XV. 1. Galvani's experiment - .... 842 2. Contraction without metals - .... 842 3. Secondary contraction - - 842 4. Demarcation and action currents with capillary electrometer 842 5. Action current of the heart - 844 6. Electrotonus - 844 7. Paradoxical contraction - 844 8. Alterations in excitability and conductivity produced in nerve by a voltaic current - 844 9. Formula of contraction - 845 10. Formula of contraction for (human) nerves in situ - 846 11. Ritter's tetanus - - 846 CHAPTER XVI. 1 . Section and stimulation of nerve-roots •* 994 2. Reflex action in the ' spinal ' frog - - 995 3. Reflex time - - 995 4. Inhibition of the reflexes - 996 5. Spinal cord and muscular tonus - 996 6. Spinal cord and tonus of the bloodvessels • - 996 7. Action of strychnine - ...... 996 8. Mammalian spinal preparation - 996 9. Decerebrate cat preparation ----- 998 10. Swallowing reflex •« 1000 xxiv CONTENTS PAGE 11. Reflex postural tonus .... IQOC 12. Reflexes in man - - 1000 13. Excision of cerebral hemispheres in the frog - - 1000 14. Excision of cerebral hemispheres in the pigeon - 1001 15. Stimulation of the motor areas in the dog - - 1001 CHAPTER XVIII. 1. Dissection of the eye - - 1101 2. Formation of inverted image on the retina - 1102 3. Helmholtz's phakoscope - 1102 4. Schemer's experiment - 1103 5. Kuhne's artificial eye - - - 1104 6. Astigmatism (ophthalmometer) - 1105 7. Spherical aberration - - 1106 8. Chromatic aberration - - 1106 9. Measurement of the field of vision - - noo ;o. Mapping the blind spot - noy EX. The yellow spot - 1107 12. Ophthalmoscope - - 1108 13. Retinoscopy - - 1109 14. Pupillo-dilator and constrictor fibres - - - - mo 15. Colour-mixing - - - . _ -mi 1 6. After-images - - --mi 17. Retinal fatigue - . IIXI 18. Visual acuity - - . mi 19. Colour-blindness - - - _ . -1112 20. Talbot's law - - - - - . -1113 21. Purkinje's figures . . m? 22. Relation of pitch and vibration frequency - ... - 1113 23. Beats . - 1113 24. Sympathetic vibration . . . -1113 25. Galton's whistle - - . . 26. Cranial conduction of sound - - . . 27- Taste - ..... x 28. Smell - ... . I1I4 29. Touch and pressure - - - - . -1114 30. Temperature sensations - . - . -1115 31- Pa^ - CHAPTER XIX. 1. Contractions of isolated uterine rings - . . - 1147 2. Comparison of changes of tone produced in uterus segments by different concentrations of adrenalin Partition of adrenalin between serum and corpuscles - 1149 A MANUAL OF PHYSIOLOGY CHAPTER I INTRODUCTION LIVING matter, whether it is studied in plants or in animals, has certain peculiarities of chemical composition and structure, but especially certain peculiarities of action or function, which mark it off from the unorganized material of the dead world around it. Chemical Composition of Living Matter. — Although we cannot analyze the living substance as such, we can to a certain, but limited, extent reconstruct it, so to speak, from its ruins. When subjected to analytical processes, which necessarily kill it, living matter invariably yields bodies of the class of proteins, exceedingly complex substances, which have approximately the following com- position: Carbon, 51-5 to 54-5 per cent.; oxygen, 20-9 to 23-5 per cent.; nitrogen, 15-2 to 17 per cent.; hydrogen, 6-9 to 7-3 per cent., with small quantities of sulphur. Nucleo-proteins, which are com- pounds of ordinary proteins with nucleic acids, a series of sulphur- free organic acids rich in phosphorus, are constantly met with. Certain carbo-hydrates, composed of carbon, hydrogen, and oxygen (the last two in the proportions necessary to form water), of which glycogen (C6Hto05)w may be taken as a type, appear to be always present. Fats, which consist of carbon, hydrogen, and oxygen, and of which tristeaiin, a compound of stearic acid with glycerin, of the formula C3H5,3(C18H3502), may be given as an example, are often, and certain lipoids, e.g., lecithin (p. 4), are always, found. Finally, water and certain inorganic salts, such as the chlorides and phosphates of sodium, potassium, and calcium, are constantly present. The Proteins. — The constitution of the protein molecule is still un- known ; but when proteins are broken down by the action of ferments, such as exist in gastric and in pancreatic juice, or by chemical methods — for example, by boiling with dilute acids — the most important of the cleavage products are various amino-acids (p. 360). It has there- fore been suggested that proteins are built up by the linking together of amino-acids, the different proteins differing quantitatively or quali- tatively as regards the amino-acids present (E. Fischer). Thus serum- albumin and egg-albumin yield no glycin or glycocoll (ammo-acetic acid, CH2.NHa.COOH), while glycin is constantly found among the cleavage products of serum-globulin. And while leucin (a-amino- 2 INTRODUCTION isobutylacetic acid) is present to the extent of about 20" 5 per cent, in the cleavage products of (horse's) serum-albumin, (hen's) egg-albumin yields only 7' i per cent. On the other hand, egg-albumin yields 8'i per cent, of alanin (amino- propionic acid, C2H4.NH2.COOH), while serum-albumin yields only 2' 7 par cent. Of the aromatic amino-acids — that is, amino-acids united to the benzene ring — phenyl-alanin (amino-propionic acid in which one atom of H is replaced by phenyl, C6H5) is obtained to the extent of 4" 4 per cent, from egg-albumin, and a little over 3 per cent, from serum- albumin. Ty rosin or oxyphenyl-alanin (amino-propionic acid in which a H atom is replaced by oxyphenyl, C6H4.OH) appears to the amount of 1*5 per cent, among the cleavage products of egg-albumin, and to the amount of 2'i per cent, among those of serum-albumin. It is an interesting point in this connection that gelatin, which yields i6'5 per cent, of glycin, yields no tyrosin at all; tryptophane, an aromatic amino-acid still more complex than tyrosin, is also absent. These facts afford an explanation of certain colour reactions of proteins long known empirically, but only recently understood (p. 8). The process by which the protein molecule is thus decomposed is called hydrolysis — that is, the molecule takes up water, and then splits into smaller molecules. The hydrolysis occurs in various stages, bodies like acid- or alkali- albumin (meta- or infra-proteins) being first formed, then proteoses, then peptones. The peptones are further split into bodies containing a relatively small number of amino-acids linked together. These bodies are called polypeptides, which finally are decomposed so as to yield the individual amino-acids, also called in this connection the peptides or monopeptides, the " building-stones " out of which the protein molecule is constructed. The inverse process can also be carried on to a certain extent, and Fischer has taken an important step towards the eventual synthesis of proteins by showing how polypeptides of increasing com- plexity can be built up by linking amino-acids together. When two amino-acids are so united, the resulting compound is called a dipeptide ; with three amino-acids we get tripeptides, etc. Still more complicated polypeptides may thus be formed in the laboratory, which give some of the characteristic reactions of peptones. The numerous substances included in the group of proteins may be classified as follows, beginning with the simplest : 1. Protamins, such as the bodies called salmin and sturin present in fish-sperm. 2. Histones, bodies separated from blood-corpuscles. Globin, the protein constituent of haemoglobin, is one of them. Unlike the other groups of proteins, they are precipitated by ammonia. 3. Albumins. 4. Globulins. 5. Sclero-proteins or albuminoids, such as gelatin and keratin. 6. Phospho-proteins, including such substances as vitellin, a body obtainable from egg-yolk, and caseinogen, the chief protein of milk. They are rich in phosphorus, but are to be distinguished from nucleo- proteins, which also contain a relatively large amount of phosphorus, by the fact that they do not yield the purin bases, the characteristic products of the decomposition of nucleo-proteins. 7. Conjugated proteins, substances in which the protein molecule is united to another constituent, usually spoken of as a ' prosthetic ' group. Thus the nucleo-proteins consist of protein united with nucleic acid, the chromo -proteins (e.g., haemoglobin) of protein united with a pig- ment, and the gluco-proteins (e.g., mucin) of protein united with a carbo-hydrate group. CHEMICAL COMPOSITION OF LIVING MATTER 3 .Among the derivatives of proteins, the most important are those already mentioned as being produced in protein-hydrolysis, viz. : (a) Meta -proteins. (6) Proteoses, including albumose, the proteose derived from albu- min; globulose, that derived from globulin; gelatose, that derived from gelatin, etc. The proteoses may be further subdivided, according to the order in which they are formed in digestion into proto-proteoses, hetero-proteoses, and deutero-proteoses. (c) Peptones. (d) Polypeptides. The majority of these are artificial products, formed by the synthesis of amino-acids, although some can be obtained from proteins by hydrolysis. Only a few of those hitherto prepared give the biuret test. However formidable the above list may appear to the student, it gives an inadequate idea of the extreme complexity of the protein class and its richness in individuals. For, apart from the fact that the list has been purposely left incomplete, especially as regards the numerous vegetable proteins, there is the best evidence that proteins of the same name from different animal species have certain properties which dis- tinguish them from each other. The serum-albumins can be crystal- lized much more easily in some animals than in others. The same is conspicuously true of the haemoglobins, which differ also in certain animals in the relative proportion of sulphur and iron in the molecule, as well as in the crystalline form. Even when no chemical or physical differences have as yet been made out, proteins of the same name from the blood or organs of different species show notable ' specific ' differ- ences when subjected to certain biological tests (see, e.g., the paragraph on Precipitins, p. 31 ; and that on Anaphylaxis, p. 32). Carbo-Hydrates. — The most important carbo-hydrates in their physio- logical relations are dextrose, levulose, galactose, lactose, maltose, sucrose (cane-sugar), starch, and glycogen. As regards their chemical constitution, the simplest carbo-hydrates are aldehydes or ketones — that is, the first oxidation products of primary and secondary alcohols respectively. Thus dextrose is the aldehyde of sorbite, a hexatomic alcohol (an alcohol containing six OH groups), while levulose is the ketone of the isomeric alcohol called mannite, and galactose the alde- hyde of the isomeric alcohol called dulcite. The sugars containing six carbon atoms are termed hexoses. They include dextrose, levulose, and galactose. The empirical formula of these three simple sugars (or monosaccharides) is the same (CjHjaOg), but, owing to the different arrangement of the atoms or groups of atoms, they have each their characteristic properties by which they can be easily distinguished. For example, dextrose rotates the plane of polarization to the right, levulose to the left. By the union or ' condensation ' of two molecules of a monosaccharide, with loss of a molecule of water, a disaccharide is formed. Cane-sugar, maltose, and lactose, all with the same empirical formula, (C^H^On), are disaccharides. Cane-sugar yields on hydro- lysis a mixture of equal parts of dextrose and levulose ; lactose, a mix- ture of dextrose and galactose ; while maltose is converted into dextrose. By the condensation of more than two molecules of monosaccharide polysaccharides are formed, such as starch, dextrin, and glycogen. The exact molecular weights of these substances are unknown. Their general formula can be written (C6H10O5)», where n represents the number of monosaccharide molecules condensed to form the poly- saccharide, in the case of starch probably some hundreds. Fats and Lipoids. — The fats are compounds of higher fatty acids with glycerin (glycerin esters). The ordinary body-fat consi:ts of a 4 INTRODUCTION mixture of three neutral fats (palmitin, stearin, and olein) which differ both chemically and physically from each other — e.g., in melting-point and in the so-called iodine value, the number which represents the amount of iodine taken up from a standard solution. Olein melts at -5° C., palmitin at 45° C., and stearin at a still higher temperature. It is, therefore, the presence of olein which keeps the body-fat liquid at the temperature of the body. The fats are soluble in ether, in hot alcohol, and in many other liquids, but insoluble in water. Besides the ordinary fats, the tissues and liquids of the body contain phospha- tides, a group of compounds which stand in close relation to the fats, but differ in containing phosphoric acid and nitrogenous bases. The most important representative of this group is lecithin (C42H84NPO9), a fat-like compound which yields on decomposition, in addition to glycerin and a fatty acid, phosphoric acid and a nitrogen-containing substance called cholin (p. 366). Lecithin, though found in all cells, is especially abundant in nervous tissues. It is associated with choles- terin and with other substances which, like lecithin and cholesterin, are soluble in ether and similar solvents of fat. For this reason these substances are often grouped together as lipoids, although some of them are chemically different from fat. Cholesterin, for instance, is an alcohol. Although usually present only in small amount, the lipoids play a very important part in the structure and in the economy of the cell. Structure of Living Matter — The Cell.* — Bioplasm is the name given to the living matter of cells. The portion of the bioplasm differentiated as the nucleus is distinguished by the term karyo- plasm, and the portion outside the nucleus by the term protoplasm or cytoplasm. Protoplasm, when examined in its most primitive undifferentiated condition in such cells as the amceba or the white blood-corpuscles, appears on first view a homogeneous, structureless mass, except for certain granules embedded in it, and consisting either of products formed by its activity or of food materials. But even here more careful study reveals a certain complexity of struc- ture. At the very least, an external layer, or ectoplasm, can be dis- tinguished from the interior mass, or endoplasm. There is reason to believe that even where no histological demonstration of an ectoplasmic layer or a definite envelope is possible, the surface of the cell is physiologically different from its interior. In many cells the protoplasm presents the appearance of a honeycomb or net- work, with granules usually situated at the nodes, and holding in its vesicles or meshes a fluid, perhaps containing pabulum, from which the waste of the living framework is made good, or material upon which it works, and which it is its business to transform. Some observers, however, maintain that the network is an artificial appearance produced by the precipitation of the colloid constituents of the protoplasm by the fixing reagent, or even by the coagulative processes associated with the act of dying, and that the unaltered living substance is a homogeneous fluid or jelly. It is known that * Space permits only the slightest sketch of this subject here. For de- tailed information the student is referred to textbooks of histology. STRUCTURE OF LIVING MATTER 5 changes of reaction occur when the living substance dies, and slight changes of reaction, i.e., changes in the relative concentration of hydrogen ions (H+) and hydroxyl ions (OH-), can bring about similar precipitates in colloid solutions. Nevertheless in some cells a certain differentiation in the structure of the protoplasm can be seen during life and before the addition of any reagent, and in such cases there can be no doubt that the structural details pre-exist and are not arte-facts. In certain respects protoplasm behaves like a liquid, and in others like a solid, a peculiarity which is un- doubtedly associated with the fact that its chief constituents exist in the colloid state, as experiments with such substances as gelatin and agar have shown. In building up our typical cell we start with a piece of protoplasm. Somewhere in the midst of this we find a body which, if not absolutely different in kind from the protoplasm of the rest of the cell or cytoplasm, is yet marked off from it by very definite morphological and chemical characters. This is the nucleus, generally of round or oval shape, and bounded by an envelope. Within the envelope lies a second network of fine threads, which do not themselves stain with nuclear dyes such as hgematoxylin. But in or on these ' achromatic ' filaments lie small, highly refractive particles, staining readily and deeply with dyes, and therefore described as consisting of chromatin. This chro- matin is either made up of nucleins (conjugated proteins particu- larly rich in nucleic acid, and therefore in phosphorus), or yields nucleins by its decomposition; and it seems to owe its affinity for certain staining substances to the presence of nucleic acid. The meshes of the nuclear reticulum contain a semi-fluid material, which does not readily stain. The nucleus is distinguished from the cytoplasm, even as regards its inorganic constituents, by the absence of potassium.* Besides the nucleus, another much smaller structure, the centrosome, is differentiated from the protoplasm of many cells. This is a minute dot staining deeply with such dyes as heematoxylin, and generally situated near the nucleus. Sur- rounding it is a clear area, the attraction sphere, in and beyond which fine fibrils radiate out into the cytoplasm. Both the attrac - tion sphere and the nucleus play an important part in division of the cell by the process known as karyokinesis, or mitosis, or in- direct division, which is by far the most common mode. When the nucleus is about to divide, the chromatin granules arrange themselves into one or more coiled filaments or skeins, which then break up into a number of separate portions called * This has been shown microchemically. The potassium is precipitated by a solution of hexanitrite of sodium and cobalt as orange-yellow crystals of the triple salt, hexanitrite of potassium, sodium, and cobalt. Where very minute traces of potassium are present, ammonium sulphide must be added, after washing out the excess of the cobalt reagent. Black cobalt sulphide is thus formed from the triple salt (Macallum, Frontispiece). 6 INTRODUCTION chromosomes. These undergo a remarkable series of transforma- tions, leading eventually to the segregation of the nuclear chromatin in two separate daughter nuclei, each surrounded by a portion of the original cytoplasm. Apart from its role in the division, and therefore in the multiplication, of the cell, the nucleus is now known to exert an influence perhaps not less important upon those chemical changes in the cytoplasm which are necessary for its normal nutri- tion and function.* It is doubtful whether any portion of proto- plasm can permanently survive the loss of its nuclear material. It must be remembered, however, that nuclear material may some- times be present in diffuse form in cells which do not show a nucleus in the histological sense. When we carry back the analysis of an organized body as far as we can, we find that every organ of it is made up of cells, which upon the whole conform to the type we have been describing, although there are many differences in details. Some organisms there are, low down in the scale, whose whole activity is confined within the narrow limits of a single cell. The amoeba sets up in life as a cell split off from its parent. It divides in its turn, and each half is a complete amoeba. When we come a little higher than the amoeba, we find organisms which consist of several cells, and ' specialization of function ' begins to appear. Thus the hydra, the ' common fresh- water polyp ' of our ponds and marshes, has an outer set of cells, the ectoderm, and an inner set, the endoderm. Through the superficial portions of the former it learns what is going on in the world; by the contraction of their deeply placed processes it shapes its life to its environment. As we mount in the animal scale, specialization of structure and of function are found con- tinually advancing, and the various kinds of cells are grouped together into colonies or organs. In some organs and tissues the bond of union is simple juxtaposition and similarity of function of the constituent cells. But in others the union is protoplasmic, pro- cesses of the cytoplasm actually passing from cell to cell. This is seen in certain epithelial tissues, and conspicuously in the cardiac muscle. The Functions of Living Matter. — The peculiar functions of living matter as exhibited in the animal body will form the subject of the main portion of this book ; and we need only say here : (i) That in all living organisms certain chemical changes go on, the sum total of which constitutes the metabolism of the body. These may be divided into (a) integrative or anabolic changes, by which complex substances (including the living matter itself) are built up from * According to Hertwig, a precursor of chromatin, ' prochromatin,' a sub- stance without characteristic staining reaction, is formed in the cytoplasm, taken up by the nucleus, and there elaborated into chromatin. From the nucleus chromatin and its derivatives return to the cytoplasm to be used in its function. FUNCTIONS OF LIVING MATTER 7 simpler materials; and (b) disintegrative or katabolic changes, in which complex bodies (including the living substance) are broken down into comparatively simple products. In plants, upon the whole, it is integration which predominates; from substances so simple as the carbon dioxide of the air and the nitrates of the soil the plant builds up its carbo-hydrates and its proteins. In animals the main drift of the metabolic current is from the complex to the simple; no animal can construct its own protoplasm from the inorganic materials that lie around it; it must have ready-made protein in its food. But in all plants there is some disintegration; in all animals there is some synthesis. The progress of biochemistry in recent years has indeed shown that the synthetic powers of animal cells have been greatly underestimated. (2) The living sub- stance is excitable — that is, it responds to certain external im- pressions, or stimuli, by actions peculiar to each kind of cell. (3) The living substance reproduces itself. All the manifold activities included under these three heads have but one source, the trans- formation of the energy of the food. It is not, however, upon the whole, peculiarities in food, but in molecular structure, that underlie the peculiarities of function of different living cells. A locomotive is fed with coal; a steam-pump is fed with coal. The one carries the mail, and the other keeps a mine from being flooded. Wherein lies the difference of action ? Clearly in the build, the structure of the mechanism, which determines the manner in which energy shall be transformed within it, not in any difference in the source of the energy. So one animal cell, whi.-a it is stimulated, shortens or con- tracts; another, fed perhaps with the same food, selects certain constituents from the blood or lymph, and passes them through its substance, changing them, it may be, on the way ; and a third sets up impulses which, when transmitted to the other two, initiate the contraction or secretion. In the living body the cell is the machine ; the transformation of the energy of the food is the process which ' runs ' it. The structure and arrangement of cells and the steps by which energy is transformed within them sum up the whole of biology. PRACTICAL EXERCISES ON CHAPTER I. Reactions of Proteins. i. General Reactions of Proteins.— Egg-albumin may be taken as a type. Prepare a solution of it by adding water to white of egg, which consists mainly of egg-albumin with a little globulin. In breaking the egg, take care that none of the yolk gets mixed with the white. Snip the white up with scissors in a large capsule, then add ten or fifteen times its volume of distilled water. The solution becomes turbid from the precipitation of traces of globulin, since globulins are insoluble in distilled water. Stir thoroughly, strain through several layers of muslin, and then filter through paper. 8 INTRODUCTION Colour Reactions. (1) Add to a little of the solution in a test-tube a few drops of strong nitric acid. A precipitate is thrown down, which becomes yellow on boiling. Cool, and add strong ammonia; the colour changes to orange (xantho-proteic reaction). The reaction depends upon the presence of aromatic groups in the protein (in phenylalanin, tyrosin, tryptophane, oxytryptophane), which are converted into nitro-compounds. (2) To a third portion add a drop or two of very dilute cupric sulphate and excess of sodium or potassium hydroxide ; a violet colour appears (Piotrowski' s test). Peptones and proteoses (albumoses) give a pink (biuret reaction).* See p. 458. (3) To another portion add Millon's reagent ;f a white precipitate comes down, which is turned reddish on boiling. If only traces of protein are present, no precipitate is caused, but the liquid takes on a red tinge. The reaction is due to tyrosin. It is given by all aromatic substances which contain the group C6H6 with at least one H replaced by OH, i.e., the hydroxyphenyl group C6H4OH. (4) Adamkiewicz's Reaction (Hopkins' s modification). — To a small quantity of the albumin solution add the same bulk of dilute glyoxylic acid.J Mix, and to the mixture add an equal volume of strong pure sulphuric acid. A purple colour is obtained. The substance in the protein molecule which gives the reaction is tryptophane (p. 360). (5) The Formaldehyde Reaction. — Add to the albumin solution a few drops of a very dilute solution of formaldehyde (i : 2,500), and then allow some strong (commercial) sulphuric acid to run from a pipette into the bottom of the test-tube. A purple ring appears at the surface of contact. This reaction depends on the presence of tryptophane in the protein. Precipitation Rtactiontt (6) Acidify another portion strongly with acetic acid, and add a few drops of a solution of potassium ferrocyanide. A white precipitate is obtained. Peptones do not give this reaction. (7) Heat a portion to 30° C. on a water-bath. Saturate with crystals of ammonium sulphate; the albumin is precipitated. Filter, and test the filtrate for proteins by (2). None, or only slight traces, will be found. The sodium hydroxide must be added in more than sufficient quantity to decompose all the ammonium sulphate. It will be best to add a piece of the solid hydroxide. Peptones are not precipitated by ammonium sulphate, but all other proteins are. (8) Add alcohol to a small quantity of the solution. The protein is * The reaction is also given, although more faintly, with the hydroxides of lithium, strontium, and barium. It is given by all substances containing at least two CONH2 groups attached to one another (as in oxamide), or to the nitrogen atom (as in biuret) , or to the same carbon atom. f Millon's reagent consists of a mixture of the nitrates of mercury with nitric acid in excess, and some nitrous acid. To make it, dissolve mercury in its own weight of strong nitric acid, and add to the solution thus obtained twice its volume of water. Let it stand for a short time, and then decant the clear liquid, which is the reagent. J A solution containing glyoxylic acid in the requisite strength can be prepared by treating half a litre of a saturated solution of oxalic acid with 40 grammes of 2 per cent, sodium amalgam in a tall cylinder. When all the hydrogen has been evolved, the solution is filtered, and diluted with twice its volume of water. Oxalic acid and sodium binoxalate are also present in the solution. PRACTICAL EXERCISES g precipitated. It can be redissolved at first, but rapidly becomes in- soluble. 2. Special Reactions of Certain Proteins — (i) Heat-Coagulable Pro- teins : (a) Albumins. — (a) Heat a little of the solution of egg-albumin in a test-tube; it coagulates. With another sample determine the tem- perature of coagulation, first very slightly acidulating with a 2 per cent, solution of acetic acid. To determine the Temperature of Coagulation. — Support a beaker by a ring which just grips it at the rim. Nearly fill the beaker with water, and slide the ring on the stand till the lower part of the beaker is im- mersed in a small water-bath (a tin can will do quite well). In this beaker place a test-tube, and in the test-tube a thermometer, both sup- ported by rings or clamps attached to the same stand. Put into the test-tube at least enough of the albumin solution to completely cover the bulb of the thermometer, and heat the bath, stirring the water in the beaker occasionally with a feather or a splinter of wood, or a glass rod, the end of which is guarded with a piece of indiarubber tubing. Note the temperature at which the solution becomes turbid, and then the temperature at which a distinct coagulum or precipitate is formed. Repeat with the unacidulated albumin solution. (£) A similar experiment may be performed with serum-albumin obtained as on p. 65. (b) Globulins. — Use serum- globulin (p. 65), or myosinogen (p. 819). Fibrinogen is also a globulin, but cannot easily be obtained in quantity. Verify the following properties of globulins: (a) They coagulate on heating. (£) They are insoluble in distilled water (p. 65). (7) They are precipitated by saturation with magnesium sulphate or sodium chloride (p. 65). They give the general protein tests (i) to (8). Both the heat-coagulated proteins and such proteins as the solid fibrin which is formed from fibrinogen in the clotting of blood give such of the general protein tests, (i), (2), (3) (p. 8), as with suitable modifica- tions can be instituted on solid substances. Thus, in performing (2), a flake of fibrin or a small piece of the boiled egg-white should be soaked for a few minutes in a dilute solution of cupric sulphate. Then the excess of the cupric sulphate should be poured off, and sodium hydroxide added, when the coagulated protein will become violet. Heat-coagu- lated proteins are insoluble in water, weak acids and alkalies, and saline solutions ; fibrin is slightly soluble in the latter. (2) Gelatin. — Add some pieces of gelatin to cold water in a test-tube. It does not dissolve. Immerse the tube in a boiling water-bath till the gelatin goes into solution. Then cool the test-tube under the tap; the solution sets into a jelly. On heating it redissolves. Try the general protein reactions (p. 8) on a dilute solution. In Piotrowski's test a violet colour is obtained. The tests which depend on the presence of tyrosin or tryptophane are not given by a solution of pure gelatin, since these ammo-acids are absent from the gelatin molecule. Commercial gelatin may give a slight reaction due to traces of other proteins. 3. Reactions of Certain Derivatives of Native Proteins — (i) Meta- Proleins : (a) Acid-Albumin. — To a solution of egg-albumin add a little o-4 per cent, hydrochloric acid, and heat to about body temperature — say 40° C. — for a few minutes. Acid -albumin is formed. It can be produced from all albumins and globulins by the action of dilute acid. Make the following tests : (a) Add to a portion of the solution in a test-tube a few drops of a to INTRODUCTION solution of litmus; the colour becomes red. Now add drop by drop sodium carbonate or dilute sodium hydroxide solution till the tint just begins to change to blue. A precipitate of acid -albumin is thrown down . Add a little more of the alkali, and the precipitate is redissolved. It can be again brought down by neutralizing with acid. (/3) Heat a portion of the solution to boiling; no precipitate is formed . (y) Add strong nitric acid ; a precipitate appears, which dissolves on heating, and the liquid becomes yellow. (b) Alkali- albumin. — To a solution of egg-albumin add a little sodium hydroxide, and heat gently for a few minutes. Alkali-albumin is produced. It can be derived by similar treatment from any albumin or globulin. (a) Neutralize, after colouring with litmus solution, by the addition of dilute hydrochloric or acetic acid. Alkali-albumin is precipitated when neutralization has been reached. It is redissolved in excess of the acid. (/3) To another portion of the solution of alkali-albumin add a few drops of sodium phosphate solution, then litmus, and then dilute acid till the alkali-albumin is precipitated. More of the dilute acid should now be required to precipitate the alkali-albumin, since the sodium phosphate must first be changed into acid sodium phosphate. (y) On heating the solution of alkali-albumin there is no coagulation. (2) Proteoses. — For preparation and reactions, see p. 458. They differ from albumins and globulins in not being coagulated by heat, and from meta -proteins in not being precipitated by neutralization. They are soluble (with the exception of hetero-albumose) in distilled water, and are not precipitated by saturation of their solutions with mag- nesium sulphate or sodium chloride. Saturation with ammonium sul- phate precipitates them. With a solution of ' commercial peptone,' which consists chiefly of albumoses, and contains only a little true peptone, perform the following tests: (a) Boil the slightly acidulated solution; there is no coagulation. (/3) Biuret reaction, p. 8. (y) - (y) To a portion of the solution add its own volume of saturated ammonium sulphate solution. The primary albumoses (proto- and hetero-albumose) are precipitated. Filter. Add a drop of sulphuric acid to the filtrate and saturate it with ammonium sulphate crystals. The secondary or deutero-albumoses are precipitated. Filter. The filtrate stilljcontains peptones. Use it for (3). (3) Peptones. — For preparation and tests, see p. 459. They differ from heat-coagulable proteins and meta-proteins in the same way as proteoses, and they differ from proteoses in not being precipitated by ammonium sulphate. On the filtrate from (2) perform the biuret test, as described in (7), p. 8; and note that the pink colour is the same as that given by proteoses. Carbo-Hydrates. i. Glucose or Dextrose. — Make a solution of dextrose in water, and apply to it Trommer's test for reducing sugar. Put some of the dextrose solution in a test-tube, then a few drops of cupric sulphate, and then excess of sodium or potassium hydroxide. The blue precipitate of cupric hydroxide which is first thrown down is immediately dissolved in the presence of dextrose and many other organic substances. Now boil the blue liquid, and a yellow or red precipitate (cuprous hydroxide or oxide) is formed. PRACTICAL EXERCISES n 2. Cane-Sugar. — Perform Trommer's test with a sample of a solution. A blue liquid is obtained, which is not changed on boiling. Now put the rest of the solution in a flask. Add ^th of its bulk of strong hydro- chloric acid, and boil for a quarter of an hour. Again perform Trom- mer's test. Remember that excess of alkali must be present after the acid is neutralized. The test now shows much reducing sugar. The cane-sugar has been ' inverted ' — i.e., changed into a mixture of dextrose and levulose. 3. Starch. — (i) Cut a slice from a well-washed potato ; take a scraping from it with a knife, and examine with the microscope. Note the starch granules with their concentric markings, using a small diaphragm. Run a drop of dilute iodine solution under the cover-slip, and observe that the granules become bluish. Examine also with a polarization microscope. (2) Rub up a little starch in a mortar with cold water, then add boiling water and stir thoroughly. Decant into a capsule or beaker, and boil for a few minutes. After the liquid has cooled, perform the following experiments: (a) Add a few drops of iodine solution to a little of the thin starch mucilage in a test-tube. A blue colour is produced, which disappears on heating, returns on cooling, is bleached by the addition of a little sodium hydroxide, and restored by dilute acid. (b) Test the starch solution for reducing sugar by Trommer's test. If none is found, boil some of the mucilage with a little dilute sulphuric acid in a flask for twenty minutes, and again perform Trommer's test. Abundance of reducing sugar will now be present. 4. Dextrin. — Dissolve some dextrin in boiling water. Cool. Add iodine solution to a portion ; a reddish-brown (port- wine) colour results, which disappears on heating. As a control, the same amount of iodine should be added to an equal quantity of water in another test-tube. The colour returns on cooling. The colour is also bleached by alkali, restored by acid. Excess of iodine should be added for the bleaching experiment (i.e., more than enough to give the maximum depth of tint). If too little iodine has been added, there may be no restoration of the colour by the acid. The addition of a little more iodine to the acid solution will then cause the port-wine colour to return, and this may be again bleached by alkali, and will now be restored by acid. 5. Glycogen. — See p. 715. 6. Molisch 's Test for Carbo-Hydrates. — This is a general test for carbo- hydrates. It is also given by proteins which contain a carbo-hydrate group. Put a drop of dextrose solution in a test-tube. Add a drop of a 10 per cent, solution of a-naphthol in methyl alcohol, and then o'5 c.c. of water. Then cautiously allow I c.c. of pure concentrated sulphuric acid to run under the mixture, and shake gently. A violet or reddish colour appears. Fats. 1. Take a little lard or olive-oil, and observe that fat is soluble in ether or warm alcohol, but not in water. Put a drop of the ethereal solution of fat on a piece of paper, and note that it leaves a greasy stain. 2. Put a little alcohol in a test-tube, and then a drop of phenol- phthalein solution and a drop or two of dilute sodium hydroxide to give the solution a red colour. Add a few drops of an ethereal solution of the lard or olive-oil. If the red colour persists, the fat is neutral; if it disappears, the fat contains free fatty acids. 3. Saponification. — Melt some lard in a porcelain dish, and pour it 12 INTRODUCTION into an alcoholic solution of potassium hydroxide previously heated on a water-bath nearly to boiling. Mix well, and keep the mixture gently boiling on the bath till saponification is complete. This only takes a short time. Remove a little of the soap solution, and drop it into dis^ tilled water in a test-tube. If unsaponified fat is present, it will rise to the top as drops of oil. In this case boiling should be continued. If all the fat has been saponified, the soap solution will mix with the water and no oil-drops will separate. 4. Fatty Acids. — Heat some 20 per cent, sulphuric acid in a small flask nearly to boiling, and drop into it some of the soap obtained in 3. The fatty acids separate put and rise to the top as an oily layer. Cool, skim off the fatty acid, and wash it with distilled water till the wash- water is no longer acid. (a) Dissolve a little of the washed fatty acid in ether. Add a few drops of an alkaline solution of phenolphthalein to a few c.c. of water in a test-tube. Drop into this the ethereal solution of fatty acid. The red colour is discharged. (6) Put a small portion of the fatty acid on a glass slide resting on a piece of white paper. Place on it a drop or two of a i per cent, solution of osmic acid (osmium tetroxide). The osmic acid is reduced to a lower oxide (which is black) by the action of oleic acid present in the fatty acid mixture, which abstracts some of the oxygen. Any. fat which contains olein or oleic acid, as body-fat does, is therefore blackened by osmic acid. (c) Add to a portion of the fatty acid some sodium hydroxide solution, and warm. Sodium soap is formed. Add warm water and shake up. A lather is produced. Keep the soap solution for 6. Keep a little of the fatty acid for 5 (b) and 6 (b). 5. Glycerin. — (a) Add to a little glycerin in a dry test-tube a few crystals of potassium bisulphate (KHSO4), and heat over the free flame. Acrolein is given off, which is recognized by its pungent odour, and by blackening a piece of filter-paper moistened with ammoniacal silver nitrate solution, and held over the mouth of the test-tube. The paper is blackened owing to the reducing action of the vapour on the silver nitrate. (b) Repeat this test with lard, and with a portion of the fatty acid from 4. Acrolein will be given off by the lard because glycerin is con- tained in neutral fat, but not by the fatty acid if it has been properly separated from the glycerin. 6. Emulsification. — (a) Take three test-tubes and label them A, B, and C. Put a few c.c. of water in A, a solution of soap in B, and a dilute solution of sodium carbonate or sodium hydroxide in C. To each add a few drops of fresh olive -oil and shake . An emulsion will be formed in B, but not in A. Probably there will be some emulsification in C also, owing to the presence in the oil of some fatty acid, which forms soap with the alkali. But if the oil is free from fatty acid, no emulsion will be formed. (b) Repeat (a) with rancid olive-oil, which contains much fatty acid, or with fresh olive-oil to which some of the fatty acid obtained in 4 has been added. A good emulsion will be produced in C as well as in B. 7. Melting-Point of Fat. — Put into a very narrow test-tube or a short piece of narrow glass tubing some finely divided mutton fat, freed, as far as possible, from connective tissue. Fasten the test-tube on to the bulb of a thermometer with a rubber band, and immerse the ther- mometer and tube in a beaker filled with water and standing on a water- bath, which is gradually heated. Observe the temperature at which the fat melts. Repeat the experiment with hog's lard and dog's fat. PRACTICAL EXERCISES 13 SCHEME FOR TESTING A SOLUTION FOR THE MORE COMMON PROTEINS AND PROTEIN-DERIVATIVES, AND FOR CARBO- HYDRATES 1. Note the reaction, and whether the liquid is coloured or colourless, clear or opalescent. A reddish colour suggests blood ; opalescence suggests glyco- gen or starch. Try one or more of the general protein tests (e.g., the xantho- proteic or biuret). If the result is positive, proceed as in 2; if negative, pass to 3. 2. Test for Proteins. — (i) If the reaction is acid or alkaline, neutralize with very dilute sodium carbonate or sulphuric acid. A precipitate=acid- or alkali-albumin, according as the original reaction is acid or alkaline. If the original reaction is neutral, no acid- or alkali-albumin can be present in solution. Filter off the precipitate, if any. (2) Boil some of the nitrate from (i) (or some of the original solution if it is neutral), acidulating slightly with dilute acetic acid. A precipitate = albumin or globulin. Filter, and keep the nitrate. (3) If a precipitate has been obtained in (2), (a) saturate some of the original solution with magnesium sulphate, or half saturate it with ammonium sulphate (i.e., add to it an equal volume of saturated ammonium sulphate solution). If there is no precipitate, globulin is absent, and therefore the precipitate obtained in (2) must be albumin. A precipitate =globulin. But albumin may also be present in the solution. To see whether this is so, filter off the globulin and boil the nitrate after acidulation with acetic acid. A precipitate = albumin. (b) Half saturate the filtrate from (2) with ammonium sulphate (i.e., add its own volume of a saturated solution of the salt). A precipitate = primary proteoses. Filter. (c) Saturate the filtrate from (6) with ammonium sulphate crystals. A precipitate = secondary proteoses. Filter. (d) To the filtrate from (c) add excess of solid sodium hydroxide in small pieces at a time. Much ammonia is given off. Allow the test-tube to stand fifteen minutes, shaking it at intervals. Then add dilute cupric sulphate, and if much of the sodium sulphate formed remains undissolved, add water to dissolve it. A well-marked rose colour = peptone. (4) If no precipitate has been obtained in (2) , the solution contains neither albumin nor globulin. To test whether primary or secondary proteose or peptone is present, apply (3) (b), (c), and (d). 3. Test for Carbo- Hydrates. — Use the original solution, freed from coagu- lable proteins, if such have been found, by acidulation and boiling. (1) Add iodine. If the solution is alkaline neutralize it before adding the iodine. A blue colour = starch. Confirm by boiling with dilute sulphuric acid and testing for reducing sugar. A reddish-brown colour with iodine = glycogen or dextrin. Glycogen gives an opalescent, dextrin a clear, solution. Glycogen is pre- cipitated by basic lead acetate, dextrin is not (p. 715). Both are changed into reducing sugar by boiling with dilute acid. (2) Add to some of the original solution cupric sulphate and excess of sodium hydroxide, and boil. Yellow or red precipitate = reducing sugar. (3) If (i) and (2) are negative, boil some of the liquid with one-twentieth of its volume of strong hydrochloric acid for fifteen minutes, and test as in (2). A red or yellow precipitate indicates that a disaccharide like cane-sugar was originally present, and has been inverted. CHAPTER II THE CIRCULATING LIQUIDS OF THE BODY IN the living cells of the animal body chemical changes are con- stantly going on; energy, on the whole, is running down; complex substances are being broken up into simpler combinations. So long as life lasts, food must be brought to the tissues, and waste products carried away from them. In lowly forms like the amoeba these functions are performed by interchange at the surface of the animal without any special mechanism; but in all complex organ- isms they are the business of special liquids, which circulate in finely branching channels, and are brought into close relation at various parts of their course with absorbing organs, with eliminating organs, and with the tissue elements in general. In the higher animals three circulating liquids have been dis- tinguished: blood, lymph, and chyle. But it is to be remarked that chyle is only lymph derived from the walls of the alimentary canal, and therefore, during digestion, containing certain freshly- absorbed constituents of the food; while both ordinary lymph and chyle ultimately find their way into the blood, and are in their turn recruited from it. The blood contains at one time or another everything which is about to become part of the tissues, and every- thing which has ceased to belong to them. It is at once the scavenger and the food-provider of the cell. But no bloodvessel enters any cell;* and if we could unravel the complex mass of tissue elements which essentially constitute what we call an organ, we should see a sheet of cells, with capillaries in very close relation to them, but everywhere separated from them by a thin layer of lymph. And to describe in a word the circulation of the food substances we may say that the blood feeds the lymph, and the lymph feeds the cell. SECTION I. — MORPHOLOGY OF THE BLOOD. The blood consists essentially of a liquid part, the plasma, in which are suspended cellular elements, the corpuscles. When the circulation in a frog's web or lung or in the tail of a tadpole is * Fine intracellular canaliculi, communicating with the blood-capillaries, and probably performing a nutritive function, since they seem to contain blood -plasma, have been described by Schafer and others in the liver cells. 14 THE BLOOD-CORPUSCLES 15 examined under the microscope, the bloodvessels are seen to be crowded with oval bodies — of a yellowish tinge in a thin layer, but in thick layers crimson- — which move with varying velocity, now in single file, now jostling each other two or three abreast, as they are borne along in the axis of an apparently scanty stream of transparent liquid. Nearer the walls of the vessels, sometimes clinging to them for a little and then being washed away again, may be seen, especially as the blood-flow slackens, a few com- paratively small, round, colourless cells. The oval bodies are the red or coloured corpuscles, or erythrocytes ; the colourless elements are the white blood-corpuscles, or leucocytes; the liquid in which they float is the plasma (Practical Exercises, p. 193). The Red Blood-Corpuscles, or Erythrocytes, differ in shape and size and in other respects in different animal groups. In amphib- ians, such as the frog and the newt, they are flattened ellipsoids containing a nucleus, and the same is true of nearly all the other ver- tebrates, except mammals. In i- mammals they are discs, hollowed out on both the flat surfaces, or biconcave, and possess no nucleus. But the red corpuscles of the llama and the camel, although non- i , j 11- -j i • i Fig. i. — Diagram showing Relative nucleated, are ellipsoidal in shape, |ize of Re« corpuscles of Various like those of the lower vertebrates. Animals. As to size, the average diameter in man is between 7 and 8 fj,.* In the frog the long diameter is about 22 /j,, while in Proteus it is as much as 60 /j,, and in Amphiuma, the corpuscles of which can be seen with the naked eye, nearly 80 ft (Frontispiece) . As regards the structure of the red corpuscles, the most prob- able view is that they are solid bodies, with a spongy and elastic structureless framework, denser at the surface of the corpuscle than in its centre, but continuous throughout its whole mass (Rollett) . The denser peripheral layer constitutes a physiological envelope which permits the passage of certain substances into or out of the corpuscles, and hinders the passage of others. In the large oval corpuscles of Necturus (see Frontispiece) the envelope can be clearly demonstrated as a detachable membrane comparable to the mem- brane surrounding the nucleus. Envelope and spongework are sometimes spoken of as the stroma of the corpuscle, in contradistinction to its most important con- stituent, a highly complex pigment, the haemoglobin. This pigment is not in solution as such, for its solubility is not nearly great enough to permit this, but either in solution as a compound with * A micro-millimetre, represented bv symbol /*, is ^-^ millimetre. 16 THE CIRCULATING LIQUIDS OF THE BODY some other unknown substance, or more probably bound in some solid or semi-solid combination to the stroma, and filling up the space within the envelope in the interstices of the spongework. Since there is good reason to believe that the haemoglobin as obtained artificially from the corpuscles is not quite the same sub- stance as the native blood-pigment within them, the latter is some- times distinguished by a separate name — hsemochrome. To the physical properties of the stroma it is usual to attribute the great elasticity of the corpuscles — that is, the power of recovering their original shape after distortion — for their elasticity is in no wise impaired by the removal of the haemoglobin. Rouleaux Formation. — When blood with disc-shaped corpuscles is shed, there is a great tendency for the corpuscles to run together into groups resembling rouleaux, or piles of coin. No satisfactory explana- tion of this curious fact has yet been given. Crenation of the corpuscles, a condition in which they become studded with fine projections, is caused by the addition of moderately strong salt solution, by the passage of shocks of electricity at high potential, as from a Leyden jar, or by simple exposure to the air. Con- centrated saline solutions, which abstract water from the corpuscles and cause them to shrink, make the colour of blood a brighter red, because more light is now reflected from the crumpled surfaces. On the other hand, the addition of water renders the corpuscles spherical; more of the light passes through them, less is reflected, and the colour becomes dark crimson (Frontispiece). The White Blood-Corpuscles, or Leucocytes. — The red corpuscles are peculiar to blood. The white corpuscles may be looked upon as peripatetic portions of the mesoderm (see Chap. XIX.), and some of them ought not in strictness to be called blood-corpuscles. They are more truly body corpuscles. Similar cells are found in many situations, and wan- der everywhere in the spaces of the connec- tive tissue. They pass into the bloodvessels with the lymph, and may pass out of them again in virtue of their amoeboid power. They consist of proto- plasm, less differ- entiated than that of any other cells in the body, and under the microscope appear as granular, colour- less, transparent bodies, spherical in form when at rest, and containing a nucleus, often tri- or multi-lobed. Many of the leuco cytes of frog's blood at the ordinary temperature, and of mam- malian blood when artificially heated on the warm stage, may be A. D. Fig. 2. — Amoeboid Movement. A, B, C, D, succes- sive changes in the form of an amoeba. THE BLOOD-CORPUSCLES if seen to undergo slow changes of form. Processes called pseudo- podia are pushed out at one portion of the surface, retracted at another, and thus the corpuscle gradually moves or ' flows ' from place to place, and envelopes or eats up substances, such as grains of carmine, which come in its way. This kind of motion was first observed in the amoeba, and is therefore called amoeboid. It is perhaps due to local alterations of surface tension; at any rate, similar phenomena can be thus produced artificially. The leuco- cytes of human blood are not all of the same size, and differ also in other respects. They may be classified according to the presence or absence of granules in their protoplasm, and the fineness or coarseness of the granules ; according to the chemical nature of the dyes with which the granules most readily stain, and according to the form of the nucleus. Five or six varieties of leucocytes may thus be distinguished in normal blood (Frontispiece] : 1. Polymorphonuclear Neutrophile Cells. — The nucleus assumes a great variety of forms, often contorted or deeply lobed, the lobes being united by fine strands of chromatin. The cytoplasm contains numerous fine refractive granules, which stain best neither with simple acid dyes like eosin nor with simple basic dyes like methylene blue, but with mixtures which roust be assumed to contain ' neutral ' stains, like Ehrlich's so-called triacid stain.* These cells make up 65 to 75 per cent, of the total number of leucocytes. Their diameter is 10 to 12 p. 2. Eosinophile Cells (12 to 15 /x in diameter), much less numerous in normal blood than the neutrophiles (less than 5 per cent, of the whole), but found in considerable numbers in the serous cavities, the connec- tive tissue, and the bone-marrow. The granules in the cytoplasm are coarser than the neutrophile granules, and stain much more deeply with eosin. The nucleus may be simple, lobed, or even divided into fragments between which no connection can be traced. It is less rich in chromatin, and stains less easily with basic dyes, like methylene blue, than the nucleus of the first variety. 3. Large Mononuclear (also called Transitional) Leucocytes, with a diameter of 12 to 15 /*. They possess a large simple or slightly lobed nucleus, poor in chromatin, surrounded by a relatively great amount of cytoplasm, with faint neutrophile granules — i.e., granules which stain with neutral dyes. They constitute 3 to 5 per cent, of the total number of leucocytes. 4. Lymphocytes of Two Varieties — (a) Small Lymphocytes. — Smaller cells than any of the preceding (diameter 6 p.), possessing a single large nucleus, surrounded by a comparatively small amount of non-granular cytoplasm; 20 to 25 per cent, of the leucocytes of the blood belong to this group. The lymphocytes are markedly deficient in the power of amoeboid motion in comparison with the other varieties of colourless corpuscles. (b) Large Lymphocytes. — The largest of all the white cells of the blood, and at least twice as large as the small lymphocytes. They possess a relatively great proportion of cytoplasm, which is devoid of granules. They constitute no more than I per cent, of the total number of the colourless corpuscles. * A mixture of orange G., acid fuchsin, and methyl green. 2 t8 THE CIRCULATING LIQUIDS Of THE BOb\ 5. ' Mast Cells,' or ' Basophiles,' the least numerous variety (o-5 per cent, of the total number). Very few are to be found in the normal blood of adults, but more in children. They are somewhat smaller than the neutrophiles (average diameter about 10 p.). The nucleus is irregularly trilobed. The protoplasm shows coarse granules, which do not glitter like the granules of the eosinophile cells, and are therefore less conspicuous hi the unstained condition. Unlike the eosinophile granules, they stain with basic dyes, such as methylene blue. Blood-Plates, or Thrombocytes. — When blood is examined im- mediately after being shed, small colourless bodies (i to 3 // in diameter) of various shapes, but usually round or oval, may be seen. These are the blood-plates or platelets, also called thrombocytes, on account of their function in the coagulation of blood. If the blood is not at once subjected to some procedure which prevents clotting, the platelets swell and then break up. There is reason to believe that in most of the methods of preventing coagulation the essential action is to hinder the break-up of the platelets (p. 37). They can be isolated by receiving a drop of blood from the finger upon a well-cleaned cover-slip, which is then laid, supported by two thin glass fibres, on a carefully cleaned slide. The plasma with the coloured corpuscles and leucocytes are washed away by irrigating the space between slide and slip with a suitable solution, e.g., a salt solution containing a certain proportion of manganese sulphate, which prevents disintegration of the platelets. The platelets stick to the cover-slip (Deetjen). They can then be fixed and stained. The blood-plates can even, like leucocytes, be kept alive on the warm stage in an appropriate medium (agar, to which certain salts have been added), and then show lively amoeboid movements (Deetjen). They have been described as nucleated cells, although the nucleus is not easy to stain, and with the ultra-microscope, a delicate means of testing whether such an object as a platelet is optically homogeneous, no evidence of the presence of a nucleus has been obtained. The origin of the platelets has been a matter of lively controversy. They are not produced by the breaking up of other elements of the shed blood, for they have been observed within the freshly excised, and therefore still living, capillaries — in the mesentery of the guinea-pig and rat (Osier). According to the best evidence, they are derivatives neither of the erythro- cytes nor of the leucocytes of the blood, but are developed from special elements (so-called megakaryocytes) of the blood-forming organs (bone-marrow) (J. H. Wright). Enumeration of the Blood-Corpuscles. — This is done by taking a measured quantity of blood, diluting it to a known extent with a liquid which does not destroy the corpuscles, and counting the number in a given volume of the diluted blood (p. 67) . The average number of red corpuscles in a cubic millimetre of blood is about 5,000,000 in a healthy man, and about 4,500,000 in THE BLOOD-CORPUSCLES 1:4-10 a healthy woman, but a variation of 1,000,000 up or down can hardly be considered abnormal. In persons suffering from profound anaemia the number may sink to 1,000,000 per cubic millimetre, or even less. In one case of pernicious anaemia, only 143,000 corpuscles per cubic millimetre were present, the lowest number recorded. In new-born children the average is over 6,000,000, and in the inhabitants of high plateaus or mountains it may rise to 7,000,000 or Fig. s.-Curveshowingthe Number eyen more. In the latter instance a of Red Corpuscles at Different residence of a fortnight in the rarefied Ages (after Sorensen's Estima- a|r js sufficient to bring about the in- tions). The figures along the , , .j £ horizontal axis are years of age, crease> and a subsequent residence of a those along the vertical axis fortnight in the lowlands to annul it.* millions of corpuscles per cubic jn certain pathological conditions a millimetre of blood. , • • j.« » i« -L £. great increase in the relative number of corpuscles (polycythsemia) is found. Over 13,000,000 erythrocytes to the cubic millimetre have been counted in a case of cyanosis (im- perfect oxygenation of the blood, with blueness of the lips, etc.), due to congenital disease of the heart. An interesting form of experi- mental polycythsemia is caused by injection of adrenalin. The number of white blood-corpuscles is on the average about 10,000 per cubic millimetre of blood, or one leucocyte for every 500 red blood- corpuscles. But if the count is made when digestion is relatively inactive, four to five hours after a meal, it gives no more than 7,000 to the cubic milli- metre. In new-born children the average number is over 18,000 per cubic millimetre. The total leucocyte count, and still more the so-called dif- ferential count, i.e., the determination of the relative number of the different kinds of leucocytes, is often resorted to in the study of pathological conditions. A distinct increase in the number is designated leucocytosis. In leukaemia the number of white corpuscles is enormously increased — on the average to about 300,000, but in extreme cases to 600,000 per cubic millimetre — while at the same time the number * In 113 apparently healthy students (male) the average number of red corpuscles was 5,190,000 per cubic millimetre. In 104 of these, the number ranged from 4,000,000 to 6,400,000; in 71 (or 63 per cent, of the whole), from 4,400,000 to 5,500,000; in 3, from 3,500,000 to 3,900,000; in 5, from 6,500,000 to 7,000,000. In one observation the number reached 7,300,000. m Fig. 4. — Curve showing Propor- tion of White Corpuscles to Red at Different Times of the Day (after the Results of Hirt). At I the morning meal was taken ; at II the midday meal; at III the evening meal. During active digestion the number of lymphocytes in the blood is greatly increased, both abso- lutely and relatively to the number of the other leucocytes. 40 THE CIRCULATING LIQUIDS OF THE BODY of the red corpuscles is diminished; and the ratio of white to red may approach 1:4. As the anaemia rapidly advances towards the fatal termination of an acute case, and the erythrocyte count falls to 1,000,000, or even less, the ratio may come still nearer to unity. An increase in the number of leucocytes has also been observed in cer- tain infective diseases as part of the inflammatory reaction. There are also physiological variations, even within short periods of time ; for example, the number of lymphocytes is increased when digestion is going on (digestive lymphocytosis) . The normal number of blood-plates varies from a quarter to half a million to the cubic millimetre, but may be greater in disease and at high levels (Kemp). Life-History of the Corpuscles. — The corpuscles of the blood, like the body itself , fulfil the allotted round of life, and then die. They arise, perform their functions for a time, and disappear. But although the place and mode of their origin, the seat of their destruc- tion or decay, and the average length of their life, have been the subject of active research and still more active discussion for many years, much yet remains unsettled. Origin of the Erythrocytes. — In the embryo the red corpuscles, even of those forms (mammals) which have non-nucleated corpuscles in adult life, are at first possessed of nuclei, and approximately spherical in form. In the human foetus, at the fourth week all the red corpuscles are nucleated. Later on the nucleated corpuscles gradually diminish in number, and at birth they have almost or altogether disappeared, some of them, at least, having been converted by a shrivelling of the nucleus into the ordinary non-nucleated form. In the newly born rat, which comes into the world in a comparatively immature state, many of the red corpuscles may be seen to be still nucleated. The first cor- puscles formed in embryonic life are developed outside of the embryo altogether. Even before the heart has as yet begun to beat, certain cells of the mesoderm (see Chapter XIX.) in a zone (' vascular area ') around the growing embryo begin to sprout into long, anastomosing processes, which afterwards become hollowed out to form capillary bloodvessels. At the same time clumps of nuclei, formed by division of the original nuclei of the cells, gather at the nodes of the network. Around each nucleus clings a little lump of protoplasm, which soon develops haemoglobin in its substance; and the new-made corpuscles float away within the new-made vessels, where they rapidly multiply by mitosis. In later embryonic life the nucleated corpuscles continue in part to be developed within the bloodvessels in the liver, allantois, spleen, and red bone-marrow, and in certain localities in the connective tissue, by mitotic division of previously existing nucleated corpuscles, in part to be formed endogenously within special cells in the liver and perhaps other organs. Still later the nucleated corpuscles give place in the blood of the mammal to non-nucleated erythrocytes. Many of these are doubtless derived from the nucleated corpuscles, but some appear to be produced in the interior of certain cells of the connective tissue, and are non-nucleated from the start. In the mammal in extra-uterine life the chief seat of formation of the red blood-corpuscles, or haematopoiesis, is the red marrow of THE BLOOD-CORPUSCLES 21 the bones of the skull and trunk, and of the ends of the long bones of the limbs. Special nucleated cells in the marrow, originally colourless, multiply by karyokinesis, take up haemoglobin or, what is much more likely, form it within their protoplasm, and are transformed by various stages into the ordinary non-nucleated red corpuscles, which then pass into the blood-stream. These blood- forming cells have received the name of erythroblasts or haemato- blasts. According to their size, erythroblasts have been distinguished as normoblasts, megaloblasts, and microblasts. The normo- blasts are most numerous, and have about the same diameter as the full-formed erythrocytes, into which they are believed to develop. The megaloblasts are larger, and the microblasts smaller, and they are thought to be the precursors of those aberrant forms of erythrocytes sometimes found in the blood in certain diseases. After haemorrhage rapid regeneration of the blood takes place, so that in a few weeks the loss of even as much as a third of the total blood is made good. The plasma is much sooner restored to its normal amount than the corpuscles. Microscopical examination shows in the red marrow the tokens of increased production of coloured corpuscles, and nucleated erythrocytes appear in the blood, the normoblasts being, as it were, hurried into the circula- tion before the transformation which normally results in the dis- appearance of the nucleus is complete. The same is true in severe pathological anaemias, e.g., pernicious anaemia. It is a matter of interest that other organs also, which in embryonic life perform a haematopoietic function, particularly the spleen, may, in such emergencies, again take on the office of forming blood-corpuscles. A constant destruction of red blood- corpuscles must go on, for the bile-pigment and the pigments of the urine are derived from blood-pigment. The bile-pigment is formed in the liver. It con- tains no iron ; but the liver cells are rich in iron, and on treatment with hydrochloric acid and potassium ferrocyanide, a section of liver is coloured by Prussian blue. Iron must therefore be removed by the liver from the blood-pigment or from one of its derivatives; and there is other evidence that the liver is either one of the places in which red corpuscles are actually destroyed, or receives blood charged with the products of their destruction. Although it cannot be doubted that in all animals whose blood contains haemoglobin the iron found in the liver bears an important relation to the building up or breaking down of the blood-pigment, the injection of haemoglobin or haemin, indeed, increasing markedly the amount of iron in the liver, as well as in the spleen, bone-marrow and other tissues, this does not seem to be the only function of the hepatic iron, for the liver of the crayfish and the lobster, which have no haemoglobin in their blood, is rich in iron. Destruction of 22 THE CIRCULATING LIQUIDS OF THE BODY erythrocytes may also take place in the spleen and bone-marrow. Although the statement that free blood-pigment exists in demon- strable amount in the plasma of the splenic vein is incorrect, red corpuscles have been seen in various stages of decomposition within large amoeboid cells in the splenic pulp; and deposits containing iron have been found there and in the red bone-marrow in certain pathological conditions. But there is no good foundation for the statement sometimes rather fancifully made that the spleen is in any special sense the ' graveyard of the red corpuscles.' Some of the coloured corpuscles may break up in the blood itself, forming granules of pigment, which may then be taken up by the liver, spleen, and lymph glands. Indeed, it is probable that a large proportion of the worn-out erythrocytes are finally destroyed in the blood- stream. The portal circulation may be more than other vascular tracts a seat of this natural decay, perhaps in virtue of the presence of substances with a hsemolytic action (p. 28) absorbed from the alimentary canal. It has been argued that the erythrocytes must be short-lived, since they are devoid of nuclei (p. 6), and attempts have been made to calculate the average time for which they survive in the circulation from the amount of haemoglobin (or of its derivative, haematin) required to furnish the daily excretion of bile-pigment. The results arrived at, however, are not sufficiently trustworthy to warrant their citation. Origin and Fate of the Leucocytes. — There has been much dis- cussion as to the origin of the white blood-corpuscles. The numerous theories fall into two groups, which have been designated somewhat pompously the monistic and the dualistic. According to the first, all the colourless corpuscles arise from a single type of parent cell, namely, the lymphocyte type, in its small or large variety. According to the dualistic school, a fundamental distinc- tion exists between the lymphocytes, or cells peculiar to lymphoid tissues, and to the blood on the one hand, and the remaining varieties of leucocytes on the other. The former are supposed to be derived from the lymphoblasts of lymphoid tissue, and the latter from the myeloblasts, the forerunners of the myelocytes of bone-marrow. The question has recently been studied by Foot by a new method, namely, by cultivating chicken marrow outside of the body, and watching the transformation of certain of its cells. He concludes in favour of the development of the polymorphonuclear leucocyte from a lymphoid type of cell existing in the marrow, a conclusion in harmony with the monistic view. As regards their immediate source, the small lymphocytes of the blood are undoubtedly derived from the lymph, and are identical with the lymph-corpuscles. That they are formed largely in the lymphatic glands is shown by the fact that the lymph coming to the glands is much poorer THE BLOOD-CORPUSCLES 23 in corpuscles than that which leaves them. The lymphatic glands, however, although the principal, are not the only seat of formation of lymphocytes, for lymph contains some corpuscles before it has passed through any gland ; and although a certain number of these may have found their way by diapedesis from the blood, others are developed in the diffuse adenoid tissue, or in special collections of it, such as the thymus, the tonsils, the Peyer's patches and solitary follicles of the intestine, and the splenic corpuscles. To a very small extent white blood-corpuscles may multiply by karyokinesis or indirect division in the blood. The fate of the leucocytes is even less known than that of the red corpuscles, for they contain no characteristic substance, like the blood-pigment, by which their destruction may be traced. That they are constantly disappearing is certain, for they are constantly being produced. Not a few of them actually escape from the mucous membranes of the respiratory, digestive, and urinary tracts. The remnants of broken-down leucocytes have been found in the spleen and lymph glands. It must be assumed that many break up in the blood-plasma itself. SECTION II. — GENERAL PHYSICAL AND CHEMICAL PROPERTIES OF THE BLOOD. Fresh blood varies in colour, from scarlet in the arteries to purple-red in the veins. It is a somewhat viscid liquid, with a saline taste and a peculiar odour. Viscosity of Blood. — The viscosity of normal dog's blood is about six times greater than that of distilled water at body temperature. It can be determined by allowing the blood to flow through a capil- lary tube of known dimensions under a definite pressure, and measuring the amount which escapes in a given time. In general the viscosity and specific gravity of the blood vary in the same direction, although there is not an exact proportionality between them. Thus, sweating, which causes a diminution of the water of the blood, causes also an increase in its viscosity. With increasing temperature the viscosity of the blood diminishes, as is the case with other liquids (Burton-Opitz). In polycythsemia, where the number of erythrocytes in propor- tion to plasma is greatly increased, the viscosity of the blood in- creases in an equal degree. In one case of polycythsemia, with a blood-count of 8,300,000, the viscosity was 9-4 times that of water; in a case of marked chlorosis it was only 2-14. But the importance of this factor in causing an abnormal blood-pressure by increasing or diminishing the resistance to the blood-flow has been exaggerated. Although it has been shown that in the living vessels, so long as 24 THE CIRCULATING LIQUIDS OF THE BODY their calibre remains constant, the flow is affected by changes in the viscosity of the blood, just as in glass tubes, compensation by adjustment of the vascular calibre is so ample and so easy thai even the greatest alterations of viscosity produce little effect on the mean blood-pressure. Reaction of Blood. — In the sense in which the term is used in physical chemistry, the reaction of a solution depends on the pro- portion between its content of hydrogen (H + ) and hydroxyl (OH — ) ions, an excess of hydrogen ions corresponding to an acid and an excess of hydroxyl ions to an alkaline reaction. It has been shown by a physical method (the determination of the electro- motive force of a cell containing blood or serum as one liquid) that hydroxyl ions are present only in small excess, and that blood is really but a little more alkaline than distilled water. Practically, it may be regarded as a neutral liquid. Under a great variety of conditions, physiological and pathological, its reaction remains almost unchanged. Yet it is known that acids (carbon dioxide, lactic, phosphoric, and sulphuric acids) are constantly being pro- duced in the normal metabolism of the tissues. The administra- tion of large quantities of acid or alkali causes a surprisingly small effect. In diabetes, even when it can be proved that an abnormal production of acid substances is taking place, the blood shows little, if any, diminution in the proportion of hydroxyl ions; it remains to all intents and purposes a neutral liquid. In diabetic coma, where the blood may in extreme cases turn blue litmus red, the true reaction is only slightly altered. The manner in which the reaction of the blood, the tissue liquids, and probably the protoplasm itself, is regulated within such narrow limits is a subject of great interest. For there is reason to believe that it is of the utmost moment that the equilibrium should be maintained not only in order that the functions of the tissues may be properly performed, but that danger to life may be averted. To be sure, the excretory organs, the lungs and the kidneys, provide the means by which the excess of acid (or of alkali) is finally, under normal circumstances, eliminated. Other regulative mechanisms also exist. For example, it has been shown that when an excessive production of acids (acidosis) occurs in conditions of disordered metabolism, or when acids are purposely administered in large amount, a greater quantity of ammonia, split off from the pro- teins, is mobilized to aid in neutralizing the acids. But very simple experiments on blood in vitro are sufficient to show that the blood itself has a great capacity, as compared with water, to resist a change in its reaction even when large amounts of acid or alkali are added to it. The secret of the reaction-regulating power lies, therefore, to a large extent in the blood itself. Two factors have been shown to be of importance: (i) The power of the proteins, GENERAL PHYSICAL AND CHEMICAL PROPERTIES 25 in virtue of their amphoteric character, to combine either with acids or with bases, so that, when excess of base is added to blood, the proteins act as acids, and neutralize the base; when excess of acid is added, the proteins act as bases, and neutralize the acid. (2) The equilibrium of certain of the inorganic constituents of the blood (carbon dioxide, the carbonates, and the phosphates) is such that even great variations in the concentration of any of these, such as may normally occur, produce scarcely any effect upon the concentrations of the hydrogen and hydroxyl ions. Thus, when phosphoric acid and sodium hydroxide are added to water in certain proportions, and the solution placed under a certain tension of carbon dioxide (which is kept constant), we get a more or less accurate imitation of blood as regards the inorganic substances concerned in the regulation of its reaction, sodium bicarbonate (NaHCO3) and disodium phosphate (NagHPC^) being present in the solution as in blood. It is found that when the quantities are so chosen that the H + concentration lies within the limits of variation of the normal blood reaction, relatively large quantities of alkalies can be added or withdrawn without causing much change in the H + concen- tration. It can be shown both theoretically and experimentally that precisely those weak acids present in blood (CO2, NaH2PO4) require the largest addition of alkali to alter the reaction to a given extent, and are therefore particularly suited to give stability to the reaction. Thus carbon dioxide requires twenty-four times, and monosodium phosphate thirty-three times, as much alkali as an equivalent solution of acetic acid to cause a given alteration of colour in rosolic acid (E. Henderson). The so-called ' titratable ' alkalinity of blood or serum, measured by the amount of standard acid which must be added before the colour of the indicator used changes from alkaline to acid, bears no necessary or fixed proportion to the actual alkalinity. When blood, for instance, is titrated with hydrochloric acid, with methyl orange as indicator, at the point where the red colour appears all the disodium phosphate and sodium bicarbonate will have been changed into monosodium phos- phate and carbon dioxide, all the alkali removed from combination with proteins, a certain amount of acid -protein compounds formed, and other minor reactions produced (Henderson). It is difficult to correlate the quantity deduced from such a titration with any physio- logical condition, although undoubtedly it bears some relation to the acid-neutralizing power of the blood, and some relation to its real reaction. Still, by titration information of value can be obtained which is not yielded by the physico-chemical method in regard to the potential ' acid capacity ' of the blood and its power of resistance against acid-poisoning. What is estimated here is the quantity of acid required to satisfy the proteins and to react with the carbonates and phosphates before that concentration of hydrogen and hydroxyl ions just necessary to cause the change of colour is established. This is not the same for different indicators, since there is a certain minimum ratio in the concentration of these ions at which each indicator turns in one or the other direction, none turning precisely at the neutral point. Thus serum appears to be acid when tested with phenolphthalein, and alkali must be added to the serum before the pink colour indicating alkalinity is produced. On the other hand, with litmus or methyl orange it gives the alkaline 26 THE CIRCULATING LIQUIDS OF THE BODY reaction, and a considerable amount of acid must be added before the colour of the indicator which denotes acidity appears. The true re- action of the serum is not, of course, at one and the same time both alkaline and acid; but it is so near neutrality that it falls just below the degree of alkalinity necessary to give the pink colour with phenol- phthalein, and just below the degree of acidity which gives the pink colour corresponding to an acid reaction with methyl orange. Certain indicators — for example, rosolic acid — turn so as to give sharp colour reactions at about the concentration of hydrogen and hydroxyl ions in the blood, and these may possibly be of use in determining the changes in the true reaction for clinical purposes (Adler) . More closely related to the true alkalinity of the blood than the titratable alkalinity is the carbon dioxide content. The estimation of the total carbon dioxide in a sample of blood throws light upon the capacity of the blood to perform one of its most important functions — the transportation of carbon dioxide — and to preserve one of its essential properties — an almost neutral reaction — in the presence of an excessive intake or production of acid substances. In herbivorous animals the carbon dioxide content of the blood is easily lessened by the administration of acids, but in carnivora and in man it is much more difficult to bring about such a decided effect, for the reason already mentioned, the acid being neutralized by ammonia. In many diseases, however, and particularly in those accompanied by fever, this protective mechanism breaks down. Specific Gravity of Blood. — The average specific gravity of blood is about 1066 at birth. It falls during infancy to about 1050 in the third year, then rises till puberty is reached to about 1058 in males (at the seventeenth year), and 1055 in females (at the four- teenth year). It remains at this level during middle life in males, but falls somewhat in females. In chlorotic anaemia of young women it may be as low as 1030 or 1035. It rises in starvation. Sleep and regular exercise increase it (Lloyd Jones).* The specific gravity of the serum or plasma varies from 1026 to 1032. The Electrical Conductivity of Blood.— The liquid portion of the blood conducts the current entirely by means of the electrolytes dissolved in it, the most important of these being the inorganic salts; and the conductivity of the serum varies, in different speci- mens of blood, within a comparatively narrow range. The con- ductivity of entire (defibrinated) blood, on the contrary, varies within wide limits. For instance, in a case of pernicious anaemia the conductivity of the blood was found to be almost double that of normal human blood, while the conductivity of the serum was normal. The most influential factor which governs this variation * In 165 students (male) the average specific gravity of the blood, as deter- mined by Hammerschlag's method (p. 62) was 1054-4. In 149 of these the variation was from 1050 to 1065; in 94 (or 57 per cent, of the whole), from 1054 to 1060; in 4, from 1046 to 1049; in 9, from 1066 to 1070. In 3 the specific gravity was only 1040 to 1042. GENERAL PHYSICAL AND CHEMICAL PROPERTIES 27 is the relative volume of the corpuscles and serum. When the blood is relatively rich in corpuscles and poor in serum, its con- ductivity is low; when it is poor in corpuscles and rich in serum, its conductivity is high. The explanation is that the corpuscle refuses passage to the ions of the dissociated molecules, which, in virtue of their electrical charges, render a liquid like blood a con- ductor (p. 428), or permits them only to pass very slowly, so that the intact red corpuscles have an electrical conductivity so many times less than that of serum, that they may, in comparison, be looked upon as non-conductors (Practical Exercises, p. 69). The Relative Volume of Corpuscles and Plasma in Unclotted Blood, or, what can be converted' into this by a small correction, the relative volume of corpuscles and serum in defibrinated blood, can be easily determined, with approximate accuracy, by com- paring the electrical conductivity of entire blood with that of its serum.* Another method, more suitable for clinical work, though not so accurate, is the so-called hgematocrite method. A small quantity of blood is centrifugalized in a graduated glass tube of narrow bore until the corpuscles have been collected into a solid ' thread ' at the outer extremity of the tube. Their volume and that of the clear plasma which has been separated from them are then read off on the scale. The hsematocrite must rotate at such a high speed (10,000 turns a minute) that separation of the corpuscles from the plasma is accomplished before clotting has occurred. Dilution of the blood with liquids which prevent clotting is not permissible for exact work (Practical Exercises, p. 68). By these and other methods too elaborate for description here, it has been shown that the plasma or serum usually makes up rather less than two-thirds, and the corpuscles rather more than one-third, of the blood. But this proportion is, of course, liable to the same varia- tions as the number of corpuscles in a cubic millimetre of blood. It depends, further, the number of corpuscles being given, on the average volume of each corpuscle. For instance, when the mole- cular concentration, and therefore the osmotic pressure (p. 427), of the plasma is reduced, as by the addition of water or the abstrac- tion of salts, water passes into the corpuscles and they swell; when the molecular concentration of the plasma is increased, by the abstraction of water or the addition of salts, water passes out of the corpuscles, and they shrink. In human serum the average * The formula p *» «rr (174 - K(b)), where p is the number of c.c. of serum in 100 c.c. of blood; K(b], K(s), the specific conductivities respectively of the blood and serum (both measured at or reduced to 5° C., and, to obtain whole numbers, multiplied by io4), may be used in the calculation. K is the specific conductivity of the liquid — i.e., the conductivity of a cube of the liquid of i centimetre side. The conductivity 01 a similar cube of mercury is 10,630. 28 THE CIRCULATING LIQUIDS OF THE BODY depression of the freezing-point below that of distilled water, which is a measure of the molecular concentration and of the osmotic pressure, is about 0-56° C. (Practical Exercises, p. 73). For clinical purposes, the determination of the relative volume of corpuscles and plasma is most useful in cases where the average size of the erythro- cytes departs from the normal, and where, accordingly, the enumera- tion of the corpuscles would give an erroneous idea of their total mass. Laking of Blood, or Haemolysis. — Even in thin layers blood is opaque, owing to reflection of the light by the red corpuscles. It becomes transparent or ' laky ' when by any means the pigment is brought out of the corpuscles and goes into true solution. Re- peated freezing and thawing of the blood, the addition of water, the passage of electrical currents, constant and induced,* putre- faction, heating the blood to 60° C., and many chemical agents (as bile-salts, ether, saponin), cause this change. Certain complex poisons of animal origin, such as snake-venoms, bee-poison, spider- poison or arachnolysin, and certain toxins produced by pathogenic bacteria — for instance, tetanolysin, formed by the tetanus bacillus — also possess decided haemolytic power. The blood-serum of certain animals acts on the coloured corpuscles of others, and sets free their pigment — for example, the serum of the dog or ox causes haemolysis of rabbit's corpuscles; the serum of the ox, goat, dog, or rabbit lakes guinea-pig's corpuscles. But rabbit's serum does not lake dog's corpuscles, and guinea-pig's serum is inactive towards the corpuscles both of the rabbit and the dog. It has been shown that in hsemolysis by foreign serum two bodies are concerned: one, which is easily destroyed by heating to about 56° C., the so-called complement, and another, the intermediary body or amboceptor, which is not affected by being heated to this temperature. Thus if dog's serum be heated to 56° C. for twenty minutes, no amount of it will lake rabbit's washed corpuscles — that is, rabbit's corpuscles freed from their own serum by repeated washing with salt solution and centrifugalization. If, however, serum which is not itself hsemolytic for rabbit's blood (e.g., rabbit's or guinea-pig's serum) be added to the washed rabbit's corpuscles, they will be laked by the heated dog's serum. Unheated dog's serum will lake rabbit's corpuscles, whether they have been washed free from their own serum or not (Practical Exercises, p. 71). The hypothesis which best explains these facts and many similar ones is that dog's serum contains both of the bodies necessary for haemolysis of rabbit's corpuscles. When the complement has been rendered inactive by heating, the amboceptor cannot cause laking * The laking action of induced currents is due simply to the heating of the blood. Condenser discharges, which cause liberation of the haemoglobin without raising the temperature of the blood as a whole to the point at which heat-laking occurs, possibly act in the same way by causing local heating of the corpuscles owing to their high resistance. GENERAL PHYSICAL AND CHEMICAL PROPERTIES 29 by itself. Rabbit's serum contains complement, but not the specific amboceptor necessary for the laking of rabbit's corpuscles. Accordingly, the addition of fresh rabbit's serum to heated dog's serum restores complement to the latter, and thus it is again ren- dered active for rabbit's corpuscles. The amboceptor is supposed to unite on the one hand with certain groups in the corpuscle and on the other with the complement, which is thus enabled to develop its hsemolytic action upon the envelope or the stroma. The com- plement is incapable of acting, even in the presence of amboceptor, if the temperature is reduced to o° C. Nevertheless, the corpuscles take up amboceptor at this temperature, and on this fact is based a method of freeing serum from amboceptor. For example, if dog's serum and excess of rabbit's washed corpuscles, both pre- viously cooled to o° C., be mixed and placed at o° C. for some hours, and the serum then removed, it will be found that it has lost the power of laking rabbit's corpuscles, washed or unwashed, at air or body temperature, although it will still do so on the addition of dog's serum in which the complement has been destroyed by heating it to 56° C. The real nature and mode of action of com- plements and amboceptors are not yet satisfactorily determined. The laws of chemical equivalents and definite proportions do not seem to be observed in the reactions into which they enter. It has therefore been suspected that the bodies in question belong to the group of ferments, or are closely related thereto, and there is some evidence that a fat-splitting enzyme, or lipase, is concerned in the complement action (Jobling). As to the manner in which haeraolytic agents cause the liberation of the blood-pigment, the fact that in so many forms of laking the cor- puscles swell up before the haemoglobin escapes indicates that the entrance of water is an important step. The entrance of water is favoured by changes produced in the chemical and physical condition of certain constituents of the superficial layer (envelope) of the cor- puscle, as well as by changes in its interior. Saponin and ether, for example, are known to be solvents of cholesterin and lecithin, and cholesterin and lecithin are important constituents of the stroma and envelope of the erythrocyte. It is easy to understand that if a portion of one or both of these substances is dissolved, or altered without being actually dissolved, profound changes may be produced in the permea- bility of the corpuscle to water and to the salts dissolved in the liquid in which the erythrocytes are suspended. In addition to this change of permeability, many laking agents, perhaps all, exert also a more direct influence on the normal relations of the native blood-pigment to the stroma. Ether and saponin, for instance, seem to act in two ways — by disorganizing the envelope through solution of its lipoids, and thus increasing its permeability to water; and by helping to dissociate the blood-pigment-stroma complex by exerting a pull on the lipoids of the stroma, while the water simultaneously exerts a pull on the pigment. The conclusion follows from this view of haemolysis, that the erythro- cytes, normally so perfectly adapted to the plasma in which they float, may, when the conditions on which th'eir equilibrium with it depends 30 THE CIRCULATING LIQUIDS OF THE BODY are altered, be rapidly and inevitably destroyed by that very plasma itself. It is, indeed, the very fact of the exquisite adaptation of liquid and cell for a strictly regulated exchange of material which constitutes the danger when the regulation is upset. A liquid like mercury, which is not adapted either to give anything to erythrocytes in contact with it or to take anything from them, would not cause haemolysis, even if the permeability of the corpuscles for water or sodium chloride were increased to any extent. The continued survival of the erythrocytes in an aqueous solution of salts and proteins like the plasma — nay, more, the protection of the corpuscles up to a certain point by the plasma against the attack of extraneous haemolytic agents — are facts we are prone to take so much for granted as to forget that they depend entirely upon a most delicate adjustment of the permeability of the corpuscles for essential constituents of the plasma. Disturb these relations to a sufficient degree, and the plasma becomes a poison to the erythrocytes not much less deadly than distilled water. When we add to blood a haemolytic substance, and see that presently the blood-pigment has left the corpuscles, we are apt to attribute the whole effect to the foreign material added, and to say that the saponin, the ether, the alien serum, has laked the blood. In a certain sense this is true, but it is not the whole truth. In reality the haemolytic agent has acted in an essential degree, although nob exclusively, by overthrow- ing the equilibrium between the corpuscles and the aqueous solution of certain substances in which they are suspended. To say that the foreign substance alone causes the haemolysis is no more accurate than it would be to say that a man swimming strongly in a rough sea, who sinks when hit and stunned by a piece of wreckage, was drowned by the blow, and not by the sea. No doubt it is true that, but for the blow, he would have continued to swim ; yet, in reality, he loses his life because he is environed by a medium deadly to him as soon as his power of adjustment to it has been too much diminished. On land, the blow would have stunned, but would not have killed him. In like manner, to glance at one phase of the natural decay of the corpuscles within the body, an erythrocyte may float secure in its watery environment through many rounds of the circulation. But its security is not static, like that of a log floating on the water. It is dynamic, a triumph of perfect physico-chemical poise, as the security of the swimmer, still more of the tight-rope dancer, is dynamic, a triumph of perfect neuro- muscular poise. The time, however, arrives when, either through changes in the corpuscle itself (the changes of cellular senility, as we may call them), or through changes in the environing medium, or through a combination of the two, the adjustment is upset, and the erythrocyte is now destroyed by the plasma in which it has so long lived. In general haemolysis by foreign serum is preceded by agglutina- tion or aggregation of the corpuscles into groups. Agglutination may be obtained without haemolysis by heating the haemolytic serum to the temperature at which the complement is destroyed, since the agglutinating agents, or agglutinins, are relatively resistant to heat. Besides the amboceptors naturally present in the blood of certain animals, and capable, in conjunction with complement, of haemolyzing the corpuscles of certain other animals, amboceptors may be produced in much greater strength by artificial means. GENERAL PHyslCAL AMD CHEMICAL PROPERTIES 31 When the corpuscles of one animal are injected intraperitoneally or subcutaneously into an animal of a different kind, the serum of the latter acquires the property of agglutinating and laking the corpuscles of an animal of the same kind as that whose corpuscles have been injected. This is especially marked if the injection is several times repeated at intervals of a few days. If, for instance, dog's corpuscles are injected into a rabbit, the rabbit's serum after a time becomes strongly hsemolytic for dog's corpuscles. It also agglutinates them. This is due to the appearance in the rabbit's serum of an amboceptor and an agglutinin which have a specific action on dog's corpuscles. Such a serum is often termed an immune serum, and the animal which has received the injections is spoken of as immunized in regard to this particular kind of corpuscles. For the reaction involved in the production of the amboceptor and agglutinin is a particular case of the peculiar and specific response which the body makes to the presence of foreign juices or cells, including bacteria, and which constitutes an attempt to render itself ' immune ' to them. Many other animal cells besides the coloured blood-corpuscles give rise, when injected, to similar specific substances (cytolysins), which cause destruction of cells of the same kind— e.g., leucocytes and spermatozoa.* The process of haemolysis is more easily followed than the cytolysis of ordinary cells. Yet in its main features it is essentially similar. In each case the specific antibody seems to be produced in response to the presence of some particular constituent of the foreign cell. The substances which on injection give rise to antibodies are spoken of as antigens. In the case of the erythrocytes there is evidence that the antigens (both the haemolysinogen, which causes the production of specific amboceptor, and the agglutininogen, the substance which gives rise to specific agglutinin) are lipoids, or are so closely associated with the lipoids of the corpuscles that they are extracted by the same solvents. Thus ethereal extracts of erythrocytes cause the production of haemol- ysin and agglutinin, just as the entire corpuscles do. The group of antibodies known as precipitins is of special interest. Precipitins. — When the serum of one animal is injected into another of a different group, the serum of the latter acquires the property of causing a precipitate in the normal serum of animals of the same group as that whose serum was injected, but not * Recent studies have tended to modify the view that the cytotoxins formed after the introduction of different foreign tissues into animals are quite specific for each tissue. Thus Lambert, using tissues cultivated on media outside of the body for testing the toxic action, finds that the plasma of guinea-pigs which have received injections of either chick embryo heart or intestine becomes toxic for both of these tissues. In like manner the plasma of guinea-pigs treated by injection of rat sarcoma, a tumour which can be propagated by inoculation in rats, acquires a toxic action on cultures of both rat sarcoma and the skin of embryo rats. And the plasma of guinea-pigs treated with rat embryo skin is also toxic for cells of both types. 32 THE CIRCULATING LIQUIDS OF THE BODY in the serum of any other kind -of animal. Thus, if human blood or serum is repeatedly injected at short intervals into a rabbit, the serum of the rabbit will cause a precipitate in diluted human blood or serum, but not in the blood or serum of other, animals, except that of monkeys, where a slight reaction may be obtained. The specific bodies which cause the precipitation are termed precipitins. The phenomenon has been made the basis of a method of dis- tinguishing human blood for forensic purposes. Other animal fluids and solutions containing tissue proteins likewise give rise to the production of precipitins. Thus, when cow's milk is injected into a rabbit, the rabbit's serum acquires the power of precipitating the caseinogen of cow's milk. Indeed, the response of the animal body to the presence of foreign proteins is so catholic, and at the same time so approximately specific, that many artificially isolated proteins, even those of vegetable origin, after as careful purifica- tion as possible, occasion, when injected, the production of anti- bodies which will precipitate from a solution only the variety of protein injected, or sometimes also, though in slighter degree, pro- teins nearly related to it. Anaphylaxis. — Under certain conditions the injection of a toxin, a serum, or a protein solution, instead of eliciting an immunity reaction which tends to combat the effects of a subsequent injection of the same material, produces the opposite result — namely, a sensitization of the animal which renders the second dose far more harmful than the first. Thus, Richet found that animals into which eel serum, or the poison contained in the tentacles of Actinaria, was subcutaneously introduced became much more sensitive to the toxic action of a second injection. This phenomenon he designated anaphylaxis, as being the opposite of the prophylaxis or protection afforded by previous treatment with the toxins hitherto studied. Later on it was discovered that the sub- cutaneous injection of a great variety of proteins alien to the animal into which they are introduced causes anaphylaxis. Only very minute amounts are necessary for the first or sensitizing dose, and an interval considerably greater than that employed in the production of an immune serum (p. 31) is allowed to elapse before the second injection is made. The symptoms induced in the sensitized animal by a subse- quent dose of the same material used in sensitization differ somewhat in different animals, but may be designated in general terms as those of collapse or shock (anaphylactic shock). They have been especially studied in the rabbit and guinea-pig, the heart being particularly affected in the former, and the lungs in the latter. The symptoms are very severe, and manifest themselves within a very short time (a few minutes) of the injection. A large proportion of the animals die. If an animal recovers, it does so suddenly, and for some time afterwards it is in- sensitive to the particular protein. While the real nature of protein sensitization or anaphylaxis is not as yet understood, it affords a new and delicate test for the detection and discrimination of proteins, and has already been utilized in a number of practical applications. For instance, the sophistication of sausages with other than the orthodox ingredients — e.g., with horseflesh — can be thus exposed, since an animal sensitized by horseflesh will exhibit anaphylaxis to horseflesh, but not to beef or pork. In like manner, the anaphylactic reaction may be GENERAL PHYSICAL AND CHEMICAL PROPERTIES 33 used for the identification of human blood. It is probable that an- aphylaxis plays an important role in certain pathological reactions. It is well known, for example, that some persons are so susceptible to particular foods .that the slightest indulgence in them brings on an attack of urticaria or nettlerash. It has been suggested that these persons have become sensitized to certain foreign proteins — such as those existing in eggs, veal, pork, strawberries, shellfish, or whatever the peccant article of diet may be — possibly by absorption at some previous time, owing to gastro-intestinal disturbance, of small quan- tities of the proteins which, have escaped complete digestion. It is only when proteins are introduced parenterally (i.e., by some other route than the alimentary canal, such as the subcutaneous tissue, the blood, or the serous cavities) that the immunity reactions already described and the phenomenon of anaphylaxis can be experimentally produced. For in digestion the protein molecule is decomposed, and although, as will be seen later on, the decomposition products are not the same for each kind of protein, the factor on which the specificity of the molecule depends does not survive the hydrolysis. COAGULATION OF THE BLOOD. Since changes begin in the blood as soon as it is shed, having for their outcome clotting or coagulation, we have to gather from the composition of the stable factors of clotted blood, or of blood which has been artificially prevented from clotting, some notion of the composition of the unaltered fluid as it circulates within the vessels. The first step, therefore, in the study of the chemistry of blood is the study of coagulation. When blood is shed, its viscidity soon begins to increase, and after an interval, varying with the kind of blood, the temperature of the air, and other conditions, but in man seldom exceeding ten, or falling below three, minutes, it sets into a firm jelly. This jelly gradually shrinks and squeezes out a straw-coloured liquid, the serum. Under the microscope the serum is seen to contain few or no red corpuscles; these are nearly all in the clot, entangled in the meshes of a kind of network of fine fibrils composed of fibrin. In uncoagulated blood no such fibrils are present ; they have accordingly been formed by a change in some constituent or constituents of the normal blood. Now, it has been shown that there exists in the plasma — the liquid portion of unclotted blood — a. substance from which fibrin can be derived, while no such substance is present in the corpuscles. In various ways coagulation can be prevented or delayed, and the plasma separated from the corpuscles. For example, the blood of the horse clots very slowly, and a low tem- perature lessens the rapidity of coagulation of every kind of blood. If horse's blood is run into a vessel surrounded by ice and allowed to stand, the corpuscles, being of greater specific gravity than the plasma, gradually sink to the bottom, and the clear straw-yellow plasma can be pipetted off. Or the addition of neutral salts to 3 34 THE CIRCULATING LIQUIDS OF THE BODY blood may be used to delay coagulation, the blood being run direct from the animal into, say, a third of its volume of saturated mag- nesium sulphate solution. The plasma may then be conveniently separated from the corpuscles by means of a centrifugal machine. Again, two ligatures may be placed on a large bloodvessel, so that a portion of it can be excised full of blood and suspended vertically (the so-called experiment of the ' living test-tube ') ; coagulation is long delayed, and the corpuscles sink to the lower end. In these and many other ways plasma free from corpuscles can be got ; and it is found that when the conditions which restrain coagulation are removed — when, for instance, the temperature of the horse's plasma is allowed to rise or the magnesium sulphate plasma is diluted with several times its bulk of water — clotting takes place, with forma- tion of fibrin in all respects similar to that of ordinary blood-clot. The corpuscles themselves cannot form a clot.* From this we con- clude that the essential process in coagulation of the blood is the formation of fibrin from some constituent of the plasma, and that the presence of corpuscles in ordinary blood-clot is accidental. In accordance with this conclusion, we find that lymph entirely free from red corpuscles clots spontaneously, with formation of fibrin ; and when fibrin is removed from newly shed blood by whip- ping it with a bundle of twigs or a piece of wood, it will no longer coagulate, although all the corpuscles are still there. What, now, is the substance in the plasma which is changed into fibrin when blood coagulates ? If plasma, obtained in any of the ways described, be saturated with sodium chloride, a precipitate is thrown down. The filtrate separated from this precipitate does not coagulate on dilution with water; but the precipitate itself — the so-called plasmine of Denis — on being dissolved in a little water, does form a clot. Fibrin is therefore derived from something in this precipitate. Now, ' plasmine ' contains two protein bodies — fibrinogen, which coagulates by heat at about 56° C., and serum- globulin, which coagulates at about 75° C., and it was at one time believed that both of these entered into the formation of fibrin (Schmidt). Hammersten, however, has shown that fibrinogen alone is a precursor of fibrin; pure serum-globulin neither helps nor hinders its formation. This observer isolated fibrinogen from blood- plasma by adding sodium chloride till about 13 per cent, was present. With this amount the fibrinogen is precipitated, while serum-globulin is not precipitated till 20 per cent, of salt is reached. After precipitation of the fibrinogen, the plasma no longer coagu- lates; and a solution of pure fibrinogen can be made to clot and to form fibrin, while a solution of serum-globulin cannot. Blood- * Bird's corpuscles, however, washed free from plasma, will form a clot when laked in various ways, as by addition of water or by freezing and thawing. COAGULATION 35 serum, too, which contains abundance of serum-globulin, but no fibrinogen, will not coagulate. So far, then, we have reached the conclusion that fibrin is formed by a change in a substance, fibrinogen, which can be obtained by certain methods from blood-plasma. It may be added that there is evidence that fibrinogen exists as such in the circulating blood; for if unclotted blood be suddenly heated to about 56° C., the tem- perature of heat-coagulation of fibrinogen, the blood for ever loses its power of clotting. The liver seems to be an important place of origin of fibrinogen, which may also be formed in the bone-marrow. That the liver is intimately concerned in the production of fibrinogen is indicated by a number of facts. In phosphorus poisoning, and notably in poisoning by chloroform, which causes necrosis, especially of the central portions of the hepatic lobules, the amount of fibrinogen in the blood is quickly diminished. The diminution is proportional to the extent of the injury to the liver, and the blood loses more or less completely its power of clotting. If the injury is repaired, the fibrinogen is rapidly regenerated (Whipple) . If the blood is allowed to circulate for a time in the head and thorax of an animal without passing into the rest of the animal's body, it becomes incoagulable, and the fibrinogen is found to be markedly deficient in amount. When the blood of an animal is defibrinated by whipping, and reinjected, regeneration of the fibrinogen does not occur if the liver has been eliminated, whereas it takes place rapidly if the liver is intact (Meek) . Since fibrinogen is readily soluble in dilute saline solutions, and fibrin only soluble with great difficulty, we may say that in coagu- lation of the blood a substance soluble in the plasma passes into an insoluble form. How is this change determined when blood is shed ? We have said that a solution of pure fibrinogen can be made to coagulate, but it does not coagulate of itself. The addition of another substance in minute quantity is necessary. This sub- stance, to which the name thrombin has been applied, can be obtained in various ways, although not in a state of purity; for example, by precipitating blood-serum, or defibrinated blood, with fifteen to twenty times its bulk of alcohol, letting the whole stand for a month or more, and then extracting the precipitate with water. All the ordinary proteins of the blood having been ren- dered insoluble by the alcohol, the thrombin passes into solution in the water, and the addition of a trace of the extract to a solution of fibrinogen causes coagulation. When purified as well as possible, thrombin still gives protein reactions, but it is not known whether it is really a protein. The action of thrombin on fibrinogen helps to explain many experiments In coagulation. Thus, transudations like hydrocele fluid do not clot spontaneously, although they contain fibrinogen, 36 THE CIRCULATING LIQUIDS OF THE BODY which can be precipitated from them by a stream of carbon dioxide or by sodium chloride. But the addition of a little thrombin causes hydrocele fluid to coagulate. So does the addition of serum, not because of the serum-globulin which it contains, as was once believed, but because of the thrombin in it. The addition of blood- clot, either before or after the corpuscles have been washed away, or of serum-globulin obtained from serum, also causes coagulation of hydrocele fluid, and for a similar reason, the thrombin having a tendency to cling to everything derived from a liquid containing it. On the other hand, serum which, although thrombin is present in it, does not of itself clot, because the fibrinogen has all been changed into fibrin during coagulation of the blood, can be made to coagulate by the addition of hydrocele fluid, which contains fibrinogen. We have thus arrived a step farther in our attempt to ex- plain the coagulation of the blood: it is essentially due to the formation of fibrin from the fibrinogen of the plasma under the influence of thrombin. Up to this point there is agreement between physiologists. Some difference of opinion exists, however, as to the manner in which thrombin is formed or activated when blood is shed, aud a? to the nature of its action upon fibrinogen once it is fully formed. The Formation of Thrombin from its Precursors. — There is good reason to believe that thrombin is formed by the interaction of three factors: (i) A substance which, since it is a precursor of thrombin, is called thrombogen, or prothrombin. It is already present in the circulating plasma. (2) A substance liberated from the formed ele- ments of the shed blood, but which can be obtained also from the cells of all tissues. Since it has been supposed to act upon throm- bogen, changing it into fully formed thrombin, much in the same way as enterokinase (p. 370) acts upon trypsinogen, changing it into fully formed trypsin, it is called thrombokinase (Morawitz). (3) Calcium ions. The following experiments illustrate the role of these three factors: The plasma obtained by drawing off bird's blood — e.g., the blood of a fowl or goose — through a perfectly clean cannula into a perfectly clean vessel, without contact with the tissues, and then rapidly centrifugal- izing off the formed elements, can be kept unclotted for days and even weeks. The addition of a small amount of tissue extract (procured by rubbing up blood-free liver, thymus, muscle, or other organs with sand, and extracting for several hours with salt solution) to the bird's plasma causes rapid coagulation. The plasma contains thrombogen and calcium salts, but is lacking in thrombokinase, which is supplied by the tissue extract. A solution of fibrinogen containing calcium will clot if serum, in which fibrin -ferment is always present, be added. It will not clot on addition of tissue-extract alone, nor on addition of bird's plasma alone (obtained as above), but will readily coagulate if both tissue extract and bird's plasma be added. Therefore, something in the bird's plasma (thrombogen), plus something in the tissue extract (thrombokinase), produce in the presence of calcium the same effect as the thrombin of serum. It can be shown that calcium is only necessary COAGULATION 37 for the formation of the thrombin, but not for its action on fibrinogen. For instance, a calcium-free solution of fibrinogen can be made to clot by serum from which the calcium has been removed. If a soluble oxalate (potassium or ammonium oxalate) is mixed with freshly drawn dog's blood to the amount of o'2 or o'3 percent., the blood remains unclotted. The plasma separated from this oxalated blood contains both thrombogen and thrombokinase, but it does not coagu- late, because the calcium has been precipitated out in the form of in- soluble calcium oxalate. In the absence of calcium the reaction of the thrombogen and thrombokinase which leads to the formation of thrombin does not take place. All that is necessary to bring about coagulation is to add calcium chloride in somewhat greater quantity than is required to combine with any excess of oxalate present. If more than a certain amount of calcium be added, clotting is hindered instead of being helped, so that it is only within certain limits of concentration that calcium favours coagulation. From oxalate plasma a nucleo- protein or a mixture of nucleo-proteins can be separated which contains thrombogen and thrombokinase, but little or no calcium, and does not cause clotting, but which on treatment with a calcium salt acquires the properties of thrombin. When sodium fluoride is added to freshly drawn blood to the amount of o'3 per cent., coagulation is also prevented. But there is this differ- ence between oxalate and fluoride plasma — that, although the calcium has been precipitated in both, the addition of calcium chloride to fluoride plasma is not sufficient to induce clotting. Tissue extract containing thrombokinase must be supplied as well. In some way or other sodium fluoride interferes with the liberation of thrombokinase from the formed elements of the blood, although in the concentration mentioned it does not hinder the action of fully formed thrombin, as is shown by the fact that fluoride plasma coagulates on the addition of a little serum, which supplies thrombin. The fluoride blood clots readily if it is diluted with water, and at the same time mixed with calcium chloride solution, for the water damages the formed elements, and thus favours the liberation of thrombokinase. Sodium citrate solution prevents the coagulation of blood run into it, although there is no precipitation of the calcium. The addition of calcium chloride to citrate plasma induces clotting, and the action of the citrate is assumed to be due to the formation of a compound with the calcium of the blood, which does not dissociate so as to yield calcium ions. It ought to be remarked, however, that in all so-called decalci- fied plasmas, as ordinarily obtained, blood-platelets are present, and that platelets disintegrate under the influence of calcium salts. It has been shown, indeed, that many of the reagents and procedures which hinder the clotting of shed blood also prevent the breaking up of the platelets. Thus, the cooling of the blood, the addition of hirudin, sodium oxalate, sodium citrate, manganese salts, etc., which are classical methods used in obtaining platelets for microscopical study, are also classical methods of hindering coagulation. These facts have not hitherto been sufficiently taken account of in interpreting experi- ments on decalcified blood. They indicate that the decalcifying agents may hinder clotting by interfering with the liberation of essential sub- stances from the platelets, and that this may be the decisive factor, and not merely the withdrawal of the calcium from the field where the already liberated thrombokinase and thrombogen would otherwise react to form thrombin. When proteoses (or peptones) are injected into the circulation of a dog or goose, the blood is deprived of the power of coagulation. The 38 THE CIRCULATING LIQUIDS OF THE BODY peptone plasma must be assumed to contain both thrombogen and thrombokinase, since it can be made to clot in various ways (e.g., by dilu- tion with water or by slight acidulation with acetic acid) without the addition of anything which could supply either of these factors. Yet a little tissue extract causes it to clot much more rapidly than simple dilution or acidulation, and more rapidly than the addition of serum. So that either the thrombokinase already present in peptone plasma is present in an unavailable form, or in some way the formation of throm- bin from its precursors is hindered. But this is not the only cause of the incoagulability of peptone plasma. It may be shown to contain an antithrombin, a body which antagonizes the action of fully formed thrombin, and which does not seem to be a ferment, since it acts quan- titatively in proportion to the amount present. This is the reason why, although peptone plasma can always be made to clot by the addition of fibrin ferment, in serum, for instance, relatively large quantities of it must be supplied (Practical Exercises, pp. 64, 65). Fig. 5. — Fibrin Formation in Horse's Plasma (Ultramicroscope) (Stiibel). Several clumps of disintegrated platelets from which the fibrin filaments radiate. An extract of the head of the medicinal leech in salt solution prevents the clotting of blood both in the test-tube and when injected into the circulation. The plasma obtained differs from peptone plasma in refusing to coagulate unless tissue extract is added. It is therefore deficient in thrombokinase, or, rather, as has been shown, the kinase present is unable to act, because neutralized by antikinase present in the leech extract. Leech extract also contains an antithrombin, which can be neutralized by a sufficient amount of thrombin. In the small wound from which the leech sucks blood this sufficient amount is not present, and the blood remains unclotted, as it also does in the alimen- tary canal of the leech. The anticoagulant substance, hirudin, has been isolated, and gives the reactions of an albumose. Sources of Thrombogen and Thrombokinase. — It has already been stated that thrombogen exists in the circulating plasma. This is shown by the fact that fluoride plasma obtained from blood drawn directly through a wide cannula into sodium fluoride solution, with COAGULATION all precautions to prevent alteration of the blood, and immediately separated from the formed elements by the centrifuge, will clot on the addition of tissue extract. The source of the thrombogen has been thought to be the blood-plates, but this has not been proved. Thrombokinase is not present in the circulating plasma. In shed and clotting blood which has not been allowed to come into contact with cut tissues, the only possible sources of thrombokinase, so far as we know, are the corpuscles and the blood-plates. The red corpuscles we may at once dismiss, for although the stromata, especially of nucleated corpuscles, contain thrombokinase, or can under artificial conditions be made to develop that action on coagulation by which we recognize its presence, not only do they remain intact under ordinary circumstances during coagulation, but there is strong evidence, as has already been pointed out, that they do not make any essen- tial contribution to the process. We have left over the leucocytes and the platelets, and it is highly probable that from the platelets thrombokinase is liberated in the first moments after blood is drawn, and, acting on the throm- bogen already present in the plas- ma, changes it into actual throm- bin. This surmise is strengthened by the fact that in freshly shed mammalian blood extensive de- struction of blood-plates takes place. Viewed with the ultra- microscope, the blood-platelets, in a drop of clotting plasma, which are at first homogeneous in appearance (optically empty), become granular. Then the platelets begin to agglutinate and swell up, and the agglutinated platelets are transformed into clumps of granules, from which needles of fibrin shoot out. Other needles and filaments of fibrin form in contact with the glass or free in the plasma, and soon the field is occupied by a felt-work of fibrin. The leucocytes have not been observed to be related to the process, at least, in the blood of mammals (Stiibel). It is true that the white layer or ' buffy coat ' which tops the tardily formed clot of horse's blood, and consists of the lighter, and therefore more slowly sinking colourless cells, causes clotting in otherwise in- coagulable liquids like hydrocele fluid much more readily than the red portion of the clot, and yields far more thrombin on treatment with alcohol. It can also be easily verified that in mammalian Fig. 6. — Fibrin Formation in Plasma from a Case of Haemophilia (Ultramicro- scope) (Stiibel). The needles of fibrin are slowly formed, and very large. 40 THE CIRCULATING LIQUIDS OF THE BODY blood collected in paraffined vessels, so as to delay clotting, and immediately centrifugalized, coagulation begins in and around the layer of white elements, and then spreads upwards in the stratum of plasma and downwards in the stratum of erythrocytes. But in this white upper layer platelets are always intermingled with leuco- cytes. It has been shown, however, that the blood of the cray- fish, which coagulates with extreme rapidity, contains certain colourless corpuscles, which immediately it is withdrawn, break up with explosive suddenness, and that substances which hinder the breaking up of these corpuscles restrain coagulation (Hardy). In the blood of another crustacean Limulus, the kingcrab, coagula- tion is preceded by an agglutination of the leucocytes which exhibit amoeboid movements. They become entangled by the interlacing of the pseudopodia which they protrude (L. Loeb). The disintegration of the platelets in shed blood has been attributed by Deetjen to an increase in the alkalinity of the blood, by escape of carbon dioxide, it may be. When blood is placed on a quartz slide and covered with a quartz cover-slip, the platelets, according to this observer, do not break up ; but if they are brought into con- tact with a medium whose OH — concentration is raised ten times or more above that of freshly drawn blood (still only a weak alka- line reaction), disintegration ensues. He supposes that the contact of glass acts harmfully on account of the alkali in it. It is im- possible to say at present whether this observation has any bearing on normal coagulation. Thrombokinase has been shown to exist not only in the leuco- cytes, the platelets, and the stromata of the coloured corpuscles, but, as already stated, in all tissues hitherto examined. Under ordinary circumstances it appears that a larger amount of thrombogen is liberated or is already present in shed blood than can be changed into thrombin by the thrombokinase set free, since serum contains a surplus of thrombogen in addition to the fully formed ferment. This is shown by the fact that the activity of a given quantity of serum in causing the coagulation of a plasma not spontaneously coagulable or of a fibrinogen solution is increased by the addition of tissue extract (containing thrombokinase). The thrombin of any particular kind of vertebrate blood has no marked specific action — that is, will cause coagulation in solutions of fibrinogen or plasma of very different origin. For example, the sera of all vertebrates hitherto investigated induce clotting in goose's plasma. On the other hand, it appears that a greater degree of specificity exists in the case of the thrombokinase and throm- bogen, the specificity being absolute in some cases, relative in others. That is to say, the thrombokinase of one animal may activate the thrombogen of an animal of another group, while it may fail to activate the thrombogen of an animal belonging to a third group. COAGULATION 41 But it will always activate the thrombogen of an animal of the same kind. To sum up, we may say that when blood is shed, thrombin is rapidly formed by the action of thrombokinase, liberated from the leucocytes, the blood-plates, and possibly to some extent from the erythrocytes, upon thrombogen, already present in the circulating plasma. Further — and this is of great practical importance — since no vessel is opened under ordinary circumstances except through a wound in the overly- ing structures, the cut tissues supply a store of thrombokinase at the point where it is required to aid in the stanching of the wound. Calcium is essential to the reaction by which thrombogen and thrombo- kinase form thrombin, but is not necessary for that action of thrombin on fibrinoges by which fibrin is produced (Practical Exercises, pp. 62-65). The Nature of the Action of Thrombin on Fibrinogen. — The usual view, first advanced by Schmidt many years ago, is that thrombin acts as an enzyme. Hence it is often spoken of as fibrin- ferment. In support of this theory it has been stated that the thrombin does not itself seem to be used up in the process, nor to enter bodily into the fibrin formed; that a small quantity of it can cause an indefinitely large amount of fibrinogen to clot; and that its power is abolished by boiling (p. 335). There has been a disposition among more recent observers to question this evidence. Accord- ing to Rettger, the quantity of fibrin formed when a small amount of thrombin is added to a fibrinogen solution tends to a fixed maxi- mum, which does not increase with the time of action.* Under certain conditions, also, it is said that thrombin is not destroyed at the temperature of boiling water. Whatever the precise nature of the reaction which leads to the precipitation of the fibrinogen in the form of fibrils, thrombin is very loosely combined if combined at all in the fibrin, since it is readily extracted by an 8 per cent, solution of sodium chloride. This, indeed, is one of the best ways of obtaining an active thrombin solution. The view which we have followed above, in accordance with Morawitz, that the substances in tissue extracts which favour coagulation do so by activating prothrombin to fully formed thrombin, has also been opposed by a number of the more recent workers. Some consider that they exert a direct action upon fibrinogen similar to, although not necessarily identical with, that of thrombin, and speak of them as coagulins (L. Loeb). Howell holds that these substances, which he prefers to term thromboplastic substances, since this makes no assumption as to their mode of action, play a quite different role, namely, that of neutralizing anti- thrombin. His observations have led him to the conclusion that * The inquiry is complicated by the fact that fibrin, once formed, tends to adsorb the remaining thrombin and so tc interfere with its further action. 42 THE CIRCULATING LIQUIDS OF THE BODY the effective thromboplastic substance in the tissues is a phos- phatide, probably kephalin, united with protein. Intravascular Coagulation Regulation of the Clotting Process, or Thrombotaxis.— So far we have been considering the problem of coagulation as if all the data for its solution could be obtained by a study of the blood itself. In other words, our main business up to this point has been the explanation of coagulation in the shed blood; it has been only incidentally, and with the object of casting light on the question of extravascular clotting, that we have touched on the coagulation of the blood within the living vessels. It is not possible here to adequately discuss, nor even to define, the differences between the two problems. All we can do is to warn the student, and to emphasize the warning by one or two illustrations, that valuable as is the knowledge derived from experiments on extravascular coagulation, it would be totally mis- leading if applied without modification to the circulating blood. For instance, we have recognized in the blood-plates an important source of the thrombokinase which plays so great a part in the clotting of shed blood; but we may be sure that blood-plates are constantly breaking down in the lymph and the blood, and we have to inquire how it is that coagulation does not occur, except in disease, within the vessels. Calcium is not wanting to the circu- lating plasma, fibrinogen is not wanting, and it has already been mentioned that thrombogen exists in perfectly fresh and, as we may say, still living blood. Why, then, does it not coagulate ? Some have said that coagulation is ' restrained ' by the contact of the living walls of the bloodvessels; but although it is certain that the contact of foreign matter — and all dead matter is foreign to living cells — does hasten the destruction of blood-plates or that alteration in them on which the liberation of the precursors of the ferment depends, it is evident that it is just this ' restraining ' in- fluence of the vessels, if it is due to anything more than the mere smoothness of their endothelial lining, which has to be explained. The best answer which can be given to the question is: First, that the quantity of thrombokinase free in the plasma at any given time must be small, since no evidence of its presence in fluoride plasma can be obtained. If thrombokinase is liberated in the circulating blood, we may assume that it is changed into some inactive sub- stance, or quickly eliminated. And it appears that, unlike the true ferments, thrombokinase acts quantitatively — i.e., in proportion to its amount — upon thrombogen. Second, an antithrombin exists in the circulating plasma, and even if fully formed fibrin-ferment were present, it could not cause coagulation until the antithrombin had been neutralized. This antithrombin is probably not manu- factured in the blood, or at least not exclusively in the blood, but in the tissues, and there is no reason to deny the vessels themselves COAGULATION 43 a share in its production, even if its presence has not hitherto been demonstrated in the internal coat (L. Loeb) . So that living blood within the living vessels may be said to be acted upon by two sets of influences, one tending to favour coagulation, the other to oppose it. In the clotting of extravascular plasma, free from corpuscles, we may indeed see the continuation, under modified conditions, of a normal process always going on within the bloodvessels. Under normal conditions, the processes that make for coagulation never obtain the upper hand. Indeed the margin of safety within which what may be called the thrombo-regulative mechanism works seems to be surprisingly wide, and the equilibrium in the circulating blood far more stable than observations on clotting outside of the body might lead us to suppose. Very considerable quantities of thrombin or of de- fibrinated blood or serum containing thrombin can be injected into the blood-stream without ill effect. According to Ho well, the presence of the abnormally great amount of thrombin causes the formation of sufficient antithrombin to neutralize it, probably by a protective reflex secretion. In like manner the injection of tissue extracts or a solution of thrombo plastic substance (thrombokinase) prepared from them by precipitation does not necessarily induce coagulation in the vessels. On the contrary, when injected slowly or in small amount into the veins of an animal, it abolishes for a time the power of coagulation of the blood; and when this ' nega- tive phase,' as it is called, has been once established, even a very large and rapid injection produces no further effect, possibly because an antibody which neutralizes the action of thrombokinase has been produced. In both cases the limits of safety can be over- stepped, and intravascular clotting induced by the injection either of thrombin or of thrombokinase. When a considerable quantity of the active substance in tissue extract is introduced at the first injection, extensive coagulation in the vessels instantly ensues; the animal dies in a few minutes; and the right side of the heart, the venae cavae, the portal vein, and perhaps the pulmonary arteries, may be found choked with thrombi. Here the injected thrombo- kinase is responsible for the clotting, thrombogen and calcium being already present. Curiously enough, intravascular coagulation fails to be produced in a certain proportion of cases when albino animals are injected with material from pigmented animals, while there is no absolute failure of coagulation when albinos are injected with material from albinos, and no failure when pigmented animals are injected with material either from other pigmented animals or from albinos. Intravascular coagulation on injection of tissue extracts is especially striking in birds. To a certain extent the action of tissue extracts in coagulation can be imitated by other substances of animal origin, such as the 44 THE CIRCULATING LIQUIDS OF THE BODY venoms of some vipers (Martin). It is not known whether these substances act on the blood-plates, leucocytes, or other cells, and thus cause an increased production or an increased liberation of one or more of the precursors of thrombin, or whether they take part directly in its formation. But there is some evidence that the venoms which favour coagulation do so in virtue of their con- taining a kinase. On the other hand, cobra- venom prevents coagula- tion by means of an antikinase — that is, a substance which antago- nizes the action of kinase, and so hinders the formation of thrombin. It does not contain an antithrombin — that is, a body which will prevent the action of thrombin already formed (Mellanby). Relation of the Liver to Coagulation.— It is not known with any degree of certainty whether the thrombo-regulative processes are especially associated with any particular organ. But there are facts which suggest that the relations of the liver to the coagulation of the blood are peculiarly close. Not only, as previously shown, does it take an important share in the formation of fibrinogen, but there is some evidence that it is closely related to the formation of antithrombin. We have already mentioned that the injection of commercial peptone, which consists chiefly of proteoses, into the blood of dogs causes it to lose its coagulability. The effect gradually passes away, till after some hours the original power of coagulation is restored (p. 63). The liver is known to be intimately concerned in the production of this remarkable result, for if the circulation through it be interrupted, the injection of proteose is ineffective. Further, if a solution of proteose is artificially circulated through an excised liver, a substance (perhaps an anti- thrombin) is formed which is capable of suspending the coagula- tion of blood outside of the body, a property which proteoses them- selves do not possess, or possess only in slight degree. It is not believed that the proteose is actually changed into this anticoagu- lant substance, but rather that the liver cells produce it as a ' reaction ' to the presence of the foreign substance, being perhaps stimulated in some way by the circulating proteose. In part the abnormally great alkalinity of the peptone blood, due to the excess of alkali secreted by the liver, is responsible for its slow coagulation. Under certain conditions, some of which are known and others not, the injection even of one or other of the purified proteoses causes not retardation, but hastening, of coagulation; and if this has been the result of a first injection, a second is equally unsuccessful. It is possible that by an effort of the organism to restore the normal coagulability of the blood, on which its very existence depends, substances which favour coagulation are produced, and that the result of an injection of proteose is determined by the relative amount of coagulant and anticoagulant secreted in a given time. Protamins (products obtained from the ripe milt of certain fishes, COAGULATION 45 and believed to be the simplest proteins) exert, when injected intravenously, a retarding influence on coagulation, and lower the blood-pressure,, just as albumoses do (Thompson). Even serum- albumin and serum-globulin possess this property in some degree. All these substances also cause a diminution in the number of leucocytes in the blood owing, in the case of albumose at any rate, to their accumulation in the abdominal vessels, and not to any actual destruction of them. It has been lately announced that the adrenal glands have a relation to the coagulation of the blood. Stimulation of the splanchnic nerve, which supplies secretory fibres to the adrenal, greatly hastens coagula- tion, but has no such effect if the adrenal on the corresponding side has been previously removed (Cannon). It is possible that this effect is exerted through the liver, since it is known that one important function of the liver, the regulation of the sugar content of the blood, is intimately dependent upon the adrenal, and is affected by excitation of its splanchnic nerve-supply. In certain pathological conditions the normal balance of the factors that make for clotting and prevent it may be upset, and the scales may tip in either direction. In patients suffering from the formation of spontaneous clots in the veins (thrombosis) it is stated that the anti- thrombin in the blood is diminished, the amount of prothrombin being normal. The mere slowing of the blood-stream in conditions where the circulatory mechanism is enfeebled may favour thrombosis. For anything which cripples the circulation, and consequently limits the free interchange between blood and tissues, interferes with the elimina- tion or neutralization of the precursors of thrombin, and with the entrance of the substances that render the fully formed thrombin in- active. This, as well as the injury caused by the ligature, which may favour the passage of thromboplastic substances into the lumen of the occluded vessel, is a possible factor in the formation of the clot on which the surgeon relies for the permanent sealing of ligated vessels. In haemophilia, a disease in which the coagulation of the blood is characteristically slow, and in which even slight wounds may occasion . severe or fatal haemorrhage, the thrombogen (prothrombin) has been found deficient in amount, and the injection of normal serum or the transfusion of normal blood has been used with temporary advantage in the treatment of the condition. In certain cases of purpura, how- ever, where haemorrhage also occurs with abnormal ease, no variation from the normal could be detected in the content of either antithrombin or prothrombin (Ho well). Some have supposed that in such conditions the fault is an unnatural fragility of the small vessels rather than a deficiency in the power of the blood to clot, but of this also no actual evidence has been adduced. Another factor on which the promptness and completeness of the sealing of wounded vessels may depend has been recently brought into notice, namely — The Vaso-Constrictor Property of Shed Blood. — It has been shown that when blood is shed and no precautions are taken to prevent clotting, it very quickly develops the power of causing marked constriction of bloodvessels. This can be demonstrated by allow- ing the serum to act on rings cut from arteries (Practical Exercises, 46 THE CIRCULATING LIQUIDS OF THE BODY p. 66), or by perfusing the hind-legs of frogs with a saline solu- tion containing serum. Plasma derived from blood in which the platelets have been prevented from breaking down, and which therefore remains unclotted, has no such effect, or a much slighter Fig. 7. — Sheep Artery Rings. At 14 and 16 Ringer's solution was replaced by citrate plasma (two different specimens). At 15 and 17 the plasmas were replaced by the corresponding sera. At 18 and 19 the sera replaced Ringer's solution directly. Time-trace, half-minutes. Tracings reduced to J. effect. But when the platelets are separated from the plasma and then decomposed, the resulting extracts of platelets are rich in vaso-constrictor material. In the sealing of wounded vessels the platelets would therefore appear to play a double role, yielding Fig 8. — Frog Perfusion Experiment with Serum. The drops of liquid flowing through the preparation are recorded. At n citrate plasma was injected; at 13 the corresponding citrate serum. The tracing is to be read from left to right. Time is marked in half -minutes. a substance which causes constriction of the vessel in the neigh- bourhood of the wound while a plug of clot is being formed, thanks to other substances liberated from the platelets, which take an essential part in coagulation. The vaso-constriction may perhaps VASO-CONSTRICTOR PROPERTY OF SHED BLOOD 47 be looked upon as a form of ' first aid ' to diminish the haemor- rhage, and also to make it less easy for the beginning clot to be washed away. It is obvious that the two processes would be mutually advantageous in dealing with those injuries of the vascular system on the prompt repair of which the very existence of the "A /v- -Ar- Z-LsjL. ~+\ ^ \l \ z x^ Jc.c Serum* citrate /cc Plasma /cc Plasma Id Seru/r'cUraie Fig. 9. — Frog Perfusion Experiment with Serum. Curves showing the flow. The number of drops per half-minute is laid off along the vertical axis, and the time (in half-minutes) along the horizontal axis; 38 drops correspond to i c.c. organism at all times depends, and it is not without interest to find that special formed elements in the blood, the platelets, are pre-eminently associated with both processes. SECTION III. — THE CHEMICAL COMPOSITION OF BLOOD. The serum of coagulated blood represents the plasma minus fibrinogen; the clot represents the corpuscles plus fibrin. Thus: Plasma - Fibrin (o gen) = Serum. Corpuscles + Fibrin = Clot. Plasma + Corpuscles — Serum +Clot Blood. Bulky as the clot is, the quantity of fibrin is trifling (0-2 to 0-4 per cent, in human blood). The plasma contains about 10 per cent, of solids, the red corpuscles about 40 per cent., the entire blood about 20 per cent. Serum contains 7 to 8 per cent, of proteins, about 0-8 per cent, of inorganic salts, and small quantities of neutral fats, soaps, cholesterin esters, lecithin, dextrose, urea, lactic acid, glycuronic acid, amino - acids, Solids Cor/7usdes[~ and other sub- stances. The chief proteins are serum- albumin and serum- globulin. In the rab- bit the former, in the horse the latter, is the more abundant ; in man they exist in not far from equal amount. A small quantity of nucleo -protein and of fibrino-globulin (which some consider a soluble product formed from Plasm* Fig. 10. — Diagram showing Relative Quantity of Solids and Water in Red Corpuscles and Plasma. 48 THE CIRCULATING LIQUIDS OF THE BODY nbrinogen in clotting) is also present. Ferments which cause hydrolysis of proteins and carbohydrates, a ferment (lipase) which acts upon fats, and certain oxidizing ferments (oxydases), have also been demonstrated. The chemical nature of the bodies which confer on serum or plasma its specific haemolytic, agglutinating, precipitating, and bactericidal properties has not been definitely determined. The quantitative composition of serum, especially as regards the inorganic salts, is remarkably constant in animals of the same species, and even in animals of different species belonging to the same, or to not very widely separated, natural groups. In cold-blooded animals the serum-albumin is scantier than in mammals, the globulin relatively more plentiful. Serum-albumin belongs to the class of native albumins. It has been obtained in a crystalline form from the serum of horse's blood. It Fig. ii. — Perspective View of Vivi-Diffusion Apparatus (Abel). This form of the apparatus contains sixteen tubes. A, arterial cannula; B, venous cannula; C, side tube for introduction of hirudin; D, inflow tube; E, outlet tube for the blood; F, G, supporting rod attached at H and K to branched V-tubes; L, burette for hirudin; M, N, tube for filling and emptying liquid in outer jacket; O, air outlet; P, dichotomous branching-point of inflow tube; Q and R, quadruple branching-points of the same ; S, S, wooden supports; T, thermometer. At each of the points H and K the blood is collected from four tubes into one, bending round to the back, and there redividing into four return flow tubes. Arrows show the direction of the flow. is soluble in distilled water, and is not precipitated by saturating its solutions with certain neutral salts. Heated in neutral or slightly acid solution, it coagulates first at 73°, then at 77°, then at 84° C. Although this is not of itself sufficient proof, there is other evidence that it consists of a mixture of proteins. Serum-globulin, also called paraglobulin, belongs to the globulin group of proteins. When heated, it coagulates at about 75° C. (p. 9). It is insoluble in distilled water, and is precipitated by saturation with such neutral salts as magnesium sulphate, or by half -saturation with ammonium sulphate. It appears that, as thus obtained, it is not a single substance, but a mixture of at least two proteins — eu-globulin, CHEMICAL COMPOSITION OF BLOOD 49 which can be precipitated from its saline solution by dialyzing off the salts, and pseudo-globulin, which cannot be so precipitated. In addition to the nitrogen represented as protein, serum (or plasma) contains non-protein nitrogen, the amount of which varies with the nature of the food and the stage of digestion. Part of this fraction is attributable to urea and other metabolites on their way to be excreted, but another portion, and an important one, is due to amino-acids absorbed from the intestine during the digestion of proteins and on their way to be utilized in the tissues. Of the inorganic salts of serum, the most important are sodium chloride and sodium bicarbonate. Small amounts of potassium, calcium, and magnesium, united with phosphoric acid or chlorine, and a trace of fluoride, are also present. A portion of the salts is loosely combined with the proteins. Our knowledge of the chemistry of the circulating plasma is likely to be notably augmented .by the method of vivi-diffusion recently intro- duced by Abel. An artery of an anaesthetized animal is connected by a cannula to a system of celloidin tubes immersed in a saline solution. Blood passes from the artery through the tubes, where it exchanges diffusible constituents with the solution, and is then returned to the animal's body by another cannula attached to a vein. Coagulation of the blood in the apparatus is prevented by hirudin, and under aseptic conditions the circulation may proceed through the tubes for a long time. The saline solution can then be analyzed for substances which have entered it from the blood — amino-acids, for example (Fig. n). The following tables give some details of the composition of blood : 1,000 GRAMMES OF PIG'S BLOOD (CORPUSCLES, 435*09; SERUM, 564*91) CONTAINED Corpuscles. Serum. Corpuscles. Serum. Water 272'2O 5l8*36 P2O5 as nuclein 0-0455 O-OI23 Solids 162-89 46*54 Na2O . . * 2-401 Haemoglobin I42'2O K2O . . 2-157 0-152 Protein 8*35 38-26 Fe203 0-696 Sugar 0-684 CaO 0-0689 Cholesterin 0-213 0-231 MgO .. 0-0656 0-0233 Lecithin 1-504 0-805 Cl 0-642 2-048 Fat . . 1-104 P205 .. 0-895 O-III Fatty acids 0-027 0-448 Inorganic P2O5 0-719 0-296 PROTEINS OF PLASMA IN 1,000 GRAMMES. Albumin. Globulin. Fibrinogen. Total. Man 40-I 28-3 4-2 72-6 Dog 31-7 22-6 6-0 60-3 Sheep 38-3 30-0 4-6 72-9 Horse 28-0 47*9 4*5 80-4 Pig 44-2 29-8 6*5 80-5 * The pig's erythrocytes are peculiar in that the sodium appears to be entirely confined to the plasma. 5<> THE CIRCULATING LIQUIDS OF THE BODY The Coloured Corpuscles consist of rather less than 60 per cent, of water and rather more than 40 per cent, of solids. Of the solids the pigment haemoglobin makes up about 90 per cent. ; the proteins and nucleo-protein of the stroma about 7 percent.; lecithin and choles- terin 2 to 3 per cent. ; inorganic salts (which vary greatly in their relative proportions in different animals, but in man consist chiefly of phosphates and chloride of potassium, with a much smaller amount of sodium chloride) about i per cent. Potassium has been demonstrated microchemically in frog's erythrocytes (Macallum) (Frontispiece] . There is evidence that a portion of the salts is more firmly combined than the rest, so that, even after the action of the most energetic laking agents, this fraction remains attached to the stroma. The erythrocytes of some animals — e.g., the dog — contain dextrose. When dextrose -is added to human blood it rapidly dis- tributes itself over corpuscles and plasma (Rona), although not exactly in proportion to their respective volumes (Masing). Hither- to the dextrose in blood has been reckoned as if it all belonged to the plasma. Hemoglobin. — Of all the solid constituents of the blood, haemoglobin is present in greatest amount, constituting as it does no less than 13 per cent., by weight, of that liquid. It is an exceedingly complex body, containing car- D c bon, hydrogen, nitro- WA gen, and oxygen in much the same pro- portions in which they exist in ordinary pro- teins (p. i). Iron is also present to the ex- Pig. 12.— Diagram of Spectroscope. A, source of light ; tent °f almost exactly B, layer of blaod; C, collimator for rendering rays one-third pi I percent., parallel ; D, prism ; E, telescope. an<3 there is alsoa little sulphur. Haemoglobin is made up of a protein element which contains all the sulphur and a pigment which contains all the iron, the protein constituting by far the larger portion of the gigantic molecule, whose weight has been estimated at more than 16,000 times that of a molecule of hydrogen. Since its percentage composition is still undetermined with absolute precision, it is impossible to give an empirical formula that is more than approximately correct. For dog's haemoglobin Jaquet gives C-58Hi203Ni95S3FeO2i8- which would make the molecular weight 16,669. Direct determinations of the molecular weight gave 15,115 for oxyhsemoglobin of the horse, and 16,321 for that of the ox (Hiifner and Gansser). Whilethese numbers need not be taken as more than a rough approximation, they at least show that the haemoglobin molecule is an exceedingly large one. The most remarkable property of haemoglobin is its power of combining loosely with oxygen when exposed to an atmosphere con- taining it, and of again giving it up in the presence of oxidizable substances or in an atmosphere in which the partial pressure of oxygen (pp. 248-253) has been reduced below a certain limit. It ChEMICAL COMPOSITION OF BLOOD D E b F i 640 630 620 610 600 SM 580 570 560 550 5W 5ti 5W\ 510 500 490 I 1 I I I I' i I i I I I II! ! I lll(lllllllllllllll[llllllllllll[lllllllllllllllllljllll|lllllillllllllllll!lllllllllllllllMllllllllllllljlllill!l!!;:i!lllll!llllllllllllllHllllllllllllllll Fig. 13. — Table of Spectra of Haemoglobin and its Derivatives (Ziemka and MiilJerj. i, Oxyhosmoglobin ; 2, reduced haemoglobin 3, methasmoglobin ; 4, ac 'd haematia ; 5, alkaline haematin; 6, haemochromogen ; 7, acid hamatoporphyrin ; 8, alkaline hajmatoporphyrin ; 9, carbon monoxide haemoglobin. 52 THE CIRCULATING LIQUIDS OF THE BODY is this property that enables haemoglobin to perform the part of an oxygen-carrier to the tissues, a function of the first importance, which will be more minutely considered when we come to deal with respiration. The bright red colour of blood drawn from an artery or of venous blood after free exposure to air is due to the fact that the haemo- globin is in the oxidized state — in the state of oxyhaemo- globin, as it is called. The amount of oxygen with which haemoglobin combines to form oxyhaemoglobin is such that one atom of iron corresponds to two atoms of oxygen. If the formula for haemoglobin given on p. 50 be represented by the symbol Hb, a molecule of oxyhaemoglobin would be re- presented as Hb02. If the oxygen is removed by means of reducing agents, such as am- monium sulphide, or by ex- posure to the vacuum of an air- pump, the colour darkens, the blood-pigment being now in the form of reduced haemoglobin. In ordinary venous blood a large proportion of the pigment is in this condition, but there is always oxyhaemoglobin present as well. In asphyxia (p. 281), however, nearly the whole of the oxyhaemoglobin may dis- appear. Crystallization of H cemoglobin. — In the circulating blood the haemo- globin is related in such a way to the stroma of the corpuscles that, although the latter are suspended in a liquid readily capable of dissolv- ing the pigment, it yet remains under ordinary circumstances strictly within them. In a few invertebrates, however, it is normally in solution in the circulating liquid. As a rare occurrence haemoglobin may form crystals inside the corpuscles (p. 71). When it is in any way brought into solution outside the body, it shows in many animals, but not in the same degree in all, a tendency to crystallization ; and the ease with which crystallization can be induced is in inverse proportion to the solubility of the haemoglobin. Thus, it is far more difficult to obtain crystals of haemoglobin from human blood than from the blood of the rat, guinea- pig, or dog, whose blood-pigment is less soluble than that of man, and for a like reason the oxyhaemoglobin of the bird, the rabbit, or the frog crystallizes still less readily than that of human blood. Fig. 14. — Oxyhaemoglobin Crystals (Frey). a, b, from man ; c, from cat ; d, from guinea- pig; e, from hamster; /, from squirrel. CHEMICAL COMPOSITION OF BLOOD 53 As to the form of the crystals, in the vast majority of animals they are rhombic prisms or needles, but in the guinea-pig they are sphenoids belonging to the rhombic system, and in the squirrel six-sided plates of tae hexagonal system (Fig. 14). Careful study of the crystallography of haemoglobin from a large number of animals has established differences and resemblances so constant and so clear-cut that they may be used for the purposes of classification and for the identification of the source of a specimen of blood (Reichert and Brown). Reduced haemoglobin can also be caused to crystallize, though with more difficulty than oxyhaemoglobin, since it is more soluble. Crystals of reduced haemoglobin were first prepared from human blood by Hiifner, who allowed it to putrefy in sealed tubes for several weeks. When a solution of oxyhsemoglobin of moderate strength is ex- amined with the spectroscope, two well-marked absorption bands are seen, one a little to the right of Fraunhofer's line D, and the other a little to the left of E. A third band exists in the extreme violet between G and H. It cannot be detected with an ordinary spectro- scope, but has been studied by the aid of a fluorescent eyepiece, by projecting the spectrum on a fluorescent screen, and by photograph- ing the spectrum. The addition of a reducing agent, such as ammonium sulphide, causes the bands in the visible spectrum to disappear, and they are replaced by a less sharply denned band, of which the centre is about equidistant from D and E. This is the characteristic band of reduced haemoglobin. The spectrum of ordinary venous blood shows the bands of oxyhaemoglobin. Carbonic oxide haemoglobin is a representative of a class of haemo- globin compounds analogous to oxyhaemoglobin, in which the loosely- combined oxygen has been replaced by other gases (carbon monoxide, nitric oxide) in firmer union. Its spectrum shows two bands very like those of oxyhaemoglobin, but a little nearer the violet end. Carbonic oxide haemoglobin is formed in poisoning with coal-gas. Owing to the great stability of the compound, the haemoglobin can no longer be oxidized in the lungs, and death may take place from asphyxia. It is, however, gradually broken up, and therefore artificial respiration may be of use in such cases. Inhalation of oxygen and especially transfusion of blood are also of great value. Methcemoglobin is a derivative of oxyhaemoglobin which can be formed from it in various ways — e.g., by the addition of ferricyanide of potassium or nitrite of amyl (Gamgee), by electrolysis (in the neigh- bourhood of the anode), or by the action of the oxidizing ferment ' echidnase ' in the poison of the viper (Phisalix) . It very often appears in an oxyhaemoglobin solution which is exposed to the air. It has been found in the urine in cases of haemoglobinuria, in the fluid of ovarian cysts, and in haematoceles. The strongest band in its spectrum is in the red, between C and D, but nearer C, nearly in the same position as the band of acid-haematin. Reducing agents, such as ammonium sulphide, change methaemoglobin first into oxyhaemoglobin and then into reduced haemoglobin. It has by some been regarded as a more highly oxidized haemoglobin than oxyhsemoglobin. Rebutting evidence has, however, been offered to the effect that the same quantity of oxygen is required to saturate both pigments, and this evidence appears to be sound. The difference lies rather in the manner in which the oxygen is united to the haemoglobin in the methaemoglobin molecule than in the quantity of oxygen which it contains. For methaemo- 54 THE CIRCULATING LIQUIDS OF THE BODY globin, unlike oxy haemoglobin, parts with no oxygen to the vacuum, while, on the other hand, in the presence of reducing agents it yields up its oxygen even more readily than oxyhaemoglobin does (Haldahe) (p. 248). By the action of acids or alkalies oxyhaemoglobin is split into a pig- ment, haematin, and a protein, globin, belonging to the histon group. It is easily precipitated from solution by ammonia. On hydrolysis, it if* \l'/''-~. • — -.^ — I; 1 / ..-•"" I Xxj> •*•<; D o E Carbonic ox-ideftb. Hafmcch romotfen Hae-matcfiorfihyrm (acidj\ ft ctd Haematin I ^«fc Alkaline Haematm r,hlef. Reduced hb, \band rig 15. — Diagram to show the Chief Characteristics by which Haemoglobin and some of its Derivatives may be recognized Spectroscopically. The position of the middle of each band is indicated roughly by a vertical line- yields a large amount of histidin, to which its basic properties are chiefly due From 100 grammes of oxyhaemoglobin about 4 grammes of hae- matin are obtained. As to the pigment moiety, when haemoglobin is acted on by acids in the absence of oxygen, h cemochromogen is first formed, which then gradually loses its iron and is changed into haemato- porphyrin. If oxy- Liv»e 29-3 TTueclts 29'2 s,heirTv lungs Zi'7 Bones 8'Z genital organs 6'3 ^Ain 27 Kidneys 1'6 Nerve centres 7'2 SpJeen "i- Fig. 16. — Diagram to illustrate the Distribution of the Blood in the Various Organs of a Rabbit (after Ranke's Measurements). The numbers are percentages of the total blood. gen be present, hae- matin is the final product. Haematin may be considered as the compound which haemochro- mogen forms with oxygen. By the action of alkalies re- duced haemoglobin yields haemochro- mogen, which is stable in alkaline solution, and gives a beautiful spectrum with two bands, bearing some resemblance to those of oxy- hsemoglobin, but placed nearer the violet end. The band next the red end is much sharper than the other (p. 76). Haemochromogen binds exactly the same amount of oxygen as the haemoglobin from which it is derived, and it is due to the haemochromogen in its molecule that the blood-pigment fulfils its function of taking up and transporting oxygen. Hesmatin (C32H 320^4. FeOH), the most frequent result of the splitting up of haemoglobin, is generally obtained as an amorphous substance with a bluish-black colour and a metallic lustre, insoluble in water, but soluble in dilute alkalies and acids, or in alcohol containing them. In addition to the iron of the haemoglobin, haematin contains the four chief elements of proteins — carbon, hydrogen, nitrogen, and oxgyen (Practical Exercises, p. 75). H (ematoporphyrin (C33H38O6N4), or iron-free haematin, may be ob- tained from blood or haemoglobin by the action of strong sulphuric acid, from haematin or haemin by the action of hydrobromic acid. It 55 is distinguished from these pigments by the fact that it contains no iron. When strong sulphuric acid is allowed to act on blood or haemo- globin solution, haematoporphyrin is also produced, as may be easily shown by the spectroscope. Its spectrum in acid solution shows two bands, one just to the left of D, the other about midway between D and E. Like oxyhsemoglobin, reduced haemoglobin, carbonic oxide haemoglobin, methaemoglobin, and other derivatives of haemoglobin, it also has a band in the ultra-violet. Hamin (C32H 330^4. Fed) is readily obtained from haematin and also from haemoglobin by heating with dilute hydrochloric acid, and also directly from blood, as described in the Practical Exercises, p. 78. It crystallizes in the form of small rhombic plates, of a brownish or brownish-black colour (Fig. 24, p. 78). They are insoluble in water, but readily soluble in dilute alkalies (Practical Exercises, p. 79). Chemistry of the White Blood-Corpuscles. — The composition of pus- cells and the leucocytes of lymphatic glands has alone been investigated. The chief constituents of the latter are a globulin coagulating by heat at 48° to 50° C. ; a nucleo-protein coagulating in 5 per cent, magnesium sulphate solution at 75° C., and causing coagulation of the blood on injection into the veins of rabbits; an albumin coagulating at 73° C. ; and a ferment with powers like the pepsin of the gastric juice. In pus- cells glycogen has been found, and it can be demonstrated micro- chemically in the leucocytes of blood by the iodine reaction in various conditions. Fats, cholesterol, and lecithin are also present, as well as the so-called protagon. The ordinary inorganic constituents have been demonstrated — namely, potassium, sodium, calcium, magnesium, and iron, united with chlorine and phosphoric acid. The total solids amount to n to 12 per cent. SECTION IV. — QUANTITY AND DISTRIBUTION OF THE BLOOD. The Quantity of Blood. — The quantity of blood in an animal is most accurately determined by the method of Welcker. The animal is bled from the carotid into a weighed flask. When blood has ceased to flow the vessels are washed out with water or physiological saline solution, and the last traces of blood are removed by chopping up the body, after the intestinal contents have been cleared away, and extracting it with water. The extract and washings are mixed and weighed ; a given quantity of the mixture is placed in a hasma- tinometer (a glass trough with parallel sides, e.g.), and a weighed quantity of the unmixed blood diluted in a similar vessel till the tint is the same in both. From the amount of dilution required, the quantity of blood in the watery solution can be calculated. This is added to the amount of unmixed blood directly determined. Since haemorrhage is immediately followed by the entrance of liquid into the bloodvessels from the lymph and tissue fluids, somewhat too ; high a result will be obtained if the bleeding is at all prolonged. T t is well, therefore, to take only a moderate amount of blood for din estimation, and to compute the balance by the colorimetric method. Many other methods have been devised on the principle of in- jecting a known quantity of some substance into the circulating blood, and then, after an interval has been allowed for mixture, determining the change produced in a sample. Thus, the specific 56 gravity of a drop of blood having been measured, a certain quantity of a solution of sodium chloride isotonic with the plasma may be injected into a vein, and the specific gravity again determined. Or the electrical resistance of a small sample of blood may be measured before and after injection of a given quantity of isotonic salt solution. The total mass of blood in a living man has been estimated by caus- ing the person to inhale a mixture of carbon monoxide with oxygen or air. The amount of carbon monoxide taken up is determined and also, in a sample of blood taken from the finger the percentage amount to which the haemoglobin has become saturated with carbon monoxide. All that remains is to estimate the volume of carbon monoxide (or, what is precisely the same thing, the volume of oxygen) which 100 c.c. of blood will take up. This latter quantity is called the percentage oxygen capacity. From these data the total volume of the blood can be calculated. If the volume is multiplied by the specific gravity the mass is obtained. Notwith- standing the elegance of this method in principle, it is by no means easy to obtain accurate results with it in practice. It unques- tionably gives values for the total quantity of blood which are too low. A better method is to inject into a vein a measured quantity of a solution of a pigment (" vital red "), which is only slowly elimin- ated and is not taken up to a sensible degree by the erythrocytes. Samples of blood are drawn before and a short time after the injec- tion and the plasma separated from each by the centrifuge. The amount of pigment is then determined which must be added to the first sample of plasma to make its tint the same as that of the second. The total quantity of plasma can thus be calculated and from it, by determining the relative proportion of corpuscles and plasma in the blood, the total quantity of blood (Rowntree, etc.). The quantity of blood in the body was greatly over-estimated by the ancient physicians. Avicenna put it at 25 lb., and many loose statements are on record of as much as 20 lb. being lost by a patient without causing death. By Welcker's method the proportion of blood to body-weight has been found to be in the dog i : 13, cat 1 : 14, horse i : 15, frog i : 17, rabbit 1 : 19, fowl i : 20. In new-born children the proportion was i : 19, in adult human beings (executed criminals) i : 13. By the ' vital red ' method, the amount of plasma was found to be one-twentieth and that of blood one- eleventh to one-twelfth of the body-weight. According to Dreyer, the blood volume is a function of the surface of the body, so that the smaller and lighter animals in any given species have a relatively greater blood volume than the larger and heavier individuals. Accordingly, he considers that the practice of expressing the volume of blood as a percentage of the body-weight should be abandoned. Fig. 16 (p. 54) illustrates the distribution of the blood in the various organs of a rabbit. The liver and skeletal muscles each con- tain rather more than one-fourth; the heart, lungs, and great vessels LYMPH AND CHYLE 57 rather less than one- fourth ; and the rest of the body about one-fifth, of the total blood. The kidney and spleen of the rabbit each contain one-eighth of their own weight of blood, the liver between one-third and one-fourth of its weight, the muscles only one-twentieth of their weight. SECTION V. — LYMPH AND CHYLE. Lymph has been denned as blood without its red corpuscles (Johannes Miiller); it resembles, in fact, a dilute blood-plasma, containing leucocytes, some of which (lymphocytes) are common to lymph and blood, others (coarsely granular basophile cells, present only in small numbers) are absent from the blood. Lymph also contains thrombocytes. The reason of this similarity appears when it is recognized that the plasma of tissue-lymph (p. 460) is derived, in large part at any rate, from the plasma of blood by a process of physiological nitration (or secretion) through the walls of the capillaries into the lymph-spaces that everywhere occupy the inter- stices of areolar tissue, while the lymph of the lymphatic vessels is in turn derived from the tissue fluid. But in addition to the con- stituents of the plasma, lymph contains substances produced in the metabolism of the tissues which pass into it directly. As collected from one of the large lymphatic vessels of the limbs, or from the thoracic duct of a fasting animal, lymph is a colourless or some- times yellowish or slightly reddish liquid of alkaline reaction. Its specific gravity is much less than that of the blood (1015 to 1030). It coagulates spontaneously, but the clot is always less firm and less bulky than that of blood. The plasma contains fibrinogen, from which the fibrin of the clot is derived. Serum-albumin and serum- globulin are present in much the same relative proportion as in blood, although in smaller absolute amount. Neutral fats, urea, and sugar are also found in small quantities. The inorganic salts are the same as those of the blood-serum, and exist in about the same amount, sodium preponderating among the bases, as it does in serum. The following table shows the results of analyses of lymph from man and the horse (Munk) : Man. Horse. Water .... 95-0 per cent. 95-8 per cent. [Fibrin O-I } O-I \ Other proteins - 4'i 2-9 Solids ] Fat - I Extractives* trace j- 5-0 o-3 trace V 4*2 O-I [Salts - --. 0-5 ) i-i ) * The term ' extractives ' is somewhat loosely applied to organic substances which exist in so small an amount, or have such indefinite chemical characters, that they cannot be separately estimated, and are extracted together from the residue by various solvents. THE CIRCULATING LIQUIDS OF THE BODY Chyle is merely the name given to the lymph coming from the alimentary canal. The fat of the food is absorbed by the lym- phatics, and during digestion the chyle is crowded with fine fatty globules, which give it a milky appearance. There may also be in chyle a few red blood-corpuscles, carried into the thoracic duct by a back-flow from the veins into which it opens. Chyle clots like ordinary lymph, the size of the clot varying according to the quantity of fat present and enmeshed by the fibrin. Wounds of the thoracic duct or of lymphatics opening into it are occasionally produced in operations on the neck, and when these remain open chyle may be readily collected. In samples obtained from a patient only a week after the section of a branch of the duct during an operation for the removal of tubercular glands, water constituted 928-90 parts in 1,000, total solids 71-10, inorganic solids 6-04, organic solids 65-06, proteins 18-52, ether extract (fatty substances) 19-30 (Sollmann). The following is the composition of a sample analyzed by Paton, and obtained from a fistula of the thoracic duct in a man : Water Solids Inorganic - Organic - Proteins Fats - Cholesterin Lecithin 953-4 46-6 40-1 13-7 24-06 0-6 0-36 The quantity of chyle flowing from the fistula was estimated at as much as 3 to 4 kilos per twenty-four hours, or nearly as much as the whole of the blood. The flow has been calculated in various animals at one- eighteenth to one-seventh of the body-weight in the twenty- four hours. The quantity of lymph in the body is unknown, but it must be very great — perhaps two or three times that of the blood. Allied to tissue-lymph, but not identical with it, are the fluids present in health in very small amount in such serous cavities as the pericardium. The synovial fluid of the joints differs from lymph especially in containing a small amount of a mucin-like substance. The aqueous humour, and still more the cerebro-spinal fluid, are characterized by a marked deficiency in solids, especially protein. In the following table (from Spiro) the differences in the composition of lymph and allied fluids from different parts of the body are illus- trated. Man : Ly:.".'r'h from FistiM ia Thigh. Horse : Lymph from Xeck during MasticStion. Aqueous Humour. Cerebro-Spinal Fluid. Water - - 06-4 to 94-3 95 98-7 99 to 99-2 Salts - - O-y ,, 0-87 o-75 0-5 to 0-8 Fat - - Protein O-O6 ,, O-22\ 2-8 ,, 4-8 J 3'7 0-72 0-02 to 0-16 FUNCTIONS OF BLOOD AND LYMPH 59 The gases of the blood and lymph will be treated of in Chapter IV., the formation of lymph in Chapter VIII., its circulation in Chapter III. SECTION VI. — THE FUNCTIONS OF BLOOD AND LYMPH. We have already said that these liquids provide the tissues with the materials they require, and carry away from them materials which have served their turn and are done with. These materials are gaseous, liquid, and solid. Oxygen is brought to the tissues in the red corpuscles ; carbon dioxide is carried away from them partly in the erythrocytes, but chiefly in the plasma of the blood and lymph. The water and solids which the cells of the body take in and give out are also, at one time or another, constituents of the plasma. The heat produced in the tissues, too, is, to a large extent, conducted into the blood and distributed by it throughout the body. The leucocytes, as will be seen farther on, aid in some measure in the absorption of certain of the food substances from the intestine. It is not known whether, apart from this, they play any role in the normal nutrition of other cells, although it is probable that they exercise an influence on the plasma in which they live. But they have impor- tant functions of another kind, to which it is necessary to refer briefly here. Phagocytosis. — Certain of the amoeboid cells of blood and lymph, and the cells of the splenic pulp, are able to include or ' eat up ' foreign bodies with which they come in contact, in the same way as the amoeba takes in its food. Such cells are called phagocytes ; and it is to be remarked that this term neither comprises all leucocytes nor excludes all other cells, for some fixed cells, such as those of the endothelial lining of bloodvessels, are phagocytes in virtue of their power of sending out protoplasmic processes, while the small, relatively immobile lymphocyte is not a phagocyte. Although it is not at present possible to assign a physiological value to all the phenomena of phagocytosis, either as regards the phagocytes themselves or as regards the organism of which they form a part, there seems little doubt that under certain circumstances the process is connected with the removal of structures which in the course of development have become obsolete, or with the neutral- ization or elimination of harmful substances introduced from with- out, or formed by the activity of bacteria within the tissues. During the metamorphosis of some larvae, groups of cilia and muscle-fibres may be absorbed and eaten up by the leucocytes. In the metamor- phosis of maggots, for example, the muscular fibres of the abdominal wall, which are used in creeping, and are therefore not required in the adult, degenerate, and are devoured by swarms of leucocytes which migrate into them. In the human subject an example of 60 THE CIRCULATING LIQUIDS OF THE BODY absorption of tissue by the aid of leucocytes is the removal of the necrosed decidua reflexa, the fold of uterine mucous membrane which envelops the ovum (Minot). The behaviour of phagocytes towards pathogenic micro-organisms is of even greater interest and importance. Metchnikoff laid the foundation of our knowledge of this subject by his researches on Daphnia, a small crustacean with transparent tissues, which can be observed under the microscope. When this creature is fed with a fungus, Monospora, the spores of the latter find their way into the body-cavity. Here they are at once attacked by the leucocytes, ingested, and destroyed. But after a time so many spores get through that the leucocytes are unable to deal with them all ; some of them develop into the first or ' conidium ' stage of the fungus ; the conidia poison the leucocytes, instead of being destroyed by them, and the animal generally dies. Occasionally, however, the leuco- cytes are able to destroy all the spores, and the life of the Daphnia is preserved. This battle, ending sometimes in victory, sometimes in defeat, is believed by Metchnikoff to be typical of the struggle which the phagocytes of higher animals and of man seem to engage in when the germs of disease are introduced into the organism. He supposes that the immunity to certain diseases possessed naturally by some animals, and which may be conferred on others by vaccina- tion with various protective substances, is, to a large extent, due to the early and complete success of the phagocytes in the fight with the bacteria; and that in rapidly- fatal diseases — such as chicken- cholera in birds and rabbits, and anthrax in mice — the absence of any effective phagocytosis is the factor which determines the result. Others have laid stress on the action of protective substances sup- posed to exist in the plasma itself. It is possible that such sub- stances are manufactured by the leucocytes, and either given off by them to the plasma by a process of ' excretion/ or liberated by their complete solution. The most recent investigations go to show that Metchnikoff's phagocytic theory of immunity requires modification, at any rate in the case of the higher animals and man, although the brilliant biological observations on which it was originally built retain all their value. He supposed that in the immunizing process the leucocytes underwent certain changes, acquired, so to speak, a sort of ' education ' that enabled them to cope with bacteria against which they were previously powerless. It seems more probable that in the presence of the substances that confer immunity, not only the leucocytes, but other cells, are stimulated to produce bodies which cut short the life, or inhibit the growth, of the bacteria (alexins), or prepare them for being taken up by the phagocytes (opsonins). It has been shown that bacteria which have been in contact with serum containing the appropriate opsonins are taken FUNCTIONS OF BLOOD AND LYMPH 61 up readily by leucocytes washed free from serum constituents by physiological salt solution, whereas the washed leucocytes either do not ingest bacteria which have not been acted on by serum, or take them up in much smaller numbers. There is some evidence that in certain bacterial infections — for example, chronic furunculosis, a condition in which crops of boils continue to appear — the grip of the bacteria on the body is perpetuated by a deficiency in the amount or in the activity of opsonins capable of acting specifically upon the micro-organisms in question. A numerical expression, which in certain cases, perhaps, gives a measure of the patient's resistance to the infection, has been worked out by Wright under the name ' opsonic index.' This index is the ratio between the average number of bacteria taken up, under certain fixed conditions, by each polymorphonuclear leucocyte in an emulsion made with the patient's serum, and the average number taken up by similar leucocytes in an emulsion made with normal serum. The significance of this index and even the practicability of the methods used to ascertain it, are still the subject of discussion. Diapedesis. — The fact that leucocytes can pass out of the blood- vessels into the tissues has a very important bearing on the subject of phagocytosis. The phenomenon is called diapedesis, and is best seen when a transparent part, such as the mesentery of the frog, is irritated. The first effect of irritation is an increase in the flow of blood through the affected region. If the irritation continues, or if it was originally severe, the current soon begins to slacken, the corpuscles stagnate in the vessels, and inflammatory stasis is pro- duced. The leucocytes adhere in large numbers to the walls of the capillaries, and particularly of the small veins, and then begin to pass slowly through them by amoeboid movements, the passage taking place at the junctions between, or it may be through the substance of, the endothelial cells. Plasma is also poured out into the tissues, the whole forming an inflammatory exudation. Even red blood- corpuscles may pass out of the vessels in small numbers. The exudation may be gradually reabsorbed, or destruction of tissue may ensue, and a collection of pus be formed. The cells of pus are emigrated leucocytes (Practical Exercises, Chap. III., p. 193). Their emigration is connected with the defence of the organism against the entrance of certain forms of bacteria at the seat of injury, and with the repair of the injured tissue, but the nature of the summons which gathers them there is not yet clearly under- stood. It is probably some sort of chemical attraction (chemio- taxis) between constituents of the bacteria or decomposition prod- ucts of the injured tissue on the one hand, and constituents of the leucocytes on the other. As for the blood-plates, it will suffice to say by way of summary that their important function in the sealing of wounded vessels (p. 46) 62 THE CIRCULATING LIQUIDS OF THE BODY is the sole office which at present can be attributed to them. And if it is permissible to consider the leucocytes as a patrol for the defence of the tissues in general against invading micro-organisms, it may perhaps not be too far-fetched an idea to look upon the blood- plates as essentially a patrol in the interests of the anatomical integrity of the vascular system itself. This does not exclude the possibility that the clotting of extravasated plasma may furnish a more favourable medium for the processes of repair in all injured tissues. PRACTICAL EXERCISES ON CHAPTER II. N.B. — In the following exercises all experiments on animals which would cause the slightest pain are to be done under complete ancesthesia. 1. Reaction of Blood. — (i) Put a drop of fresh dog's or ox blood on a piece of glazed neutral litmus paper (the litmus paper can be glazed by dipping it into a neutral solution of gelatin and allowing it to dry). Wash the blood off in 10 to 30 seconds with distilled water. A bluish stain will be left, showing that fresh blood is alkaline. (2) Repeat with dog's or ox serum. It is not necessary to wash the serum off, as it does not obscure the change of colour. (3) Repeat (i) with human blood. With a clean suture-needle or a good-sized sewing-needle which has been sterilized in the flame of a Bunsen burner, prick one of the fingers behind the nail. Bandaging the finger with a handkerchief from above downwards, so as to render its tip congested, will often facilitate the getting of a good-sized drop, but for quantitative experi- ments, like 2, 10, and 17 (4), this should not be done. 2. Specific Gravity of Blood — Hammer schlag's Method. — (i) Put a mixture of chloroform and benzol of specific gravity i -060 into a small glass cylinder. Put a drop of dog's or ox defibrinated blood into the mixture by means of a small pipette. If the drop sinks add chloroform, if it rises add benzol, till it just remains suspended when the liquid has been well stirred. Then with a small hydrometer measure the specific gravity of the mixture, which is now equal to that of the blood. Filter the liquid to free it from blood, and put it back into the stock-bottle. (2) Obtain a drop of human blood as in i, and repeat the measurement of the specific gravity. 3. Coagulation of Blood.* — (i) Take three tumblers or beakers, label them a, £, and y, and measure into each 100 c.c. of water. Mark the level of the water by strips of gummed paper, and pour it out. (If a sufficient number of graduated cylinders is available, they may of course be used, and this measurement avoided.) Into a put 25 c.c of a saturated solution of magnesium sulphate, into ft 25 c.c. of a i per cent, solution of potassium or ammonium oxalate in 0-9 per cent, solution of sodium chloride, and into y 25 c.c. of a 1-2 per cent, solution of sodium fluoride in 0-9 per cent, salt solution If the dog provided is a large one, these quantities may be all doubled ; for a small dog they may be all halved. * This experiment requires two laboratory periods, the various blood mix- tures being obtained during the first and worked up during the second. (2) Insert a cannula into the central end of the carotid artery of a dog anaesthetized with morphine* and ether, or A.C.E. mixture. f To put a Cannula into an Artery. — Select a glass cannula of suitable size, feel for the artery, make an incision in its course through the skin, then isolate about an inch of it with forceps or a blunt needle, carefully clearing away the fascia or connective tissue. Next pass a small pair of forceps under the artery, and draw two ligatures through below it. If the cannula is to be inserted into the central end of the artery, tie the ligature which is farthest from the heart, and cut one end short. Then between the heart and the other ligature compress the artery with a small clamp (often spoken of as ' bulldog ' forceps). Now lift the artery by the distal ligature, make a transverse slit in it with a pair of fine scissors, insert the cannula, and tie the ligature over its neck. Cut the ends of the ligature short. If the cannula is to be put into the distal end of the artery, both ligatures must be between the clamp and the heart, and the bulldog must be put on before the first ligature (the one nearest the heart) is tied, so that the piece of bloodvessel between it and the ligature may be full of blood, as this facilitates the opening of the artery. (3) Run into a, /3, and y enough blood to fill them to the mark. Shake the vessels, or stir up once or twice with a glass rod, to mix the blood and solution. (4) Take a small thin copper or brass vessel, and place it in a freezing mixture of ice and salt. Run into it some of the blood from the artery. It soon freezes to a hard mass. Now take the vessel out of the freezing mixture and allow the blood to thaw. It will be seen that it remains liquid for a short time and then clots. (5) Run some of the blood into a porcelain capsule, stirring it vigorously with a glass rod. The fibrin collects on the rod; the blood is defibrinated and will no longer clot. (6) Now let some blood run into a small beaker or jar. Notice that the blood begins to clot in a few minutes, and that soon the vessel can be tilted without spilling it. Note the time required for clotting to occur. Set the coagulated blood aside, and observe next day that clear yellow serum has separated from the clot. (7) Weigh out a quantity of Witte's ' peptone ' equivalent to 0-5 gramme for every kilo of body-weight of the dog. Dissolve the peptone in about twenty times its weight of 0-9 per cent, salt solution. Put a cannula into the central end of a crural vein (p. 212). Fill the cannula with the peptone solution and connect it with a burette. Put 15 drops of the peptone solution into a test-tube labelled ' Peptone A.' Put the rest into the burette, and see that the connecting tube is filled with the solution and free from air. Run into the test-tube about 5 c.c. of blood from the cannula in the carotid. Now let the peptone solution flow from the burette into the vein. Feel the pulse over the heart as the solution is running in. If the heart becomes very weak, stop the injection ; otherwise the animal may die from the great lower- ing of blood-pressure (p. 214). As soon as the injection is finished, draw off a sample of 5 c.c. of blood into a test-tube labelled ' Pep- tone B,' and let it stand. In ten minutes collect five further samples of 5 c.c. (' Peptone C, D, E, F, G '), and a large one, H; in half an hour * One to 2 centigrammes of morphine hydrochlorate per kilogramme of body-weight should be injected subcutaneously about half an hour before the operation. Ten c.c. of a 2 per cent, solution is sufficient for a dog of good size. Note that diarrhoea and salivation .are caused by such a dose. For directions for fastening the dog on the holder, see footnote on p. 199. f A mixture of i part of alcohol, 2 of ether, and 3 of chloroform. 64 THE CIRCULATING LIQUIDS OF THE BODY another set of five small samples, and at as long an interval as possible thereafter five more. Now letting the dog bleed to death, observe that the flow of blood is temporarily increased by pressure on -the abdominal walls, which squeezes it towards the heart, by passive mavements of the hind-legs, and also during the convulsions of asphyxia, which soon appear. Add to the peptone blood D '5 c.c. of serum, to E a little sodium chloride extract of liver, to F a little extract of muscle, and to G 15 drops of a 2 per cent, solution of calcium chloride, and put C, D, E, F, and G into a water-bath at 40° C. Treat the other sets of small samples in the same way. Note how long each specimen takes to clot, and report your results.* (8) Observe that the blood in a, 8, and y has not coagulated. Label four test-tubes ' Oxalate A, B, C, D,' and put into each about 5 c.c. of the oxalated blood. Add to A and B 5 or 6 drops of a 2 per cent, solution of calcium chloride, to C 12 drops, and to D as much as there is of the blood. Leave A at the ordinary temperature, put the other test-tubes in a water-bath at 40° C., and note when clotting occurs. (9) To 10 c.c. of the fluoride blood add a little more CaCl2 than is re- quired to combine with the excess of fluoride present . Label four test- tubes ' Fluoride A, B, C, D,' and into each put about 2 c.c. of this ' recalcified ' fluoride blood. To B add i c.c. liver extract, to C I c.c. muscle extract, and to D 4 c.c. water. Label two more test-tubes ' Fluoride E and F.' Into each put 2 c.c. of the fluoride blood without CaCl2. Add also to E i c.c. liver extract and to F i c.c. serum. Put all the tubes in a bath at about 40° C., and observe in which and in what time coagulation takes place. (10) By means of a centrifuge (Fig. 17) separate the plasma from the corpuscles in a, ft, and y, and also from the peptone blood. With the oxalate plasma from /3, and the fluoride plasma from y, repeat the observations in (8) and (9), using smaller quantities of the plasma, if necessary, in small test- tubes. With the plasma from a perform the following experiments: Put a small quantity of the plasma (i c.c.) into four test-tubes, labelling * Sometimes the injection of peptone hastens coagulation instead of hinder- ing it. It has been asserted that this is only the case when small doses are used (less than 0*02 gramme per kilo of bod y- weight) . But in 2 dogs out of ii a dose of 0^5 gramme per kilo has been seen to hasten coagulation, and in i out of 12 to leave it unaffected; in the other 9 coagulation was markedly retarded Fig. 17. — Centrifuge (Jung), The four cylinders shown at the top of the figure are so swung that they become horizontal as soon as speed is up. PRACTICAL EXERCISES 65 them ' Magnesium Sulphate A, B, C, D.' Dilute B with four times, C with eight times, and D with twenty times as much distilled water as was taken of the plasma. Observe in which, if any, coagulation occurs, and the time of its occurrence, and report the result. (n) With peptone plasma from H and from the peptone blood obtained later repeat the experiments done in (7). In addition dilute i c.c. of the plasma with three volumes of water and i c.c. of it with ten volumes of water, and put in the bath at 40° C. Observe whether clotting occurs. Instead of dog's blood, the blood of an ox or pig may be obtained at the slaughter-house. 4. Preparation of Fibrin-Ferment. — Precipitate blood- serum with ten times its volume of alcohol. Let it stand for several weeks, then extract the precipitate with water. The water dissolves out the fibrin- ferment, but not the coagulated serum proteins. 3. Preparation of Tissue Extracts containing Thrombokinase. — In a dog or rabbit killed by bleeding insert a cannula into the lower end of the thoracic aorta. Fill the cannula with 0-9 per cent, salt solution, and connect it with a bottle also containing salt solution. Wash put the vessels of the lower portion of the body, making an opening in the inferior vena cava above the diaphragm to allow the liquid to escape. For the sake of cleanliness, a cannula armed with a piece of rubber tubing should be inserted for this purpose into the inferior vena cava. Continue the injection till the liquid issues colour- less. Then remove portions of liver and muscle. Mince each separately. Rub up with sand in a mortar. Add 0-9 per cent, sodium chloride solution and rub up again. Put into bottles and keep in the ice-chest. For use take off some of the liquid from the top with a pipette, or strain through cheese-cloth. 6. Serum. — Test the reaction, and determine, both by the hydrom- eter and the pycnometer, or specific gravity bottle, the specific gravity of the serum provided, or of the serum obtained in experi- ment 3. Serum Proteins. — (i) Saturate serum with magnesium sulphate crystals at 30° C. The serum-globulin is precipitated. Filter off. Wash the precipitate on the filter with a saturated solution of mag- nesium sulphate. Dissolve the precipitate by the addition of a little distilled water, and perform the following tests for globulins : (a) Satu- rate with magnesium sulphate. A precipitate is obtained, (b) Drop into a large quantity of water, and a flocculent precipitate falls down. (c) Heat. Coagulation occurs. Determine the temperature of coagu- lation (p. 9). (2) To a portion of the filtrate from (i) add sodium sulphate to saturation. The serum-albumin is precipitated. (Neither magnesium sulphate nor sodium sulphate precipitates serum-albumin alone, but the double salt sodio-magnesium sulphate precipitates it, and this is formed when sodium sulphate is added to magnesium sulphate.) (3) Dilute another portion of the filtrate from (i) with its own bulk of water. Very slightly acidulate with dilute acetic acid, and de- termine the temperature of heat coagulation. (4) Precipitate the serum-globulin from another portion of serum by adding to it an equal volume of saturated solution of ammonium sulphate. Filter. Precipitate the serum-albumin from the filtrate by saturating with ammonium sulphate crystals. (5) Dilute serum with ten to twenty times its volume of distilled water, and pass through it a stream of carbon dioxide. The serum- globulin is partially precipitated. 5 66 THE CIRCULATING LIQUIDS OF THE BODY (6) Acidulate some serum with dilute acetic acid and boil. Filtei off the coagulum, and to the filtrate add silver nitrate. A non-protein precipitate insoluble in nitric acid, but soluble in ammonia, indicates the presence of chlorides. 7. Action of Serum on Artery Rings. — Cut a number of rings about i£ millimetres wide from a fresh carotid artery of the sheep, obtained from the slaughter-house. Keep the rings in a dish in Ringer's solu- tion.* They should be as nearly as possible of uniform width. A small cylindrical glass vessel is supported on a stand in such a way that it can be easily lowered into a bath of water kept at a temperature of about 39° to 40°. A stock of Ringer's solution is kept in a beaker or bottle immersed in the bath. A ring of the artery is put into the small cylinder, where it is held between two aluminium hooks, one fastened to the bottom of the cylinder, the other (the upper one) con- nected with the short arm of a lever, the long arm of which is arranged to write on a slowly revolving drum. A time-trace, say in half- minutes, is recorded below. The small cylinder is now filled with warm Ringer's solution and lowered into the bath. Oxygen is bubbled through the solution by means of a side-tube near the bottom of the cylindrical vessel. The artery ring is now stretched for five minutes by a weight of 10 grammes attached to the long arm of the lever at the same distance from the axis as that at which the ring is attached. After the stretching period the weight is removed, and a little tune allowed to elapse tUl the writing-point traces a horizontal line on the drum. Then a bent pipette is filled with serum already heated to bath temperature in a vessel immersed in the bath. The pipette is intro- duced into the small cylinder so that its point is at the bottom, without disturbing the ring, and the serum is allowed to run in till the Ringer solution is displaced. The ring shortens under the influence of the con- strictor substance in the serum, and the tracing is continued till the shortening has reached its maximum and the trace is again horizontal (Fig. 3, p. 46). Various dilutions of the serum are now made with Ringer's solution, and the greatest dilution in which the serum will still cause a percep- tible constriction of the rings is determined. This affords a measure of comparison with other sera of the strength of the constrictor action. For each dilution of serum a separate ring must be used. It must be remembered that comparisons of this kind can only be made with arteries of the same sensitiveness, and different arteries vary much in this regard. 8. Comparison of the Action of Serum and Adrenalin (Epinephrin) on Artery Rings. — Tracings showing the effect of various dilutions of adrenalin chloride on artery rings may now be taken for comparison with the serum effects. The adrenalin dilutions should be made just before use, as adrenalin is rapidly oxidized. Or a separate experiment on the action of adrenalin may be made under ' Circulation,' as on p. 216. 9. Comparison of the Action of Serum and Plasma on Artery Rings. — Citrate plasma is obtained as follows: A cannula and attached rubber tube are boiled, oiled inside with fresh olive-oil, and filled with a citrate * This is the name given to a solution containing the most important of the inorganic constituents of blood -serum in approximately the normal pro- portions. The various ' Ringer's solutions ' used by different workers have varied slightly. That recommended by Locke (for perfusion of the isolated heart) contains NaCl, 0-9 per cent.; KC1, 0*042 per cent.; CaCl2, 0-024 per cent.; NaHCO3, o-oi to 0-03 per cent.; with in addition o'i per cent, of dextrose, which can be omitted for such experiments as 7. PRACTICAL EXERCISES «7 solution made by dissolving sodium citrate in Ringer's solution to the extent of- 2 per cent. The solution is prevented from escaping by a clip on the tube. The cannula is inserted into the carotid of a dog, the end of the rubber tube dipped below a quantity of citrate- Ringer solution in a beaker, and a volume of blood equal to that of the solution run in. Then the blood and solution are at once stirred gently, but sufficiently to insure proper admixture. Some blood is now run into another vessel, defibrinated, and measured. An equal volume of the citrate-Ringer solution is added to it while the mass of fibrin is still floating in the blodd. After mixing, the fibrin is removed. Plasma is then separated by the centrifuge from the first specimen of blood, and serum from the second, and comparison experiments are made with each on artery rings. If the plasma has been properly obtained, it will have little constrictor effect on the rings in comparison with the serum. In making the comparison, arteries which give a decided effect with serum should be employed. The defibrinated blood and the unclotted citrate blood may also be used for the comparison. Fig. 18. — Thoma-Zeiss Haemocytometer. M , mouthpiece of tube G, by which blood is sucked into S; E, bead for mixing; a, view of slide from above; b, in section; c, squares in middle of B, as seen under microscope. 10. Enumeration of the Blood - Corpuscles. — Use the Thoma-Zeiss apparatus (Fig. 18). (i) Suck a drop of ox or dog's blood up into the capillary tube 5 to the mark i. Wipe off any blood which may adhere to the end of the tube. Then fill it with 3 per cent, sodium chloride to the mark 101. This represents a dilution of 100 times. Mix the blood and solution thoroughly, then blow put a drop or two of the liquid to remove all the solution which remains in the capillary tube. Now fill the shallow cell B with the blood mixture. Put the cover-glass on, taking care that it does not float on the liquid, but that the cell is exactly filled. Put the slide under the microscope (say Leitz's oc. III., obj. 5), and count the number of red corpuscles in not less than ten to twenty squares. Sixteen squares is a good routine number. The greater the number of squares counted, the nearer will be the approximation to the truth. Now take the average number in a square. The depth of the cell is ^ mm., the area of each square jj^ sq. mm. The volume of the column of liquid standing upon a square is 1T^ cub. mm. One cub. mm- of the diluted blood would therefore contain 4,000 times as many corpuscles as one square. But the blood has been diluted 100 times, therefore i cub. mm. of the 6R THE CIRCULATING LIQUIDS OF THE BODY undiluted blood would contain 400,000 times the number of corpuscles in one square. Suppose the average for a square is found to be 13. This would correspond to 5,200,000 corpuscles hi i cub. mm. of blood. Compare your result with the true number supplied by the demon- strator. (2) Prick the finger to obtain a drop of blood, and repeat the crunt as in (i).* To Count the White Corpuscles. — Add to i part of blood 9 parts of i per cent, acetic acid, in order to lake the coloured corpuscles and render it easy to see the leucocytes. II. Relative Volume of Corpuscles and Plasma by Haematocrite. — (i) For practice, fill the two graduated glass tubes with the defibrinated blood of an animal. The rubber tube with mouthpiece supplied with the apparatus is to be attached to the glass tube, and the blood sucked up. Press the tip of the index-finger against the pointed end, and care- fully remove the rubber tube . Place the tube in the haematocrite frame , blunt end outwards — that is, farthest from the axis of rotation — and then slip the pointed end down into position against the spring. Instead of the rubber tube, a special suction pipette for automatically filling the graduated tubes may be employed (Daland) . Attach the haemato- crite frame to the centrifuge, and rotate till the volume of sediment Fig. 19.— Haematocrite. A, haematocrite attachment with graduated tubes; B, auto- matic pipette for filling the tubes (Daland). (corpuscles) ceases to diminish. The graduations are best read with a hand lens. The leucocytes will be seen to form a thin whitish line proximal to the column of red corpuscles. (2) Prick the finger or the lobe of the ear, fill the tubes as in (i), and centrifugalize. Everything must be done as rapidly as possible, so that the blood may not clot till the separation of plasma and corpuscles is completed. The centrifuge must rotate very rapidly (about 10,000 revolutions a minute) for two or three minutes. For clinical purposes it is best to rotate the centrifuge always at the same speed for the same length of time rather than to aim at reaching a constant length of the column of corpuscles. In this way useful comparative results can be obtained. It is well, to avoid the risk of accident, to rotate the centrifuge under a guard. 12. Electrical Conductivity of Blood. — (i) Fill a small U-tube with blood up to a mark. In each limb insert a platinum electrodef con- * If the tube has not been properly filled, blow the blood out immediately. On no account permit it to clot in the capillary tube. t If the platinum electrodes are of good size and the resistance of the tube of liquid considerable, it is not necessary to platinize them — i.e., to cover them by electrolysis of a solution of platiuic chloride with a layer of platinum-black. PRACTICAL EXERCISES 69 nected with a holder, which insures that the electrode shall always dip to the same depth into the tube. Arrange the U-tube so that it is immersed at least to the mark in water of constant temperature. Water running freely from the cold-water tap into and out of a large vessel will have a sufficiently constant temperature for the purpose. A ther- mometer must be fixed in the water with its bulb in contact with the U-tube. Connect the electrodes with a resistance -box in the Wheat- stone's bridge arrangement (Fig. 231, p. 726), so that the U-tube occupies the position of the unknown, resistance CD. Instead of the battery F, connect the poles of the secondary of a small induction-coil, arranged for an interrupted current, with A and C, and instead of the galvanometer G insert a telephone. The resistances AB and AD will be obtained by taking out two plugs from the appropriate part of the resistance-box. Whether AB and AD should be equal (say, 10 : 10, 100 : 100, or 1,000 : 1,000 ohms) or unequal (say, 10 : 100, or 100 : 1,000, or 10 : 1,000 ohms) will depend upon the resistance of the tube of liquid to be measured. Take out from the part of the box corresponding to BC a plug representing a resistance something like that which the tube of blood is expected to have. Close the primary circuit of the induction-coil, and apply the telephone to the ear. A buzzing sound will be heard, which will be louder the farther from the true resistance of the tube the resistance taken out of the box is. Go on altering the resistance in the box by taking out or putting in plugs till the sound disappears, or is reduced to a minimum. The tempera- ture of the water should now be read off. The resistance of the tube of blood for this temperature can easily be calculated from the formula on p. 726. It increases about 2 per cent, for each degree Centigrade of diminution of temperature. The conductivity is the reciprocal of the resistance. By determining once for all the resistance of the tube when filled with a standard solution of a salt whose conductivity is known, the specific conductivity of the blood can be expressed in definite units, but this is not necessary for the purposes of the student. Compare the resistances of defibrinated blood, serum, 0*9 per cent, sodium chloride solution, and a sediment of blood-corpuscles separated by centrifugalization. (2) Instead of the resistance-box a wire mounted on a scale may be used for the resistances AB, AD, the ends of the wire being connected at B and D. A slider with an insulated handle moving along the graduated wire is joined by a flexible wire with one pole of the secondary coil, the other pole being connected at C. The resistance BC is consti- tuted by a rheostat from which a fixed resistance can be taken out. Instead of obtaining the minimum sound in the telephone by varying the resistance BC in the box, the measurement is made by varying the position of the slider; iu other words, by changing the ratio AB: AD. (3) If no rheostat is available instructive comparative measurements may still be made with the graduated wire by substituting for the resistance BC a U-tube of another liquid. If the tubes are of the same dimensions, and the liquids with which they are filled are approximately at the same initial temperature, it is not necessary to immerse them in water at constant temperature. It is sufficient to place them side by side in the air. Perform the following experiments in this way : (a) Label the tubes A and B. Fill them both to the mark with 0-9 per cent. NaCl solution. Connect as in the figure, and move the slider along the wire till the sound is a minimum. Probably the two tubes are not exactly of the same dimensions, and therefore the slider will not be exactly in the middle of the wire. Suppose it is at 49-0 TO THE CIRCULATING LIQUIDS OF THE BODY the total length of the wire being 100. Then resistance of A : resistance of B:: 49-0: 51-0, i.e., resistance of A = — resistance of B. (6) Fill A with defibrinated blood, keeping B filled with NaCl solu- tion, and repeat the measurement. The slider must now be moved much farther away from the zero of the scale. Suppose the minimum sound is obtained with the slider at 70*0. Then resistance of blood = * x 2- resistance of the NaCl solution. 7 49 (c) Compare in the same way the resistance of serum with that of the NaCl solution. It will be found much less than that of the blood. (d) Centrifugalize some of the blood for as long as is convenient, and compare the resistance of the blood from the top of the tubes and from the bottom of the tubes with that of the NaCl solution. The resistance of the blood from the bottom of the tubes will be found much greater than that of the blood from the top. 13. Opacity of Blood. — Smear a little fresh blood on a glass slide, and hold the slide above some printed matter. It will not be possible to read it, because the light is reflected from the corpuscles in all directions, and little of it passes through. 14. Laking of Blood by Chemical and Physical Agents. — (i) Put a little fresh blood into three test-tubes, A, B, and C. Dilute A with an equal volume, B with two volumes, and C with three volumes, of dis- tilled water, and repeat experiment 9. The print can now be read probably through a layer of A, but certainly through B and C, since the haemoglobin is dissolved out of the corpuscles by the water and goes into solution, the blood becoming transparent or laked. That the difference is not due merely to dilution can be shown by putting an equal quantity of blood in two test-tubes, and gradually diluting one with distilled water and the other with a 0-9 per cent, solution of sodium chloride, which does not dissolve out the haemoglobin. Print can be read through the first with a smaller degree of dilution than through the second. Examine the laked blood with the microscope for the ' ghosts,' or shadows of the red corpuscles. The addition of a drop or two of methylene blue will render them somewhat more distinct. (2) Heat a little dog's or ox blood in a test-tube immersed in a water- bath. Put a thermometer in the test-tube, taking care that there is enough blood to cover the bulb. Keep the temperature about 60° C. In a few minutes the blood becomes dark and laking occurs. (3) (a) Put a little blood into each of four test-tubes. To one add a little ether, to another a little chloroform, to the third dilute acetic acid in 0-9 per cent. NaCl, and to the fourth a dilute solution of bile salts (or of sodium taurocholate) in 0-9 per cent. NaCl solution. Laking occurs in all. (6) To 5 c.c. of blood add 0-5 c.c. of a 3 per cent, solution of saponin in 0-9 per cent. NaCl solution, and put the mixture at 40° C. Laking soon occurs. (c) Using a 10 per cent, dilution of blood (blood to which nine volumes of NaCl solution have been added) or a 5 per cent, suspension of washed corpuscles in NaCl solution (i.e., a suspension of corpuscles which have been washed free from serum by being repeatedly mixed with NaCl solution and centrifugalized), determine the minimum dose of 0-3 per cent, saponin solution which will just cause complete laking. Keep the tubes at about 40° C., and observe them from time to time. Now add to some of the 10 per cent, dilution or the 5 per cent, suspension of blood an equal volume of serum from the same kind of blood, and repeat the determination of the minimum dose of saponin necessary for laking. PRACTICAL EXERCISES 71 It will be found that more is now required. The cholesterin in the serum neutralizes the action of some of the saponin. (4) (a) Put i c.c. of blood into each of two test-tubes. To one add I c.c. of 2 per cent, aqueous solution of urea, and to the other 3 c.c. Laldng will take place in the second, whether this has been the case in the first or not. (b) Repeat the experiment with a 2 per cent, solution of urea in 0-9 per cent. NaCl solution. Laking does not occur. This shows that the urea in the first experiment did not act as a haemolytic agent. Laking occurred because urea penetrates the corpuscles easily, and therefore, although the freezing-point of the urea solution is not very different from that of the NaCl solution, its actual osmotic pressure, in relation to the envelopes of the corpuscles, is very much less, and the laking is really water-laking. (5) Put some blood into a flask or test-tube, cork up, and let it stand till it begins to putrefy. It becomes laked. The same occurs when the blood is collected aseptically in a sterile tube and sealed up, although it takes a longer time for the laking to become complete. (6) With blood containing nucleated corpuscles (necturus, frog or chicken) diluted with isotonic salt solution, perform the following experiments under the microscope : (a) With a glass rod drawn to a fine point put a small drop of blood on a slide, and near it a drop of distilled water. Carefully lower the cover-slip and observe the interface with the microscope, first with the low and then with the high power. Then mix and see complete laking. Add a little methylene blue. Note that the nuclei still stain. (6) Place a small drop of a 3 per cent, solution of saponin in isotonic salt solution on a slide, and near it a small drop of blood. Observe as in (a). Repeat with a 2 per cent, solution of sodium taurocholate in salt solution. If necturus corpuscles, which are splendid objects for such experiments on account of their great size, have been used, intracorpuscular crystallization of the haemoglobin may be observed. (c) Repeat (a) and (b) with mammalian blood. Note that the cor- puscles swell before being laked by the saponin . If any of the corpuscles are crenated, it may be seen that before being laked by the saponin the crenations disappear, the corpuscles becoming round, while in the taurocholate solution they may remain crenated till laking has occurred. This indicates that the permeability of the envelopes is not affected in the same way by the two laking agents. 15. Haemolysis and Agglutination by Foreign Serum. — (i) To a small quantity of rabbit's blood add an equal volume of dog's serum. Mix and let stand at 40° C. The colour of the blood is soon darker than before, and it can be seen to be laked. Examine microscopically. (2) Place a small drop of rabbit's blood and a somewhat larger drop of the dog's serum on a slide, near, but not quite in contact with, each other. Now put on a cover-slip, so that the drops just come together, and examine at once with the microscope with a moderately high power. Where the two drops mingle, the red corpuscles will be seen first to become agglutinated into groups, and then to fade out, leaving only their ' ghosts.' A few of the corpuscles which come into contact with the, as yet, undiluted serum may be entirely dissolved. (3) Heat some of the dog's serum to 60° C. for ten minutes, and repeat (i) and (2). No laking will now be produced in the rabbit's corpuscles, but agglutination may be observed as before. (4) Repeat (i) and (2) with dog's bloo^ and rabbit's serum. The blood will not be laked. although sometimes the dog's corpuscles may become crenated. There will be no agglutination. 72 THE CIRCULATING LIQUIDS OF THE BODY (5) With a 5 per cent, suspension of rabbit's washed corpuscles perform the following experiments:* Put into each of six small test-tubes i c.c. of the suspension. Label the tubes A, A', B, B', C, C'. (a) To A and A' add respectively o-i c.c. and 0-5 c.c. ox serum. (b) To B and B' add respectively o-i c.c. and 0-5 c.c. dog's serum. (c) To C and C' add respectively o-i c.c. and 0-5 c.c. of 0-9 per cent, sodium chloride solution. Put all the tubes in a bath at 40° C. Compare the amount of laking and agglutination in the various tubes at intervals of two minutes or less. Repeat (a), (b), and (c) with guinea-pig's washed corpuscles and serum of ox and dog. Determine which of these sera has the strongest haemolytic power.! (6) Heat i c.c. of ox and dog's serum respectively to 56° C., keeping it at that temperature, or not more than a couple of degrees above it, for tenj minutes, and repeat experiment (5), labelling the tubes D, D', E, E', F, F'. Save the rest of the heated sera for (8). There is no laking in any of the tubes, but probably agglutination in D, D', and E, E'. (The complement is destroyed, but not the intermediary body or amboceptor, or the agglutinin — p. 28.) (7) Put half of the contents of tubes D, D', E, E', into four separate test-tubes, and add to each 0-2 c.c. of rabbit's serum. If there is laking now it is because the rabbit's serum contains complement. Save the balance of D, D', E and E' for (8). (8) Allow 0-5 c.c. of ox serum to act at o° C. on the rabbit's washed corpuscles contained in 5 c.c. of the 5 per cent, suspension after removal of the sodium chloride solution. The ox serum and rabbit's corpuscles are separately cooled to o° C. before being mixed, and the mixture is then kept at o° C. for one hour. Centrifugalize the serum off rapidly. Label it ' Serum S.' To 0-2 c.c. of the original 5 per cent, suspension of rabbit's washed corpuscles add o;i c.c. of this serum (labelling the tube G), and put at 40° C. with a control-tube containing the same amount of suspension plus salt solution instead of serum. Add the rest of the serum S, cooled to o° C., to the same cooled rabbit's cor- puscles, and leave for a further period at o° C. Then Centrifugalize rapidly, and to 0-2 c.c. of the original suspension of washed rabbit's corpuscles add o-i c.c. of serum S (labelling the tube H), and put at 40° C. with a sodium chloride tube as control. There may be no * The material obtained from one medium-sized dog, two rabbits, and one guinea-pig is enough for fifty or sixty students, working together in sets of two, to perform experiments (5) to (8). In order to obtain a serum more strongly haemolytic for rabbit's corpuscles than normal dog's serum, a dog may be ' immunized ' by previous injection of all the washed corpuscles obtainable from a rabbit. The injection should be made under the skin or, better, into the peritoneal cavity — of course, with aseptic precautions. It should be repeated not less than twice, with an interval of ten days between the successive injections, and the dog's blood should be drawn off about ten days after the last injection. f To determine the amount of laking at any given moment, drop the small test-tubes into the metallic centrifuge cups after shaking them up, and centrif- ugalize. A very short time is sufficient to separate a clear supernatant liquid, from the tint of which the extent of the haemolysis can be deduced. Before replacing the tubes in the thermostat, they should, of course, be shaken up. Small test-tubes of about 8 mm. internal diameter and short enough to go conveniently into the centrifuge cups are the most serviceable. J For exact work a longer time is recommended. But for the student the time is made as short as possible, and it is only in exceptional cases that ten minutes is not enough. PRACTICAL EXERCISES 73 laking in either G or H, or if there is laking it may be greater in G than in H. The amboceptor has been removed from serum S by the rabbit's corpuscles. Add o-i c.c. of this ' inactivated ' serum to the balance of D, D', and E, E' (left from 6). Laking will occur because the serum S contains complement, and the heated serum added in (6) to these tubes contains amboceptor. Wash the rabbit's corpuscles which have been treated with ox serum at o° C. with cooled sodium chloride solution. Add to them some of serum S (that from the top of tube H will do if no more is left), and put at 40° C. Laking will occur, showing that the amboceptor was fixed by the rabbit's corpuscles at o° C. To a further portion of the washed rabbit's corpuscles which were treated with ox serum at o° C. add normal rabbit's serum, and put at 40° C. If laking occurs it is because the rabbit's serum contains complement. Dog's serum may be used instead of ox serum for experiment (8). 1 6. Osmotic Resistance of the Coloured Corpuscles. — Fill a burette with a I per cent, solution of sodium chloride and another with dis- tilled water. Take a series of ten test-tubes and run into the first 6 c.c. of the NaCl solution, into the second 5-8 c.c., into the third 5-6 c.c., and so on, always making a difference of 0-2 c.c. between successive test-tubes. From the other burette run in enough distilled water to make up 10 c.c. of solution in each tube — that is, 4 c.c. of dis- tilled water for the first tube, 4-2 c.c. for the second, and so on. Shake up. The tubes now contain a series of solutions of salt differing in strength by 0-02 per cent, in successive tubes, the strongest being 0-6 per cent., and the weakest 0-42 per cent. Number the tubes i to 10, beginning with the strongest solution. Put into each tube one drop of perfectly fresh blood. Shake moderately so as to mix the blood and salt solution, and allow the tubes to stand for ten to thirty minutes. Observe the colour of the clear liquid above the sediment of corpuscles. Determine in which tube the first tinge of haemoglobin appears. The next higher concentration of the salt solution is that in which all the corpuscles are just able to retain their haemoglobin, and is a measure of the minimum osmotic resistance of the corpuscles, or the resistance of the weakest corpuscles. Repeat with blood which has stood at room temperature for twelve to twenty -four hours. For clinical purposes tubes, each containing 5 c.c. of salt solution, may be used. A single drop of blood can then be distributed between the tubes with a fine pipette or a glass rod, beginning with the most concentrated solution, and passing down to the less concentrated. The blood must be dis- tributed rapidly before coagulation occurs. Only such concentrations of the salt solution as are known to correspond to the possible variations of the osmotic resistance for any particular disease or for any particular variety of blood need be employed. 17. Blood-Pigment — (i) Preparation of Haemoglobin Crystals. — (a) To a little dog's blood in a narrow test-tube add its own volume or twice its volume of chloroform. Invert the tube ten or twelve times so as to allow the chloroform to act on the blood, but avoid violent shaking. When the tube is now allowed to stand for a few minutes the laked blood all rises to the top. Remove a little of the layer of blood without taking with it any of the chloroform layer, and examine the oxyhaemoglobin crystals with the microscope. They form long rhombic prisms and needles (Fig. 14, p. 52). (b) Add a little crude saponin to dog's blood in a test-tube. Shake up well, and allow it to stand till the colour becomes dark. Then shake vigorously, and a mass of haemoglobin crystals will be formed. (c) Put a small drop of guinea-pig's blood on a slide. Mix with a 74 drop of Canada balsam and cover. Tetrahedral crystals of oxy- hsemoglobin will form after a time. The slide may be kept. (2) Spectroscopic Examination of Haemoglobin and its Derivatives. — (a) With a small, direct-vision spectroscope look at a bright part of the sky or a white cloud. Focus by pulling out or pushing in the eye- piece until the numerous fine dark lines (Fraunhofer's lines), running vertically across the spectrum, are seen. Narrow the slit by moving the milled edge till the lines are as sharp as they can be made. Note especially the line D in the orange, the lines E and 6 in the green, and F in the blue. Always hold the spectroscope so that the red is at the left of the field. Now dip an iron or platinum wire with a loop on the end of it into water, and then into some common salt or sodium carbonate, and fasten or hold it in the flame of a fishtail burner. On examining the flame with the spectroscope, a bright yellow line will be seen occupying the position of the dark line D in the solar spectrum. This is a convenient line of reference in the spectrum, and in studying the spectra of haemoglobin and its derivatives, the position of the absorption bands with regard to the D line should always be noted. The dark lines in the solar spectrum are due to the absorption of light of a definite range of wave-lengths by metals in a state of vapour in the sun's atmosphere, and of course no dark lines are seen in the spectrum of a gas-flame. Put some defibrinated blood into a test-tube. Fig. 20. — Direct Vision Spectroscope ot Simple Type. A , slot in which a pin on the eyepiece C slides in focussing the spectrum; B, milled head, by the rotation of which the slit is narrowed or widened. Fasten it vertically in a clamp in front of the flame and examine it with the spectroscope, holding the latter in one hand with the slit close to the test-tube, and focussing the eyepiece with the other. Or arrange the spectroscope, test-tube and gas-flame on a stand as in Fig. 21. Nothing can be seen till the blood is diluted. Pour a little of the blood into another test-tube, and go on diluting till, on focussing, two bands of oxyhesmoglobin are seen in the position indicated in Fig. 13, p. 51 . Draw the spectrum; then dilute still more, and observe which of the bands first disappears. Now put 5 c.c. of the blood into another test-tube, and dilute it with four times its volume of water. Take 5 c.c. of this dilution, and again add four times as much water, and so on till the solution is only faintly coloured. Note with what degree of dilution the bands disappear. Then examine each of the solutions with the spectroscope and draw its spectrum. (b) Make a solution of blood which shows the oxyhaemoglobin bands sharply. Add some ammonium sulphide solution to reduce the oxy- haemoglobin. Heat gently to about body temperature. A single, ill-defined band now appears, occupying a position midway between the oxyhEemoglobin bands, and the latter disappear. This is the band of reduced hemoglobin (Fig. 13). (c) Carbonic Oxide Hemoglobin. — Pass coal-gas through blood for PRACTICAL EXERCISES 75 a considerable time. Examine some of the blood (after dilution) with the spectroscope. Two bands, almost in the position of the oxyhaemoglobin bands, are seen; but no change is caused by the addition of ammonium sulphide, since carbonic oxide haemoglobin is a more stable compound than oxyhaemoglobin. (d) Methcemoglobin. — Put some blood into a test-tube, add a few drops of a solution of ferricyanide of potassium, and heat gently. On diluting a well-marked band will be seen in the red. On addition of ammonium sulphide this band disappears; the oxyhaemoglobin bands are seen for a moment, and then give place to the band of reduced haemoglobin (Fig. 13, p. 51). (e) Acid Hcematin. — To a little diluted blood add strong acetic acid and heat gently. The colour becomes brownish. The spectrum shows a band in the red between C and D, not far from the position of the band of methaemoglobin. The addition of a drop or two of ammonium sulphide causes no change in the spectrum, and this is a means of distinguishing acid haematin from methaemoglobin. If more ammonium sulphide be added, haematin will be precipitated when the acid solution has been rendered neutral, and a further addition of ammonium sulphide or sodium hydroxide will cause the haematin to be again dis- solved, a solution of alkaline haematin being formed. This A in its turn may be reduced by an excess of ammonium sul- phide, and the spectrum of Jest' tube haemochromogen may be ob- tained (Fig. 13, p. 51). Since the watery solution of acid haematin obtained as above is usually somewhat tur- bid, a solution in acid ether is sometimes employed for spec- troscopic examination. Add to a little undiluted defibrinated blood about half its volume of glacial acetic acid, and then not less than an equal volume of Spectroscope Solution. 0 Fig. 21. — Spectroscopic Examination of Blood-Pigment. ether. Mix well, pour off the ethereal extract and examine it with the spectroscope, diluting, if necessary, with ether and glacial acetic acid. It shows a strong band in the red somewhat farther from the D line than the methasmoglobin band. On dilution, three additional fainter bands may be seen. (/) Alkaline Hcematin. — To diluted blood add strong acetic acid and warm gently for a few minutes. Then, when the spectroscopic ex- amination of a sample shows that acid haematin has been formed, neutralize with sodium hydroxide. A brownish precipitate of haematin is thrown down, which dissolves in an excess of sodium hydroxide, giving a solution of alkaline haematin (or alkali haematin). Or add sodium hydroxide to blood directly, and warm for a couple of minutes after the colour has changed decidedly to brownish-black. The spectrum of alkaline haematin is a broad but ill-defined band just overlapping the D line, and situated chiefly to the red side of it (Fig. 13). The solution should be shaken up with air before being examined, as 76 THE CIRCULATING LIQUIDS OF THE BODY some of the alkali haematin is changed into haemochroraogen by re- ducing substances formed by the action of the alkali on the blood. (g) Hcemochromogen. — To a solution of alkaline haematin add a drop or two of ammonium sulphide. The band near D disappears, and two bands make their appearance in the green (Fig. 13, p. 51). (A) Hcsmatoporphyrin. — Put some strong sulphuric acid into a test- tube. Add a few drops of blood, agitate the test-tube till the blood dissolves, and examine the purple liquid, diluting it, if necessary, with sulphuric acid. Its spectrum shows two well-marked bands, one just to the left of D, and the other midway between D and E (Fig. 13). (3) Guaiacum Test for Blood.-^A test for blood — much used in hospitals, and, indeed, a delicate one, but quite untrustworthy unless certain precautions be taken — is the guaiacum test. A drop of freshly- prepared tincture of guaiacum is added to the liquid to be tested, and then peroxide of hydrogen. If blood be present, the guaiacum strikes a blue or greenish-blue colour. The decomposition of the peroxide by the blood is due mainly to the haemoglobin of the corpuscles. Any derivative of haemoglobin which still con tarns the iron will act, and boiling does not abolish this power. On the other hand, oxydases or oxidizing ferments present, not only in the formed elements of blood, but elsewhere, e.g., in fresh vegetable protoplasm, milk, seminal fluid, and pus, will cause the same colour (p. 271), but not if they have been previously boiled.* The test has been considered chiefly of value as a negative test. When the blue colour is not obtained, we have good evidence that blood is absent. But, according to Buckmaster, if the precaution of first boiling the liquid suspected to contain blood be adopted, it is also a good positive test. It is, however, far inferior to the haemin test (p. 78) where that can be obtained, and of course in- ferior to the identification of erythrocytes with the microscope, or to the spectroscopic identification of the blood-pigment where the material is suitable for this. (4) Quantitative Estimation of Haemoglobin — (a) By Haldane's Modi- fication of Gowers' Hcemoglobinometer. — Place in the graduated tube B (Fig. 22.) an amount of water less than will ultimately be required to dilute the blood to the required tint. Puncture the finger or lobe of the ear with one of the small lancets in F, and fill the capillary pipette D to a little beyond the mark 20. Wipe the point of the pipette and dab it on a piece of filter-paper till the blood stands exactly at the mark. Blow the blood into the water in B, and rinse the pipette with the water. Attach the cap of tube G to a gas-burner. Introduce the rubber tube into B nearly to the level of the water, and allow gas to pass for a few seconds. Withdraw the tube while the gas is still passing. Immediately close the end of B with the finger, and move the tube so that the liquid passes from end to end of it at least a dozen times, to saturate the haemoglobin with carbonic oxide. While this is being done, the tube should be held in a cloth, otherwise it will become heated, and liquid will spurt out when the finger is removed. Water is now added * The formed elements of blood really contain no less than three ferments of interest in this connection: (i) A catalase which decomposes peroxide of hydrogen into water and molecular oxygen (i.e., oxygen not in the atomic or nascent state). This reaction is given by both blood and pus. (2) An oxydase (also spelled oxidase), which oxidizes guaiacum and similar substances without the presence of hydrogen peroxide. This reaction is obtainable even from aqueous extracts of leucocytes. (3) A peroxydase (also spelled peroxi- dase) which causes the oxidation of these substances only in the presence of hydrogen peroxide, a reaction also given by leucocytes. These ferments are all inactivated iby boiling (Kastle). PRACTICAL EXERCISES drop by drop with the pipette stopper of the bottle E, which is used for holding the water, the tube being inverted after each, addition, till the tint in B is the same as that in A. In comparing the tubes, they should be held against the light from the sky or from an opal glass lamp-shade. It is necessary to transpose the tubes repeatedly. The level at which the tints are equal is read off on B half a minute after the addition of the last drop of water. Water is now again added by drops till the tint in B is just roticeably weaker than in A, and the mean of the two readings is taken. The result is the percentage actually present of the average proportion of haemoglobin in the blood of healthy adult males. Healthy women give an average of only 89 per cent., and healthy children an average of only 87 per cent., of the proportion in men. The liquid in A is a i per cent, solution of blood containing the average percentage of haemoglobin found in the blood of healthy Fig. 22. — Haldane's Modification of Gowers' Haemoglobinometer. adult males, and having an oxygen capacity of 18 5 per cent. — i.e., 100 c.c. of the blood with which the standard was made would take up in combination from air 18-5 c.c. of oxygen. The solution in A has been saturated with carbonic oxide. This method is probably more accurate than any other used in clinical work, the error, in the hands of an experienced observer, not exceeding i per cent. (b) By Fleischl's Hcemometer (Fig. 23). — Fill with distilled water that compartment a' of the small cylinder (above the stage) which is over the tinted wedge. Put a little distilled water into the other compart- ment a. Now prick the finger and fill one of the small capillary tubes with blood. See that none of the blood is smeared on the outside of the tube. Then wash all the blood into the water in compartment a, and fill it to the brim with distilled water. By means of the milled 78 THE CIRCULATING LIQUIDS OF THE BODY head T move the tinted wedge K till the depth of colour is the same in the two compartments. The percentage of the normal quantity of haemoglobin is given by the graduated scale P. For example, if the reading is 90, the blood contains 90 per cent, of the normal amount; if 100, it contains the normal quantity. The observations should be made in a dark room, the white surface S, arranged below the compart- ments a and a', being illuminated by a lamp. Or the instrument may be placed in a small box, lighted by a candle. It is best that each result should be the mean of two readings, one just too large and the other just too small. In any case the instrument does not give readings accurate to less than 5 per cent. (c) Hoppe-Seyler's Method. — Two parallel-sided glass troughs are used. In one is put a standard solution of oxy haemoglobin of known strength, in the other a measured quantity of the blood to be tested. The latter is diluted K with water until its tint appears the same as that of the standard solution, when the troughs are placed Fig. 23. — Fleischl's Hasmometer. Fig. 24.— Crystals of Hamin (Frey). side by side on white paper. From the quantity of water added it is easy to calculate the proportion of haemo- globin in the undiluted blood. Greater accuracy is obtained if the haemoglobin in the standard solution and that of the blood are converted into carbonic oxide haemo- globin by passing a stream of coal-gas through them. (d) Tallquist's Method. — In this method the tint produced by a drop of blood on a piece of white filter-paper is compared with a scale representing 10 percentages of haemoglobin (from 10 to 100 per cent.). The standard filter-paper is supplied in the form of a book with the scale. To make an estimation, all that is necessary is to touch a drop of blood with a piece of the filter-paper, and allow the blood to diffuse slowly through the paper, so as to give an even stain. The position of the stain is then determined by the scale; e.g., it may be deeper than 90, but fainter than 100, in which case the percentage of haemo- globin lies between 90 and 100. The method is by no means a very accurateone, but more accurate than it appears at first sight. (5) Microscopic Test for Blood-Pigment. — Put a drop of blood on a slide. Allow the blood to dry, or heat it gently over a flame, so as to evaporate the water. Add a drop of glacial acetic acid ; put on a cover- PRACTICAL EXERCISES . 79 glass, and again heat slowly till the liquid just begins to boil. Take the slide away from the flame for a few seconds, then heat it again for a moment; and repeat this process two or three times. Now let the slide cool, and examine with the microscope (high power). The small black, or brownish-black, crystals of haemin will be seen (Fig. 24, p. 78). This is an important test where only a minute trace of blood is to be examined, as in some medico-legal cases. If a blood-stain is old, a mimite crystal of sodium chloride should be added along with the glacial acetic acid. Fresh blood contains enough sodium chloride. A blood-stain on a piece of cloth may first of all be soaked in a small quantity of distilled water, and the liquid examined with the spectro- scope or the micro-spectroscope (a microscope in which a small spectro- scope is substituted for the eyepiece). Then evaporate the liquid to dryness on a water-bath, and apply the hasmin test. Or perform the haemin test directly on the piece of cloth. In a fresh stain the blood- corpuscles might be recognized under the microscope. Very few liquids, however, are available for washing out the blood, as all ordinary solutions, and even serum itself, cause laking of dried corpuscles (Guthrie). Absolute alcohol, or 35 per cent, potassium hydroxide, may be used to soak and rub up the cloth in. CHAPTER III THE CIRCULATION OF THE BLOOD AND LYMPH THE blood can only fulfil its functions by continual movement. This movement implies a constant transformation of energy; and in the animal body the transformation of energy into mechanical work is almost entirely allotted to a special form of tissue, muscle. In most animals there exist one or more rhythmically contractile muscular organs, or hearts, upon which the chief share of the work of keeping up the circulation falls. SECTION I. — PRELIMINARY ANATOMICAL AND PHYSICAL DATA. Comparative. — In Echinus a contractile tube connects the two vascu- lar rings that surround the beginning and end of the alimentary canal, and plays the part of a heart. In the lower Crustacea and in insects the heart is simply the contractile and generally sacculated dorsal bloodvessel; in the higher Crustacea, such as the lobster, it is a well- defined muscular sac situated dorsally. A closed vascular system is the exception among invertebrates. In most of them the blood passes from the arteries into irregular spaces or lacunae in the tissues, and thence finds its way back to the heart. In the primitive vertebrate heart five parts can be distinguished as we proceed from the venous to the arterial end: (i) The sinus venosus, into which the great veins open; (2) the auricular canal, from the dorsal wall of which is developed — (3) the auricle; (4) the ventricle; (5) the bulbus arteriosus, from which the chief artery starts (Fig. 25, p. 81). Amphioxus, the lowest verte- brate, has a primitive lacunar vascular system; a contractile dorsal bloodvessel serves as arterial or systemic heart, a contractile ventral vessel as venous or respiratory heart. From the latter, vessels go to the gills. Fishes possess only a respiratory heart, consisting of a venous sinus, auricle, ventricle, and bulbus arteriosus. This drives the blood to the gills, from which it is gathered into the aorta; it has thence to find its way without further propulsion through the systemic vessels Amphibians have a venous sinus, two auricles, a single rentricle, and an arterial bulb; reptiles, two auricles and two incompletely-separated ventricles. In birds and mammals the respiratory and systemic hearts are completely separated. The former, consisting of the right auricle and ventricle, propels the blood through the lungs; the latter, consisting of the left auricle and ventricle, receives it from the pul- monary veins, and sends it through the systemic vessels. The sinus venosus seems to be represented in the mammalian heart by certain small portions of tissue, especially the so-called sino-auricular node, a little knot of primitive fibres near the mouth of the superior 80 ANATOMICAL AND PHYSICAL DATA 81 vena cava. The auricular canal is probably represented by the auriculo-ventricular bundle (conveniently designated as the a. -v. bundle), which will again be referred to in relation to the conduction of the heart- beat from auricles to ventricles (p. 147). This bundle starts from a clump of primitive tissue, the auriculo-ventricular node (a.-v. node) at the base of the interauricular septum on the right side, below and to the right of the coronary sinus, and runs down the interventricular septum. The sino-auricular and the auriculo-ventricular nodes are connected by fibres which run in the interauricular septum, so that it may be considered that the primitive cardiac tube is still represented from base to apex of the adult mammalian heart, although only by very slender threads of tissue, amidst the massive secondary developments of auricular and ventricular muscle (Keith and Flack). General View of the Circulation in Man. — The whole circuit of the blood is divided into two portions, very distinct from each other, both anatomically and function- ally— the respiratory or lesser circulation, and the systemic or greater circulation. Starting from the left ventricle, the blood passes along the systemic vessels — ar- teries, capillaries, veins — and, on returning to the heart, is poured into the right auricle, and thence into the right ventricle. From the latter it is driven through the pulmonary artery to the lungs, passes through the capillaries of these organs, and returns through the pulmonary veins to the left auricle and ventricle. The portal Fifl ^.-Diagram of Primitive Vertebrate _.- .- •*•_ 1-1 OOT-T <^*-im Kin in cr H AQT111"AQ Trmniri in T l-nn system, which gathers up the blood from the intestines, forms a kind of loop on the systemic circulation. The lymph-current is also in a sense a slow and stag- nant side -stream of the blood - circulation; for substances are constantly passing from the bloodvessels into the lymph -spaces, and returning, although after a com- paratively long interval, into the blood by the great lymphatic trunks. Physiological Anatomy of the Vascular System. — The heart is to be looked upon as a portion of a bloodvessel which has been modified to act as a pump for driving the blood in a definite direction. Morpho- logically it is a bloodvessel; and the physiological property of auto- matic rhythmical contraction which belongs to the heart in so eminent a degree is, as has been mentioned (p. 80), an endowment of blood- vessels in many animals that possess no localized heart. Even hi some mammals contractile bloodvessels o'ccur; the veins of the bat's wing, for example, beat with a regular rhythm, and perform the func- tion of accessory hearts. Heart, combining Features found in the Eel, Dogfish, and Frog (Flack, after Keith), a, Sinus venosus; b, auricular canal; c, auricle; d, ventricle; e, bulbus cordis; /, aorta; i-i, sino-auricular junc- tion and venous valves; 2-2, junction of canal and auricle; 3-3, annular part of auricle; 5, bulbo-ventricular junction. 82 THE CIRCULATION OF THE BLOOD AND LYMPH The whole vascular system is lined with a single layer of endothelial cells. In the capillaries nothing else is present; the endothelial layer forms the whole thickness of the wall. In young animals, at any rate, the endothelial cells of the capillaries are capable of contracting when stimulated; and changes in the calibre of these vessels can be brought about in this way. The walls of the arteries and veins are chiefly made up of two kinds of tissue, which render them distensible and elastic: non-striped muscular fibres and yellow elastic fibres. The muscular fibres are mainly arranged as a circular middle coat, which, especially in the smaller arteries, is relatively thick. One conspicuous layer of elastic fibres marks the boundary between the middle and inner coats. In the larger arteries elastic laminae are also scattered freely among the muscular fibres of the middle coat. The outer coat is composed chiefly of ordinary connective tissue. The veins differ from the arteries in having thinner walls, with the layers less distinctly marked, and containing a smaller proportion of non-striped muscle and elastic tissue ; although in some veins, those of the pregnant uterus, for instance, and the cardiac ends of the large thoracic veins, there is a greater development of muscular tissue. Further, and this is of prime physiological importance, valves are present in many veins. These are semilunar folds of the internal coat projecting into the lumen in such a direction as to favour the flow of blood towards the heart, but to check its return. In some veins, as the venae cavae, the pulmonary veins, the veins of most internal organs, and of bone, there are no valves ; in the portal system they are rudimentary in man and the great majority of mammals. The valves are especially well marked in the lower limbs, where the venous circulation is uphill. When a valve ceases to perform its function of supporting the column of blood between it and the valve next above, the foundation of varicose veins is laid; the valve immediately below the incompetent one, having to bear up too great a weight of blood, tends to yield in its turn, and so the condition spreads. The smallest veins, or venules, are very like the smallest arteries, or arterioles, but somewhat wider and less muscular. The transition from the capillaries to the arterioles and venules is not abrupt, but may be considered as marked by the appearance of the non-striped muscular fibres, at first scattered singly, but gradually becoming closer and more numerous as we pass away from the capillaries, until at length they form a complete layer. In the heart the muscular element is greatly developed and differ- entiated. Both histologically and physiologically the fibres seem to stand between the striated skeletal muscle and the smooth muscle. In the mammal the cardiac muscular fibres are generally described as made up of short oblong cells, devoid of a sarcolemma, often branched, and arranged in anastomosing rows, each cell having a single nucleus in the middle of it. But it has recently been shown that the muscle fibrils run right through the apparent cell boundaries, and form a con- tinuous sheet of tissue anastomosing in every direction. The fibres are transversely striated, but the striae are not so distinct as in skeletal muscle. A sarcolemma is not absent, although it is more delicate than in skeletal muscle, and perhaps of a different nature. Many fibres pass from one auricle to the other, and from one ventricle to the other. In the frog's heart the muscular fibres are spindle-shaped, like those of smooth muscle, but transversely striated, like those of skeletal muscle. From the sinus to the apex of the ventricle there is a con- tinuous sheet of muscular tissue. ANATOMICAL AND PHYSICAL DATA 83 The problems of the circulation are partly physical, partly vital. Some of the phenomena observed in the blood-stream of a living animal can be reproduced on an artificial model ; and they may justly be called the physical or mechanical phenomena of the circulation. Others are essentially bound up with the properties of living tissues ; and these may be classified as the vital or physiological phenomena of the circulation. The distinction, although by no means sharp and absolute, is a convenient one — at least, for purposes of description; and as such we shall use it. But it must not be forgotten that the physiological factors play into the sphere of the physical, and the physical factors modify the physiological. Considered in its physical relations, the circulation of the blood is the flow of a liquid along a system of elastic tubes, the bloodvessels, under the influence of an intermittent pressure produced by the action of a central pump, the heart. But the branch of dynamics which treats of the movement of liquids, or hydrodynamics, is one of the most difficult parts of physics, and even in the physical portion of our subject we are forced to rely chiefly on empirical methods. It would, therefore, not be profitable to enter here into mathematical theory, but it may be well to recall to the mind of the reader one or two of the simplest data connected with the flow of liquids through tubes: Torricelli's Theorem. — Suppose a vessel filled with water, the level of which is kept constant; the velocity with which the water will escape from a hole in thejside of the vessel at a vertical depth h below the surface will be v= * TaioiF scl- mm- The outflow will be ^ gsffo x i = 05^0 cub. mm. per sec. The outflow from the capillary of ,».o /* diameter would be sixteen times as much, from the 5 /* capillary only one-sixteenth as much. Some idea of the extremely minute scale on which the blood-flow through a single capillary takes place may be obtained if we consider that for the capillary of 10 /* diameter a flow of -2500TJ cub. mm. per sec. would scarcely amount to I cub. mm in six hours, or to i c.c. in 250 days. When the initial energy is obtained in any other way than by means of a ' head ' of water in a reservoir — say, by the descent of a piston which keeps up a constant pressure in a cylinder filled with liquid — the results are exactly the same. Even when the horizontal tube is distensible and elastic, there is no difference when once the tube has taken up its position of equilibrium for any given pressure, and that pressure does not vary. Flow with Intermittent Pressure. — When this acts on a rigid tube, everything is the same as before. When the pressure alters, the flow at once comes to correspond with the new pressure. Water thrown by a force-pump into a system of rigid tubes escapes at every stroke of the pump in exactly the quantity in which it enters, for water is practically incompressible, and the total quantity present at one time in the system cannot be sensibly altered. In the intervals between the strokes the flow ceases; in other words, it is intermittent. It is very different with a system of distensible and elastic tubes. During each stroke the tubes expand, and make room for a portion of the extra liquid thrown into them, so that a smaller quantity flows out than passes in. In the intervals between the strokes the distended tubes, in virtue of their elasticity, tend to regain their original calibre. Pressure is thus exerted upon the liquid, and it continues to be forced out, so that when the strokes of the pump succeed each other with sufficient rapidity, the outflow becomes continuous. This is the state of affairs in the vascular system. The intermittent action of the heart is toned down in the elastic vessels to a continuous steady flow. SECTION II. — THE BEAT OF THE HEART IN ITS PHYSICAL OR MECHANICAL RELATIONS. Events in the Cardiac Cycle. — In the frog's heart the contraction can be seen to begin about the mouths of the great veins which open into the sinus venosus. Thence it spreads in succession over the 86 THE CIRCULATION OF THE BLOOD AND LYMPH sinus and auricles, hesitates for a moment at the auriculo- ventric- ular junction, and then with a certain suddenness invades the ventricle. In the mammalian heart the contraction likewise com- mences, so far as can be ascertained by inspection or the study of tracings, in the region near the mouths of the veins opening into the auricles. It will be seen, when the question of the origin of the rhythmical beat is being discussed (p. 141), that the actual starting- point is probably the sinus tissue of the right auricle (p. 142) near the opening of the superior vena cava, which is richly provided with muscular fibres akin to those of the heart. But the wave advances so rapidly that it is difficult to trace in its course a regular progress from base to apex, although the ventricular beat undoubtedly follows that of the auricle, and in a heart beating normally the electrical change associated with contraction of the ventricle begins at the base, then reaches the apex (p. 853), and finally passes towards the orifices of the great arteries. The most conspicuous events in the beat of the heart, in their normal sequence, are: (i) the auricular contraction or systole, (2) the ventricular contraction or systole, each followed by relaxation. (3) the pause. The auricles, into which, "and beyond which into the ven- tricles, blood has been flowing during the pause from the great thoracic veins, contract sharply, the right, perhaps, a little before the left. The contraction begins in the muscular tissue that surrounds the orifices of the veins, so that these, destitute of valves as they are, are functionally, at least, if not anatomically, sealed up for an instant, and regurgitation of blood into them is to a great extent, if not entirely, prevented. Apparently, complete closure of the inferior cava is unnecessary, the pressure of the blood in it being sufficiently high to hinder any important back-flow. The action of the circular fibres of the veins in closing their orifices is reinforced by the contraction of a band of muscle (the tania ter- minalis) in the roof of the right auricle. This band compresses especially the mouth of the superior vena cava. The filling of the ventricles is thus completed. The actual amount of extra blood injected into the ventricles by the auricular contraction is not large. The ventricles are already nearly charged, but the auricles, so to speak, ram the charge home. The ventricular contraction follows hard on the relaxation of the auricles. The mitral and tricuspid valves, whose strong but delicate curtains have during the diastole been hanging down into the ventricles and swinging freely in the entering current of blood, are floated up as the intraventricular pressure begins to rise, so that, in the first moment of the sudden and powerful ventricular systole, the free edges of their segments come together, and the auriculo-ventricular orifices are completely closed (Fig. 98, p. 206). In the measure in which the pressure in the contracting ventricles increases, the contact of the valvular seg- MECHANICS OF THE HEART-BEAT 87 merits becomes closer and more extensive; and their tendency to belly into the auricles is opposed by the pull of the chordae tendineae, whose slender cords, inserted into the valves from border to base, are kept taut, in spite of the shortening of the ventricles by the con- traction of the papillary muscles. The arrangement and connec- tions of the muscular fibres of the heart are such that during the auricular systole the auriculo-ventricular groove moves towards the base of the heart, while during the systole of the ventricles it moves towards the apex, which constitutes a relatively fixed point on account of the mutual action of the numerous fibres which converge here and constitute the ' whorl.' The line joining the apex and the origin of the aorta does not shorten when the ventricles contract, but all parts of the heart are drawn towards this line. The apex is, therefore, pushed forwards, while the rest of the ventricular surface is being drawn inwards. During the systole, the ventricles change their shape in such a way that their combined cross-section — which in the relaxed state is a rough ellipse with the major axis from right to left — becomes approximately circular, and they then form a right circular cone. As soon as the pressure of the blood within the con- tracting ventricles exceeds that in the aorta and pulmonary artery respectively, the semilunar valves, which at the beginning of the ventricular systole are closed, yield to the pressure, and blood is driven from the ventricles into these arteries. The ventricles are more or less completely emptied during the contraction, which seems still to be maintained for a short time after the blood has ceased to pass out. The contraction is followed by sudden relaxation. The intraventricular pressure falls. The lunules of the semilunar valves slap together under the weight of the blood as it attempts to regurgitate, the corpora Arantii seal up the central chink, and the aorta and pulmonary artery are thus cut off from the heart. Then follows an jnterval during which the whole heart is at rest, namely, the interval between the end of the relaxa- tion of the ventricles and the beginning of the systole of the auricles. This constitutes the pause. The whole series of events is called a cardiac cycle or revolution (see Practical Exercises, p. 201). It will be easily understood that the time occupied by any one of the events of the cardiac cycle is not constant, for the rate of the heart is variable. If we take about 70 beats a minute as the average normal rate in a man, the ventricular systole will occupy about 0-3 second; the diastole,* including the ventricular relaxation, about * The term ' diastole ' is variously used, as meaning the pause, the pause plus the period during which relaxatidn is occurring, or the period of re- laxation alone. We shall employ it in the second sense. Henderson refers to the period during which the ventricular muscle is at rest, from the end of its relaxation to the onset of the auricular systole, as the ' diastasis ' and the period during which the relaxation is taking place as the 'diastole,' a termin- ology which seems worthy of general adoption. 88 THE CIRCULATION OF THE BLOOD AND LYMPH 0-5 second. The systole of the auricle is one-third as long as that of the ventricle. This rhythmical beat of the heart is the ground phenomenon of the circulation. It reveals itself by certain tokens — sounds, surface- movements or pulsations, alterations of the pressure and velocity of the blood, changes of volume in parts — all periodic phenomena, continually recurring with the same period as the heart -beat, and all fundamentally connected together. And if we hold fast the idea that when we take a pulse-tracing, or a blood-pressure curve, or a plethysmographic record, we are really investigating the same fact from different sides, we shall be able, by following the cardiac rhythm and its consequences as far as we can trace them, to hang upon a single thread many of the most important of the physical phenom- ena of the circulation. The Sounds of the Heart. — When the ear is applied to the chest, or to a stethoscope placed over the cardiac region, two sounds are heard with every beat of the heart. They follow each other closely, and are succeeded by a period of silence. The dull booming ' first sound ' is heard loudest in a region which we shall afterwards have to speak of as that of the ' cardiac impulse ' (p. 90) ; the short, sharp, ' second sound ' over the junction of the second right costal cartilage with the sternum. The heart-sounds can be registered by placing over the chest a microphone receiver connected with a string galvanometer. The magnified sounds are translated into electrical changes which cause movements of the fibre of the galvanometer, and the movements are photographed on a travelling plate (Einthpven). The record is called a cardiophonogram. When this is studied, a third sound can be detected, and it is probable that it is present in all persons, although it is as a rule inaudible to auscultation. It occurs early in diastole very shortly after the second sound. In those persons in whom it is audible it is most distinct over the region of cardiac impulse. It is described as softer and of lower pitch than the second sound (Thayer). There has been much discussion as to the cause of the first sound. That a sound corresponding with it in time is heard in an excised bloodless heart when it contracts is certain ; and therefore the first sound cannot be exclusively due, as some have asserted, to vibra- tions of the auriculo-ventricular valves when they are suddenly rendered tense by the contraction of the ventricles, for in a bloodless heart the valves are not stretched. Part of the sound must accord- ingly be associated with the muscular contraction as such. Again, the fact that the first sound is heard during the whole, or nearly the whole, of the ventricular systole is against the idea that it is exclusively due to the vibrations of membranes like the valves, which would speedily be damped by the blood and rendered in- audible. But while there is good reason to believe that the vibra- tion of the suddenly-contracted ventricles is the fundamental factor, the shock sets up vibrations also in the blood, the chest-wall, and MECHANICS OF THE HEART-BEAT 89 perhaps the resonant tissue of the lungs. Further, as we shall see later on (p. 760), the sound caused by a contracting muscle readily calls forth sympathetic resonance in the ear, and the peculiar boom- ing character of the first sound may be due to the superposition of these various resonance tones upon the muscular note. But, in addition, the vibration of the auriculo-ventricular valves un- doubtedly contributes to the production of the sound, and some observers have been able to distinguish in the first sound the valvular and the muscular elements, the former being higher in pitch than the latter, but a minor third below the second sound. In the excised empty heart the deeper tone of the first sound is alone heard, while the higher note is elicited when in a dead heart the auriculo-ventric- ular valves are suddenly put under tension (Haycraft) . When the mitral valve is prevented from closing by experimental division of the chordae tendineas, or by pathological lesions, the first sound of the heart is altered or replaced by a ' murmur.' This evidence is not only important as regards the physiological question, but of great practical interest from its bearing on the diagnosis of cardiac disease. It may be added that the point of the chest-wall at which the first sound is most easily recognized is also the point at which a changed sound or murmur connected with disease of the mitral valve is most distinctly heard. The sound is, therefore, best conducted from the mitral valve along the heart to the point at which it comes in contact with the wall of the chest. Changes in the first sound con- nected with disease of the tricuspid valve are heard best, in the com- paratively rare cases where they can be distinctly recognized, in the third to the fifth interspace, a little to the right of the sternum. The second sound is caused by the vibrations of the semilunar valves when suddenly closed, ' the recoiling blood forcing them back, as one unfurls an umbrella, and with an audible check as they tighten ' (Watson). The sharpness of its note is lost, and nothing but a rushing noise or bruit can be heard, when the valves are hooked back and prevented from closing. It is altered, or replaced by a murmur, when the valves are diseased. As there is a mitral and a tricuspid factor in the first sound, so there is an aortic and a pul- monary factor in the second. The place where the second sound is best heard (over the junction of the second right costal cartilage and sternum) is that at which any change produced by disease of the aortic valves is most easily recognized. The sound is conducted up from the valves along the aorta, which comes nearest to the surface at this point. Changes connected with disease of the pulmonary valves are most readily detected over the second left intercostal space near the edge of the sternum, for here the pulmonary artery most nearly approaches the chest- wall. The first sound is ' systolic ' — that is, it occurs during the ventricular systole; the second is ' diastolic/ beginning at the commencement of the diastole THE CIRCULATION OF THE BLOOD AND LYMPH Various explanations of the third sound have been given, but, as the authors who have studied it are not even agreed as to whether it is produced at the auriculo-ventricular orifices or at the aortic and pulmonary orifices, it would not be useful to discuss them at present. The Cardiac Impulse. — A surface-movement is seen, or an impulse felt, at every cardiac contraction in various situations where the heart or arteries approach the surface. The pulsation, or impulse, of the heart, often styled the apex-beat, is usually most distinct to sight and touch in a small area lying in the fifth left intercostal space, between the mammary and the parasternal line,* and gener- ally, in an adult, about an inch and a half to the sternal side of the former. It is due to the systolic hardening of the ventricles, which are here in contact with the chest-wall, the contact being at the same time rendered closer by their change of shape, and by a slight movement of rotation of the heart from left to right during the contraction (Practical Exercises, p. 207). When the left ventricle is in contact with the chest at the position of the apex-beat, as is usually the case, an important element in the impulse is the actual forward thrust of the apex. When the apex-beat corresponds in position with the right ventricle, there is no actual forward movement, although the hardening of the ventricle may be felt as a thrust by the ringer. Even in health the position of the impulse varies somewhat with the position of the body and the respiratory movements. In children it is usually situated in the fourth intercostal space. In disease its displacement is an important diagnostic sign, and may be very marked, especially in cases of effusion of fluid into the pleural cavity. It is sometimes, though not invariably, a little lower in the standing than in the sitting position, and shifts an inch or two to the left or right when the person lies on the corresponding side. Various instruments, called cardiographs, have been devised for magnifying and recording the movements produced by the cardiac impulse. Marey's cardiograph (Fig. 27) consists essentially of a small * The mammary line is an imaginary vertical line supposed to be drawn on the chest through the middle point of the clavicle. It usually, but not necessarily, passes through the nipple. The parasternal line is the vertical line lying midway between the mammary line and the corresponding border of the sternum. Fig. 27. — Diagram of Marey's Cardiograph. MECHANICS OF THE HEART -BEAT gi chamber, or tambour, filled with air, and closed at one end by a flexible membrane carrying a button, which can be adjusted to the wall of the chest. This receiving tambour is connected by a tube with a recording tambour, the flexible plate of which acts upon a lever writing on a travelling surface — a uniformly-rotating drum, for example — covered with smoked paper. Any movement communicated to the button forces in the end of the tambour to which it is attached, and thus raises the pressure of the air in it and in the recording tambour; the flexible plate of the tatter moves in response, and the lever transfers the movement to the paper. The tracing, or cardiogram, obtained in this way shows a small elevation corresponding to the auricular systole, succeeded by a large abrupt rise corresponding to the beginning of the first sound, and caused by the ventricular systole. This ventricular elevation is the essential portion of the curve ; it is alone felt by the palpating hand, and the auricular elevation is often absent from the cardiogram in man. The rise is maintained, with small secondary oscillations, for about 03 of a second in a tracing from a normal man, then gives way to a sudden de- scent, that marks the relaxation of the ventricles, the beginning of the second sound, and the closure of the semilunar valves. An interval of about 0-5 second elapses before the curve begins again to rise at the next auricular contraction. Such was the interpretation which Chauveau and Marey put upon their tracings. . Although neither their results nor their deductions from them have Pi& 28.— Cardiogram taken with Marey s escaped the criticism of succeed- Cardiograph A, auricular systole; ine investigators it is doubtful V> ventncular systole ; D, diastole. whether *Sv Seonate Reason The arrow shows the directi°Q * which whether any adequate reason the tracing is to be read. has been brought forward for discarding them, and Chauveau has furnished further proofs of their accuracy. The difficulties that beset the subject are great, for the cardiogram is a record of a complex series of events. The very rapid variation of pressure within the ventricles, the change of volume and of shape of the heart, the slight change of position of its apex, must all leave their mark upon the curve, which is besides distorted by the resistance of the elastic chest-wall, the inertia of the recording lever, and the compression of the air in the connecting tubes. It is only by comparing in animals the cardiographic record with the changes of blood-pressure in the heart and arteries that our present degree of knowledge of the human cardiogram has been attained. Could we register directly the fluctuations of pressure in the interior of the human heart, the cardiographic method would be rarely employed. For cl.nical purposes the receiving tambour can be advantageously replaced by a small glass funnel or a small metal cup, the open end of which is applied without a membrane over the cardiac impulse, the stem being connected with the recording tambour. In cases in which the right ventricle is in contact with the chest-wall at the position of the apex- beat the cardiogram is ' inverted ' — that is to . say, the chest-wall is drawn in during systole and protruded during diastole of the ventricles. Inversion of the cardiogram is, therefore, not an infallible sign of the pathological condition known as adherent pericardium (Mackenzie). Aur:\~- J I; A , 92 Endocardiac Pressure. — The function of the heart is to maintain an excess of pressure in the aorta and pulmonary artery sufficient to overcome the friction of the whole vascular channel, and to keep up the flow of blood. So long as the semilunar valves are closed, most of the work of the contracting ventricles is expended in raising the pressure of the blood within them. At the moment when blood begins to pass into the arteries, nearly all the energy of this blood is potential; it is the energy of a liquid under pressure. During a cardiac cycle the pressure in the cavities of the heart, or the endo- cardiac pressure, varies from moment to moment, and its variations afford important data for the study of the mechanics of the circulation. Manometers. — For the study of the endo- cardiac pressure, the ordinary mercurial manometer (p. no) is unsuitable, since, owing to the rela- tively great amount of work required to produce a given dis- placement of the mer- cury, it does not readily follow rapid changes of pressure, and the mercurial column, once dis- placed, continues for a time to execute vibrations of its own, which are compoun- ded with the true oscillations of blood-pressure. But by introducing in the connection between the manometer and the heart a valve so arranged as to oppose the passage of blood towards the heart, while it favours its passage towards the manometer, the maximum pressure attained in the cardiac cavities during the cycle may be measured with considerable accuracy. When the valve is reversed the apparatus becomes a minimum manometer. In this way it has been found that in large dogs the pressure in the left ventricle may rise as high as 230 to 240 mm. of mercury, and sink as low as — 30 to - 50 mm. ; while in the right ventricle it may be as much as 70 mm., and as little as — 25 mm. In the right auricle a maximum pressure of 20 mm. of mercury has been recorded, and a minimum pressure of — 10 mm. or even less. But these results were obtained under somewhat exceptional experimental conditions, and the normal maximum pressures in the heart cavities in man are probably not so high, especially in the right auricle and ventricle. Our knowledge of the maximum and minimum pressure attained in the cavities of the heart, even if it were far more precise than it actually is, would only carry us a little way in the study of the endo- cardiac pressure-curve, for it would merely tell us how far above the Vt nt: V AS VS Fig. 29. — Curves of Endocardiac Pressure taken with Cardiac Sounds. Aw., auricular curve; Vent., ven- tricular curve; AS, period of auricular systole, in- cluding relaxation; VS, of ventricular systole, including relaxation; D, pause. MECHANICS OF THE HEART-BEAT 93 base-line of atmospheric pressure the curve ascends, and how far below the base-line it sinks. To exhaust the problem, we require to have tracings of the exact form of the curve for each of the cavities of the heart, and to know the time -relations of the curves so as to be able to compare them with each other, and with the pressure-curves of the great arteries and great veins. To obtain satisfactory tracings of the swiftly- changing endocardiac pressure i is a task of the highest techni- / cal difficulty, and it is only in vvwvzittAJ. very recent years that it has been accomplished, with any ap- proach to accuracy by the use of elastic manometers, in which the blood-pressure is counter- balanced, not by the weight of a column of liquid aSKin.the Fig. 3o.-DiagrL of Hurthle's Elastic ManoT mercurial manometer, but by J^g, r> °small chamber covered by mem. the resistance to compression brane; /t tube communicating with interior 01 a small column of air or the Of heart; L, compound lever to magnify the tension of an elastic disc or of movements of the membrane, a spring. Modifications in the nature and dimensions of the elastic resistance of the recording apparatus and of the size of the cavity have produced successive improvements, as, e.g., in the manometers of Hiirthle (Fig. 30). The penetrating analysis of the principles of manometer construction by Frank has recently stimulated renewed investigation of the whole subject with the aid of instruments whose movements are optically re- Fig. 31. — Diagram of Optical Manometer (Wiggers). A is a vertical glass tube surmounted by a hollow brass cylinder, B, which contains a stopcock, C, whose lumen comes into apposition with a plate, a, having a small opening in it. By opening the stopcock more or less, the pulsations will be ' damped ' to a smaller or greater extent. Above a the cylinder ends in a segment capsule b (i.e., a capsule cut away at one side) 3 mm. in diameter, covered with rubber dam. Upon this a small piece of celluloid carrying a little mirror, c, is fastened, so that it pivots on the chord side of the capsule. Over the capsule and its recording mirror is mounted a support bearing an inclined reflecting mirror, E, adjustable about a horizontal axis by a screw, so that the image of the recording mirror appears within it. Upon this image a strong light is focussed. The incident rays are doubly reflected, as shown in the figure, and the movements of the capsule are thus greatly magnified. The beam of light is photographed on a moving plate. c-- corded on a photographic plate, so as to eliminate all unnecessary fric- tion. Fig. 31 is a diagram of the manometer devised by Wiggers on this principle. Hurthle's spring manometer consists of a sn^all drum covered with an indiarubber membrane, loosely arranged so as not to vibrate with a period of its own. The drum is connected with the heart or with a vessel, and the blood -pressure is transmitted to a steel spring by means of a light metal disc fastened on the membrane. The spring 94 THE CIRCULATION OF THE BLOOD AND LYMPH acts on a writing lever. The instrument is so constructed that for a given change of pressure the quantity of liquid displaced is as small as possible, and it is on this that its capacity to follow sudden varia- tions of pressure chiefly depends. The manometer is connected with the cavity of the heart by an appropriately curved cannula of metal or glass, which, after being filled with some liquid that prevents co- agulation (Practical Exercises, p. 211), is pushed through the jugular vein into the right auricle or ventricle, or through the carotid artery and aorta into the left ventricle. Some observers fill only the cannula with fluid, and leave the capsule of the elastic manometer and as much of the connections as possible full of air. Others fill the whole system with liquid. And around the question of the relative merits of ' trans- mission ' by liquid and by air has raged a controversy which, however, now shows signs of coming to an end. For there is reason to suppose that the character of the curves obtained is modified among other circumstances by the manner in which the pressure is transmitted, as it is certainly modified by the dimensions and mass of the moving parts and the method of recording. As Wiggers has pointed out, the differ- ences in the records obtained by different observers, even with the latest methods of optical registration, are determined largely by the sensitive- ness and degree of damping of the manometer. The Ventricular Pressure-Curve. — Thus, the pressure-curve of the ventricle, according to most of those who have employed mano- meters with liquid transmission and small inertia of the moving parts (Fig. 33), remains after the first abrupt rise, which undoubtedly corresponds to the ventricular systole, well above the abscissa line for a considerable time, and then 'descends somewhat less suddenly than it rose. This systolic ' plateau,' although usually broken by minor heights and hollows, which may be partly due to inertia oscilla- tions of the liquid or the recording apparatus, would indicate that the ventricular pressure, after its first swift rise, maintained itself at a considerable height throughout the greater part of the systole. The tracings yielded by most of the manometers with air trans- mission show the same suddenness in the first part of the upstroke and the last part of the descent — that is, the same abruptness in the beginning of the contraction and the end of the relaxa- tion. But they differ totally in the intermediate portion of the curve, which, climbing ever more gradually as it nears its apex, remains but a moment at the maximum, then immediately descend- ing forms a ' peak/ and not a plateau. It ought to be distinctly understood, however, that the use of the term ' plateau ' must not be taken to imply that the pressure remains constant and the curve parallel to the abscissa during this interval. Wiggers, using the optical method of recording the pressure- curve in the right ventricle (p. 93), finds that when the auricular pressure and the pressure in the pulmonary artery are normal the curve of intraventricular pressure may be divided into (i) an auric- ular period; (2) a period of rising pressure while the ventricle is contracting and its cavity is closed by the auriculo- ventricular and semilunar valves; (3) an ejection period during which the pressure MECHANICS OF THE HE ART -BE AT 95 still rises, reaches a summit, and then slowly falls; and (4) a relaxa- tion period (Fig. 32). Without entering further into a technical discussion, we may say Gtrotid Fig. 32. — Intraventricular Pressure Curves with Optical Recording (Wiggers). Three types of normal curves are reproduced, taken with manometers of different degrees of sensitivene'ss. The second at the left-hand side was taken with the least sensitive, a — b, auricular systolic; 6 — d, isometric period, during which the auriculo-ventricular and the semilunar valves are both closed; d — /, ejection period; after/, diastole. the bulk of the evidence goes to show that the plateau is not, as the advocates of the peak have claimed, an artificial phenomenon, but does in reality correspond to that continuation of the systole of the Fig- 33- — Simultaneous Record of Pressure in Left Ventricle (V) and Aorta (A). (Hurthle.) The tracings were taken with elastic manometers; o indicates a point just before the closure of the mitral valve; i, the opening of the semilunar valve ; 2, beginning of the relaxation of the ventricle ; 3, the closure of the semi- lunar valve; 4, the opening of the mitral valve. The ventricular curve shows a ' plateau.' ventricle, that dogged grip, if we may so phrase it, which it seems to maintain upon the blood after the greater portion of it has been expelled. g6 THE CIRCULATION OF THE BLOOD AND LYMPH This conclusion is essentially in accordance with the results of Chauveau and Marey, obtained long ago by means of their" ' cardiac sound,' which was in principle an elastic manometer. It consisted of an ampulla of indiarubber, supported on a frame- work, and communicating with a long tube, which was connected with a recording tambour. The ampulla was introduced into the heart (of a horse) through the jugular vein or carotid artery in the way already described. Sometimes a double sound was employed, armed with two ampullae, placed at such a distance from each other that when one was in the right ventricle the other was in the auricle of the same side. Each ampulla communicated by a separate tube in the common stem of the instrument with a recording tambour, and the writing points of the two tambours were arranged in the same vertical line. When any change in the blood-pressure takes place, the degree of compression of the ampullae is altered, and the change is transmitted along the air-tight connections to the recording tambours. On most of the endocardiac pressure tracings taken with modern manometers, whether the curves belong to the type of the peak or of the plateau, no sudden change of curvature, no nick, or crease, or undulation reveals the moment of opening or closure of any valve. This has been considered by some writers a striking tribute to the smooth working of the cardiac pump. There is reason to think, however, that the smoothness of the curve is still in some degree artificial, and on some of the records obtained by optical methods (Fig. 32) indications of changes of curvature, associated with the action of the valves, may be observed. But even in the absence of such indications, by experimentally graduating a pair of elastic manometers, and obtaining with them simultaneous records of the pressure in auricle and ventricle, or by using a ' differential ' mano- meter, in which the pressures in two cavities are qpposed to each other, so that the movement of the membrane corresponds to their difference, we can calculate at what points of the ventricular curve the pressure is just greater than and just less than the pressure in the auricle. The first point, it is evident, will correspond to the instant at which the mitral or tricuspid valve, as the case may be, is closed, and the second to the instant at which it is opened. And in like manner, by comparing the pressure-curve of the aorta with that of the left ventricle, the moment of opening and closure of the semi- lunar valves may be determined (Figs. 33 and 34). According to the best observations, the closure of the semilunar valves takes place at a time corresponding to a point on the upper portion of the descend- ing limb of the intraventricular curve. On the blood-pressure curve of the aorta, simultaneously registered, the corresponding point is near the bottom of the so-called ' aortic ' notch (p. 105) which precedes the dicrotic elevation. For clinical purposes, in man the moment of closure of the semilunar valves (denoted by the abbreviation S.C. point) may be taken as 0-03 second before the bottom of the aortic notch in sphygmographic tracings from the carotid, this being approximately the average time occupied MECHANICS OF THE HEART-BEAT 97 by the pulse-wave in travelling from the aorta to the carotid. The S.C. point, the A.O. point, or moment of opening of the auriculo- ventricular valves, and the beginning of the ventricular systole, are three important points of reference in the measurement and inter- pretation of pulse-tracings in clinical work. The A.O. point in man may be taken as a point ' 0-03 second in advance of the summit of the dicrotic wave ' on the carotid pulse-tracing (Lewis). But this is the most difficult of the three standard points to determine clinically with anything like accuracy. The study of the curves of endocardiac pressure enables us to add precision in certain points to the description of the events of the cardiac cycle which we have already given, and, as regards the ventricles, to divide the cycle into four periods: (1) A period dtiring which the pressure is lower in the ventricles than either in the auricles or the arteries, and the auricula-ventricular valves are consequently open, and the semilunar valves closed. This is the period of ' filling ' of the heart, or the pause. (2) A period, beginning with the ventricular systole, during which the pressure is increasing abruptly in the ventricles, while they are as yet completely cut off from the auricles on the one hand and the arteries on the other by the closure of both sets of valves. This is the period of ' rising pressure, ' during which the ventricles are, so to say, ' getting up steam.' The interval between the beginning of the ventricular systole and the opening of the semilunar valves is termed the ' presphygmic ' interval. (3) A period during which the pressure in the ventricles overtops that in the arteries, and the semilunar valves are open, while the aiiriculo- ventricular valves remain shut. This is the period of ' discharge ' or ' sphygmic ' period. (4) A period during which the pressure in the ventricles is again less than the arterial, while it still exceeds the auricular pressure, and both sets of valves are closed. This is the period of rapid relaxation. The interval between the closure of the semilunar and the opening of the auriculo-ventricular valves is sometimes called the ' post-sphygmic ' interval. Of the four periods, the second and fourth are exceedingly brief. The third is relatively long and constant, being but slightly depen- dent on either the pulse-rate or the pressure in the arteries. The duration of the first period varies inversely as the frequency of the heart ; with the ordinary pulse-rate it is the longest of all. From records taken in a person with a defect in the chest-wall which rendered the heart accessible the following results were obtained as to the duration of the various events of the cardiac cycle : First and fourth periods together, 0-445; third period, 0-254; second period (pre- sphygmic interval), 0-051 second, the pulse-rate being 80 a minute (Tigerstedt) . In another case with a similar defect the first period lasted 0-32, the fourth period (post-sphygmic interval) 0-06, the second and third periods together 0-4, and the auricular systole o-i second, the pulse-rate being 66. 7 98 THE CIRCULATION OF THE BLOOD AND LYMPH The Auricular (and Venous) Pressure-Curve. — The fluctuations of pressure in the auricles, although confined within narrower limits than in the ventricles, are of equal interest. They have been studied in considerable detail both in animals and by indirect methods in No fewer than three distinct elevations or ' positive waves,' man. separated or followed by three depressions or ' negative waves,' have been described on the curve of intra-auricular pressure. The first elevation corresponds to the systole of the auricle. The second coincides with the onset of the ventricular systole, and is Fig. 34. — Schematic Comparison of Pressure Curves in the Auricle (or Superior Vena Cava), the Ventricle and the Aorta in the Dog (Fredericq). In the auricular curve are to be distinguished ab, the first positive or presystolic wave, corre- sponding to the auricular systole (a wave) ; bb' or be, second positive wave or first systolic wave, which corresponds with the beginning of the ventricular systole (c wave); b'cd, the steep negative wave of which the beginning corre- sponds to the opening of the semilunar valves; def, the third positive wave (v wave), more or less serrated, ending at/, the point of opening of the auriculo- ventricular valves; fg, a negative wave corresponding to the relaxation of the ventricle. The time is indicated along the abscissa in tenths of a second, the pressure along the vertical axis at the left in mm. of mercury. probably due to the sudden bulging of the auriculo- ventricular valve into the auricle, or even to a slight regurgitation of blood from the ventricle through the valve before it has completely closed. The cause of the third elevation, which occurs during the period occupied in the ventricular pressure-curve by the plateau, is less clearly made out. In man, the events taking place in the right auricle during its MECHANICS Of THE HEARTBEAT systole can be followed to some extent by recording the venous pulse in the jugular veins, especially the internal jugular, at the root of the the neck (Fig. 36). Successful tracings can be obtained, not only in certain pathological conditions, but in many normal individuals, and it is probably only a matter of improved technique to obtain them in all. The jugular venous pulse-tracing Ifop fhfr If! " "" Identical features are observed on records of the normal venous pulse taken from veins of dogs near the heart, and on records of the pulse taken by a sound in the oesophagus. The oesophagus pulse is related to the pul- sation of the left auricle, the venous pulse to the changes of pressure in the right auricle. The first elevation, called the a (auricular) or p (presystolic) wave, begins with, and is the result of, the auricular systole. It is probably produced by stasis in the veins due to the contraction of the auricle, as well as to the effect of the impact of the auricular systole. The downstroke on the curve which succeeds this first elevation corresponds to the first negative wave or presystolic fall, which is due to the auricular relaxation . Thi* Fig- 35- — Schema of Events in the Cardiac Cycl«, in Relation to the Venous Pulse (Ewing). i, Tracing from Vena Cava, showing presystolic rise and fall, PR, PF (a wave) ; SR, systolic rise and fall (c wave); O', first onflow wave and, DR, diastolic rise and fall (v wave); O*, second onflow wave; 2, auricular myogram (tracing of contraction of auricle) ; 3, ventricular myogram (tracing of con- traction of ventricle); 4, record of the movement of the auriculo-ventricular septum; 5, ventricular volume curve (plethysmographic curve of dis- charge of the ventricles ) ; 6, curve of aortic pressure ; 7, intraventricular pressure-curve. fall of pressure is terminated by a rise — the second positive wave — which begins at the same moment as the ventricular systole, and is the ex- pression on the venous pulse-curve of that second elevation of the intra -auricular pressure whose probable cause has already been found in the sharp protrusion of the auriculo-ventricular valve into the auricular cavity under the stress of the ventricular systole while the semilunar valve are still closed. In addition to the actual bulging of the auriculo-ventricular valves, the impact of the sudden contraction of the ventricle on its contents transmitted through the valve to the contents of the auricle may aid in producing the rise of venous, pressure. The second elevation has been termed the c wave by certain too THE CIRCULATION OF THE BLOOD AND LYMPH writers, who studied it on jugular tracings, because they supposed it to be simply transmitted from the pulse in the adjacent carotid artery. This, however, has been shown to be erroneous, although it is true enough that pulsations transmitted from the great arteries of the thorax and neck may augment or distort the second elevation of the venous pulse. It has been proposed that the second positive wave should be called the s (systolic) wave. It lasts practically throughout the presphygmic period of the ventricular systole ; the opening of the semilunar valves, as indicated by the appearance of the pulse in the innominate artery, occurs just before the end of the second elevation (Porter, Ewing, etc.). The rapid discharge of the ventricle through the open semilunar valves, and its consequent diminution in size, especially in its longitudinal diameter, is associated with a dilatation of the auricular cavity and a fall of intra- auricular pressure which is expressed on the venous pulse -curve as the downstroke succeeding the second positive wave. This second negative wave gives place to the third positive wave, due to the steady inflow of blood into the auricle from the veins. According to Ewing, the third positive wave, the v Vs \\o\o Pulse Fig. 36. — Normal Apex and Venous Pulses, Photographically Recorded (reduced nine- tenths) (Niles and Wiggers). P, presystolic (or a) wave ; S, systolic (or c) wave ; Dlf first diastolic (v) wave. In this venous record a second diastolic wave, D2, is present. S1, vibrations corresponding to first sound ; S2, to second sound. wave of Mackenzie, really consists of two waves, the " first onflow wave " and the " diastolic rise " or d wave. This last is terminated by the third negative wave or diastolic fall of venous pressure coincident with the opening of the auriculo-ventricular valves. The re-examina- tion of the venous pulse with apparatus of which the moving parts have an exceedingly small mass and optical methods of recording (see p. 93) has confirmed the existence on the phlebogram of three essential waves. A fourth is sometimes added when the cardiac cycle is long. The first wave is clearly presystolic, the second systolic, as agreed by all observers who have used polygraph tracings. The third wave, however, is diastolic, as is, of course, the fourth when it exists. The position of these waves can be definitely fixed by simultaneous heart apex tracings, since on such optically recorded cardiograms the heart- sounds are represented by distinct vibrations, except where the chest wall is too thick or the heart overlaid by emphysematous lung. Some- times heart-sound vibrations may be present also on the record of the venous pulse (Wiggers). MECHANICS OF THE CIRCULATION IN THE VESSELS 101 The jugular curve, when properly interpreted, affords valuable information as to the action of the auricle, information of the same kind as that afforded by the arterial pulse-tracing and the cardiogram as to the action of the ventricle. In the in|^rprft°^on «•>*•'* he venous^ pulse-tracings, a simultaneous .record of the ja.rfia.1 or-Jae&er. the carotid puls'eTor ofjJTg_^pex-beat--i£ always import aiit>^ajid^often inj3rlpPn~n'hl'n. ?"r it^ryflhfp'i ; th? tilTIP nf nnSpt "f *^p wn?rTriilar "ystol°jt£Lbg ^Rrk"^ "p^" *hp phlphngram (the venous trace), and truTiacUitates the identification of the a wave, which must immedi- ately precede, and the c or s wave, which should coincide with the beginning of the contraction of the ventricle. The student must, however, be warned that the proper interpretation of such tracings in the study of cardiac disease is often difficult and requires special knowledge and training.* Suction Action of the Ventricle. — We have already said that a negative pressure may be detected in the cardiac cavities by means of a special form of mercurial manometer. This is confirmed by an examination of the tracings written by good elastic manometers, for the curves of both ventricles may often descend below the line of atmospheric pressure. The cause of this negative pressure has been much discussed. In part it may be ascribed to the aspiration of the thoracic cage when it expands during inspiration (p. 226). But since the pressure in a vigorously-beating heart may still become negative, when the thorax has been opened, and the influence of the respiratory movements eliminated, we must conclude that the recoil of the some- what narrowed, or at least distorted, auriculo-ventricular rings, and of elastic structures in the walls of the ventricles, exerts of itself a certain suction upon the blood. This, however, is not an important factor in the maintenance of the circulation. SECTION III. — PHYSICAL OR MECHANICAL PHENOMENA OF THE CIRCULATION IN THE BLOODVESSELS. The Arterial Pulse. — At each contracton of the heart a quantity of blood, probably varying within rather wide limits (p. 139). is forced into the already full aorta. If the walls of the bloodvessels were rigid, it is evident (p. 85) that exactly the same quantity would pass at once from the veins into the right auricle. The work of the ventricle would all be spent within the time of the systole, and only while blood was being pumped out of the heart would any enter it. Since, however, the vessels are extensible, some of the blood forced into the aorta during the systole is heaped up in the arteries, beyond which, in the narrow arterioles and in the capillary tract, with its relatively great surface, the chief resistance lies. The arteries are accordingly distended to a greater extent than before the systole, and, being elastic, they keep contracting upon their contents until the next systole over-distends them again. • In this way, during the pause, the walls of the arteries are executing a kind of elastic systole, * The necessary details must be sought in such works as Mackenzie's Diseases of the Heart.' loz THE CIRCULATION OF THE BLOOD AND LYMPH and driving the blood on into the capillaries. The work done by the ventricle is, in fact, partly stored up as potential energy in the tense arterial wall, and this energy is being continually transformed into work upon the blood during the pause, the heart continuing, as it were, to contract by proxy during its diastole. Thus, the blood progresses along the arteries in a series of waves, to which the name of ' blood-waves ' or ' pulse-waves ' may be given. ^Wherever the pulse-wave spreads it manifests itself in various ways — by an increase of blood-pressure, an increase in the mean velocity of the blood-flow, an increase in the volume of organs, and by the visible arrd palpable signs to which the name of pulse is commonly given in a restricted sense. The intermittence in the flow with which the pulse-wave is necessarily associated is at its height at the beginning of the aorta. In middle-sized arteries, such as the radial, it is still well marked, but it dies away as the capillaries are reached, and only under special conditions passes on into the veins, where, however, as has just been mentioned, pulsatory phenomena of a different origin maybe detected. The pulse was well known to the Greek physicians, and used by them to a certain extent as an indication in practical medicine. Harvey demonstrated with some clearness the relation of the pulse to the contraction of the heart, but Thomas Young was the first to form a proper conception of it as the outward token of a wave prop- agated from heart to periphery. When the finger is placed over a superficial artery like the carotid, the radial, or the temporal, a throb or beat is felt, which, without measurement, seems to be exactly coincident with the cardiac impulse. In certain situations the pulse can be seen as a distinct rhythmical rise and fall of the skin over the vessel. The throbbing of the carotid, especially after exertion, is familiar to everyone, and the beat of the ulnar artery can be easily rendered visible by extend- ing the hand sharply on the wrist. When the pulse is felt by the finger, it is not the expansion, but the hardening of the wall of the vessel, due to the increase of arterial pressure, that is perceived ; and even a superficial artery, when embedded in soft tissues so that it cannot be compressed, gives no token of its presence to the sense of touch. Sometimes an artery is longitudinally extended by the pulse-wave, and this extension may be far more conspicuous than the lateral dilatation. This is particularly seen when one point of the vessel is fixed and a more distal point offers some obstruction to the blood- flow, as at a bifurcation or in an artery which has been ligatured and divided. By means of the sphygmograph, the lateral movements of the arterial wall, or, rather, in man, the movements of the skin and other tissues lying over the bloodvessel, can be magnified and recorded. It would be very unprofitable to enumerate all the sphygmographs which ingenuity has invented and found names for. The first attempt MECHANICS OF THE CIRCULATION IN THE VESSELS 103 Fig. 37. — Scheme of Marey's Sphygmo- graph. A, toothed wheel connected with axle H, and gearing into toothed upright B; C, ivory pad which rests over blood- vessel and is pressed on it by moving G, a screw passing through the spring J ; E, writing-lever attached to axle H, and moved by its rotation. E writes on D, a travelling surface moved by clockwork F. to magnify the movements of the pulse was made by loosely attaching a thin fibre of glass or wax to the skin with a little iard, in order to demonstrate the venous pulse which appears under certain conditions. In all modern sphygmographs there is a part, usually button-shaped, which is pressed over the artery by means of a spring, as in Marey's and Dudgeon's sphygmographs, or by a weight, or by a column of liquid. In Marey's instrument, the button acts upon a toothed rod gearing into a toothed wheel, to which a lever, or a system of levers, is attached. The lever has a writing-point which records the movement on a smoked plate, or a plate covered with smoked paper, drawn uniformly along by clockwork (Figs. 37, 100). Special forms of sphygmographs (poly- graphs) have been devised, which, by the addition of one or more recording tambours, permit the simultaneous record of movements from two or more points of the vascular system — for example, the radial artery and the jugular vein, or the radial or carotid artery, jugular vein, and the apex of the heart. In rare cases, with de- fect of the chest wall, a tracing may be ob- tained even from the aorta (Fig. 40). In a normal arterial pulse-tracing (Fig. 38) the ascent or ana- erotic limb is abrupt and unbroken ; the descent or katacrotic limb is more gradual, and is interrupted by one, two, or even three or more, second- ary wavelets. The most important and constant of these is Fig. 38. — Pulse-Tracings, i, primary elevation ; 2, predi- crotic or first tidal wave ; 3, dicrotic wave. The depression between 2 and 3 is the dicrotic or aortic notch ; 3 is better marked in B than in A. C, dicrotic pulse with low arterial pressure; D, pulse with high arterial pressure — summit of primary elevation in the the one marked 3, which has received the name of the dicrotic wave. Usually less marked, and some- times absent, is the wavelet 2 between the dicrotic elevation and the apex of the curve. It is generally termed the predicrotic wave. Oscillations, due to vibrations of the recording apparatus, appear on many pulse- form of an ascending plateau. E, systolic anacrotic pulse; the secondary wavelet « occurs during the upstroke corresponding to the ventricular systole. F, presystolic anacrotic pulse; a occurs just before the systole of the ventricle. In this rarer form of anacrotism, a may sometimes be due to the auricular systole when the aortic valves are incompetent. io4 tracings, and it is important to recognize their cause, so that no weight may be given to them. In the explanation of the pulse-tracing, a fundamental fact to be borne in mind is the elasticity of the vessels. When an incompres- sible fluid like water is injected by an intermittent pump into one end of an elastic tube a wave is set up, which is transmitted to the other end of the tube. It is a positive wave — that is, it causes an increase of pressure and an expansion of the tube wherever it arrives; and if a series of levers be placed in contact with the tube, they will rise and sink in succession as the wave passes them. After the passage of this primary wave the walls of the tube, instead of coming instantly to rest in their original position, regain it by a series of oscillations, first shrinking too much, then expanding too much, but at each movement coming nearer to the position of equilibrium. Each vibration of the elastic wall is of course accompanied by a change of pressure in the contents of the tube. This change of pressure runs along the tube as a wave; and such waves, succeeding the primary one, may be called secondary waves of oscillation. These secondary waves will be set up in an elastic system whether the distal end of the system be closed or open. But if it is closed, or sufficiently obstructed without being actu- ally closed, secondary waves of another kind may also be generated, the primary wave on arriving at the distal end being reflected there. The reflected wave running back towards the central end may there again undergo reflexion, and pass out once more towards the distal end as a centrifugal, twice-reflected wave. When the liquid ceases to enter the tube at the end of the stroke, a wave of diminished pressure — a negative wave — is generated at the central end, and is propagated to the distal end, where it may be reflected just like the posi- tive wave. Although under certain conditions the dicrotic wave is so marked that the double beat of the pulse was discovered and named by physicians long before the in- vention of any sphygmograph, perhaps no physiological question has been more discussed or is less understood than the mechanism of its production. Two points, however, seem to be clear : (i) That it is a centrifugal, and not a centripetal, wave — that is to say, it travels away from, and not towards, the heart ; (2) that the aortic semilunar valves have something to do with its origin. It is not a centripetal wave, for in tracings taken at all parts of the arterial path, no matter what the distance from the heart and the Fig. 39. — Pulse - Tracings from Different Arteries (v. Frey). T, temporal; R, radial ; P, artery of foot. Fig. 40. — Pulse-Curve from Human Aorta (after Tiger - stedt). MECHANICS OF THE CIRCULATION IN THE VESSELS 105 capillaries (e.g., the origin of the carotid and the radial at the wrist), the dicrotic wave is separated by the same interval from the begin- ning of the primary elevation. This can only be explained by supposing that it has the same point of origin, and travels with the same velocity and in the same direction as the primary wave. It is not, then, a wave reflected directly from the peripheral distribution of the artery from which the pulse-tracing is taken. Some writers have contended that it is a centrifugal twice-reflected wave, and, indeed, traces of such waves may be detected in the vessels of newly-killed animals when changes of pressure of the same order of magnitude as the arterial pulse are artificially produced by a pump and recorded by elastic manometers connected with the interior of an artery. It has been supposed that these secondary waves are reflected first from peripheral points at which the blood-flow is particu- larly obstructed (the bifurcations of the larger arteries, and the small arteries and capillaries in general), and that, running towards the heart, they are again reflected outwards from the semilunar valves. It has been urged in support of this view that in very small animals (guinea- pigs) no dicrotic elevation occurs on the pulse-tracing, since the path which the reflected wave has to follow is so short that it arrives at the root of the aorta before the primary elevation is over. But this argument is by no means conclusive, and, indeed, the great difference in the distance from the heart of the numerous points at which reflection must take place is one of the chief difficulties of the hypothesis. For it is not easy to understand how the reflected fragments of the primary wave, arriving at different intervals at the heart, can be integrated into the single considerable dicrotic elevation. The explanation that best takes account of the facts and renders most clear the role of the semilunar valves is somewhat as follows: When the systole abruptly comes to an end and the outflow from the ventricle ceases, the column of blood in the aorta tends still to move on in virtue of its inertia, and a diminution of pressure, accom- panied by a corresponding contraction of the aorta, takes place behind it, just as a negative wave is set up in the central end of the elastic tube when the stroke of the pump is over. At the same moment, and while the semilunar valves are still for an instant in- completely closed, the diminution of pressure in the beginning of the aorta is intensified by the aspiration of the relaxing ventricle, which sucks the blood back against the valves, and draws them a little way into its cavity. A negative wave, therefore — a wave of diminished pressure, represented in the pulse-curve by the ' aortic notch ' — • travels out towards the periphery. The diminution of pressure is quickly followed by a rebound, as always happens in an elastic system. The recoiling blood meets the closed semilunar valves. The aorta expands again, and the expansion is propagated once more along the arteries as the dicrotic elevation. Lt is possible that this elevation may be reinforced by a reflected wave produced in the manner described. io6 THE CIRCULATION OF THE BLOOD AND LYMPH When the semilunar valve becomes incompetent in disease, or is rendered insufficient in animals by the artificial rupture of one or more of its segments, the dicrotic wave, as will be readily understood from the manner in which it is produced, either disappears altogether or is markedly enfeebled. But apart from any anatomical lesion or functional defect in the aortic valves, the prominence of the wave varies with a great number of circumstances, some of which are in a measure understood, while others remain obscure. It varies in par- ticular with the abruptness of discharge of the ventricle and the ex- tensibility of the arteries. The conditions are usually favourable when the arterial pressure is low, for the blood then passes quickly from the ventricle into the arteries, which, already only moderately tense, are easily dilated by the primary wave, then sharply collapse, and are again abruptly distended when the dicrotic wave arrives. And, in fact, an exaggeration of the dicrotic wavelet may be artificially produced by nitrite of amyl (Fig. 102, p. 209), a drug which lessens the blood-pressure by dilating the small arteries. Muscular exercise (Fig. 101, p. 209), running or bicycling, for instance, has a similar effect on the sphygmo- gram, although the explanation can scarcely be the same, since the blood- pressure mounts rapidly when moderate exercise begins, and only gradually falls during its continuance, with an abrupt decline to normal or below it on cessation of work (Bowen). The increase in the pulse- rate may have something to do in this case with the exaggeration of the dicrotism, which is very frequently, although by no means invariably, associated with a rapidly -beating heart, and therefore is often seen in fever. On the other hand, in certain diseases associated with a high arterial pressure, the dicrotic elevation almost disappears. Ather- omatous arteries, being very inextensible, do not allow a dicrotic pulse. Since the pulse represents a periodical increase and diminution in the amount of distension of an artery at any point, the line joining all the minima of the pulse-curve will vary in its height above the base-line, or line of no pressure, according to the amount of permanent distension, i.e., permanent blood-pressure, which the heart in any given circumstances is able to maintain. Any circumstance that tends to lessen the permanent distension will cause a fall of the line of minima, and any circumstance tending to increase the distension will cause that line to rise. If, for example, the arm be raised while a pulse-tracing is being taken from the wrist, the line of minima falls because the permanent pressure in the radial artery is diminished. The form of the pulse-curve varies in the different arteries, and therefore in making comparisons the same artery should be used. When the wave of blood only enters an artery slowly, the ascending part of the curve will be oblique. This is normally the case in a pulse-curve of a distant artery, such as the posterior tibial. The height of the wave is also less than in an artery nearer the heart, such as the carotid, or even the axillary, where the primary elevation is relatively abrupt (Fig. 39, p. 104). Anacrotic Pulse. — As a rule, the ascent of the tracing is unbroken by secondary waves, but in certain circumstances these may appear on it. This condition, which, when well marked at any rate, may be considered pathological, is called anacrotism (Fig. 38). It is seen when the discharge of the left ventricle into the aorta is slow and difficult — e.g., in cases where the orifice of the aorta has been MECHANICS OF THE CIRCULATION IN THE VESSELS 107 narrowed from disease of the semilunar valves (aortic stenosis). Since this condition is associated with hypertrophy and dilatation of the left ventricle, the slow emptying of the ventricle is partly due to the greater quantity of blood which it contains. In whatever way the delay in the emptying of the ventricle is brought about, the most probable explanation of the anacrotic pulse is that the delay affords time for one or more secondary waves to be developed in the arterial system before the summit of the curve has been reached, and that these are superposed upon the long-drawn primary elevation. In aortic insufficiency, where the left side of the heart is never cut off entirely from the aorta, the auricular impulse is sometimes marked on the pulse-curve as a distinct elevation; and this gives rise to a peculiar kind of anacrotic pulse, especially in the arteries nearest the heart (Fig. 38, F, p. 103). Frequency of the Pulse. — In health, the working of the cardiac pump is so smooth and apparently so self-directed that it needs a certain degree of attention to perceive that the rate of the stroke is not absolutely constant. It is, in reality, affected by many internal conditions and external influences. At the end of foetal life the rate is given as 144 to 133 ; from birth till the end of the first year, 140 to 123 ; from 10 to 15 years, 91 to 76 ; from 20 to 25 years, 73 to 69. It remains at this till 60 years, and increases again somewhat in old age.* At all ages the pulse is some- what quicker in the female than in the male, the excess amounting to about 8 beats a minute. So that if we take the average rate for a man (in the sitting position) as 72, the average for a woman will be 80. The difference is partly due to the fact that the average man is taller than the average woman; and it is known that in persons of the same sex and age the pulse-rate has an inverse relation to the stature. But there may be, in addition, a real sexual difference. It must not be forgotten that a certain number of perfectly healthy persons, who may even be noted for their powers of physical en- durance, have an habitually slow pulse, not above 50 in the minute. The position of the body exercises a slight, but relatively constant, influence on the rate, which is greater in the standing than in the sitting posture, and greater in the latter than in the recumbent position. And this is true even when muscular action is as far as possible eliminated by fastening the person to a board. The pulse * It must be remembered that these numbers are merely averages. Some healthy individuals have a much lower pulse-rate than 72 per minute, and some a rate considerably greater. Thus, while the average pulse-rate (taken in the sitting position) of 87 healthy (male) students, whose ages ranged from 18 to 36 years, was 73, the extreme variation was from 54 to 89. In the standing position the average was 80, and the variations 64 to 105. In the supine position, average 69, and variations 48 to .98. After a short spell of muscular exercise (generally running up and down some nights of stairs) the average in the standing position was 119, the variations 75 to 164, and the average increase 32. lo8 THE CIRCULATION OF THE BLOOD AND LYMPH is further affected by the respiratory movements, especially when they are exaggerated in forced breathing, being accelerated during each inspiration (p. 293). It is also increased by the taking of food, and especially of alcoholic stimulants, by muscular exercise, in fever and many other pathological conditions, and by a high external temperature. A warm bath, for example, causes a very distinct acceleration of the heart; and Delaroche found that in air at the temperature of 65° C. his pulse went up to 160. A cold bath may depress the pulse-rate to 60, or even less. During sleep it may fall to 50. It is greatly influenced by psychical events, and that in the direction either of an increase or a decrease. Finally, it ought to be remembered as of some practical importance that the pulse-rate in women and children, but particularly in the latter, is less steady than in men, and more apt to be affected by trivial causes. And it is a good general rule to let a short interval elapse after the finger is laid on the artery before beginning to count the pulse, so that the acceleration due to the agitation of the patient may have time to subside. Rate of Propagation of the Pulse-Wave. — When pulse-tracings are taken simultaneously at two points of the arterial system unequally distant from the heart, by two sphygmographs whose writing-points move in the same vertical straight line, it is found that the ascent of the curve begins later at the more distant than at the nearer point. Since waves like the pulse-wave travel with approximately the same velocity in different parts of an elastic system like the arterial ' tree,' this ' delay ' must be due to the difference in the length of the two paths. The difference in length can be measured; the time- value of the ' delay ' can be deduced from the rate of movement of the re- cording surface; dividing the length by the time, we arrive at the rate of propagation of the pulse-wave. The average rate has been found to be about 7 metres per second in man in the arteries of the upper limb, and 8 metres in those of the lower limb, the difference being due to the smaller distensibility of the latter. In sleep the velocity diminishes almost a metre a second. It increases in arterio- sclerosis, where the distensibility of the arteries is diminished, and in chronic nephritis with hypertrophy of the heart, in which the blood-pressure is increased. The mean velocity of the pulse-wave is about the same as the speed of a moderately fast steamship (say, 17 miles an hour), but less than that of a wave of the sea in a strong gale. The velocity of the pulse-wave must not be confounded with that of the blood-stream itself, which is not one-thirtieth as great. A ripple passes over the surface of a river at its own rate — a rate that is independent of the velocity of the current. The passage of the ripple is not a bodily transference of the particles of water of which at any given moment the wave is composed, but the propaga- tion of a change of relative position of the particles. The mere fact MECHANICS OF THE CIRCULATION IN THE VESSELS 10$ that the ripple can pass upstream as well as down is sufficient to illustrate this. The pulse-wave does not, however, correspond in every respect to a ripple on a stream, for the bodily transfer of the blood depends upon the series of blood-waves which the heart sets travelling along the arteries. Every particle of blood is advanced, on the whole, by a certain distance with every pulse-wave in which for the time it takes its place. But no particle continues in the front of the pulse- wave from beginning to end of the arterial system. The ' delay ' or ' retardation ' of the pulse (the interval, say, between the beginning of the ascent of the carotid and radial curves) is practically constant in the same individual, not only in health, but also in most diseases. But the retardation is markedly increased when the pulse-wave has to pass through a portion of an artery whose lumen is either greatly widened (in aneurism) or greatly constricted (in endarteritis obliterans). The Blood-Pressure Pulse in the Arteries. — In man it is only possible to trace the pulse-wave along the arteries by movements of the walls of the vessels transmitted through the overlying tissues. In animals the changes of pressure that occu" in the blood itself can be directly registered, and these changes may be spoken of as the blood-pressure pulse. At bottom, as already pointed out, the phenomenon is exactly the same as that we have been dealing with in our study of the external pulse. We are only now to follow, by a more direct, and in some respects a more perfect method, the same wave of blood along the same channel. Measurement of the Arterial Blood-Pressure. — Hales was the first to measure the blood-pressure. This he did by connecting a tall glass tube with the crural artery of a horse. The height to which the blood rose in the tube indicated the pressure in the vessel. Poiseuille, nearly half a century later, applied the mercury manometer, which had already been used in physics, to the measurement of blood-pressure. Ludwig and others improved this method by making the manometer self- registering by means of a float in the open limb, supporting a style which writes on a revolving drum, or kymograph. (For the method of taking a blood-pressure tracing, see p. 210.) For reasons already mentioned, the mercurial manometer is better suited for measuring the mean blood -pressure, or for recording changes in the pressure which last for some time, than for following the rapid variations of the pulse -wave. For the latter purpose, one of the class of elastic manometers is required (p. 93). A blood-pressure tracing taken from an artery with a manometer of this sort yields the truest picture of the pulse-wave which it is possible to obtain, because the reproduction of it is the most direct. The fact that such a tracing shows a close agreement with the trace of a good sphygmograph properly applied to the corresponding artery on the other side is a striking proof of the general accuracy of the sphygmographic method for physiological purposes, and enables us to guide ourselves in transferring to man, in whom, of course, the sphyg- mograph can alone be used, the information derived from direct manometric observations in 110 THE CIRCULATION OF THE BLOOD AND LYMPH For the same reason it is unnecessary to discuss the manometric tracings, as regards the pulsatory phenomena, in all their details. It will be sufficient to say that, while the form of the blood-pressure pulse-curve varies with the mean blood-pressure, the dicrotic wave is always marked on it, preceded by one or more oscillations falling within the period of the systole, and followed by one or more within the period of the diastole. When the blood-pressure is low, the first or primary elevation is the highest of the whole curve (Fig. 42). When the blood-pressure is high, the maximum falls later coinciding with one of the secondary systolic waves, bux always preceding the dicrotic wave; and the curve assumes an ana- erotic character. That all the secondary oscillations, including the dicrotic wavelet, are cen- trifugal, and not centrip- etal, may be shown, just as in the sphygmographic method, by recording the blood -pressure simultane- ously at two points of the arterial system at differ- ent distances from the heart — e.g., in the crural and carotid arteries. The secondary waves are found, by measuring the tracings, to reach the more distal point later than the more central. The increase of pres- sure during the systole, as indicated by the height of the primary elevation, is always very large, much larger than it appears in a tracing taken with a mer- cury manometer. In the rabbit this pulsatory variation is one -third to one -fourth of the minimum pressure. In the dog it is still greater, owing to the slower rate of the heart, and often amounts to 50 mm. of mercury, while under favourable conditions (low minimum pressure and slowly - beating heart) the systolic increase of pressure may be actually more than double the minimum (Hiirthle). Pick found also, by means of his spring man- ometer, that the pulsatory variations of blood-pressure were greater than the respiratory variations (p. 289), although in the records of the mercury manometer the reverse appears often to be the case. Landois, too, in the course of experiments in which a divided artery was allowed to spout against a moving surface, and to trace on it a sort of pulse-curve painted in blood (a haemautogram as it is called), Fig. 41.- — Arrangement for taking a Blood-Pressure Tracing. M, manometer; Hg, mercury; F, float armed with writing-point; A, thread attached to a wire projecting from the drum and supporting a small weight. The thread keeps the writing- point in contact with the smoked paper on the drum. B is a strong rubber tube connecting the manometer with the artery; C, a pinchcock on the rubber tube, which is taken off when a tracing is to be obtained. MECHANICS OP THE CIRCULATION IN THE VESSELS in observed that the rate of escape of the blood was nearly 50 per cent, greater during the systole than during the pause of the heart. The existence of the dicrotic wave on this tracing was long looked on as the best proof that it was not an artificial phenomenon. The wave of increased pres- sure, as it runs along the arterial system, carries with it wherever it arrives an increase of potential energy. But this excess of po- tential energy is continually being worn down, owing to the friction of the vascular bed ; and although in the comparatively large arteries the loss of energy is not great, it rapidly increases as the arteries approach their termination, and begin to break up into the narrow arterioles which feed the capillary network. For not only is the ratio of the total surface to the total cross-section, and therefore the friction, increased with every bifurcation, but the mere change of direction and division of the wave cannot take place without loss of energy. For this reason the fluctuations of blood-pres- sure are greater in the large arteries near the heart than in arteries smaller and more remote. In the wide and much-branched capillary bed the pulse-wave disappears altogether, and the blood-pressure becomes relatively constant or permanent. And it is for some purposes convenient to look upon the blood-pressure in the arteries as made up of a permanent element, with pulsatory oscillations super- posed on it. Since no portion of the arterial system is more than partially emptied in the interval between two blood-waves, the minimum or per- manent pressure is always positive — i.e., always above that of the atmosphere, the beats of the heart succeeding each other so rapidly that the successive waves over- lap or ' interfere,' and are only separated at their crests. If the heart is stopped while a blood-pressure tracing is being taken— and we shall see later on how this can be done (p. 157) — the minimum line of the tracing goes on falling towards the zero-line. Fig. 42. — Curves of Blood-Pressure taken with a Spring Manometer from the Carotid Artery of a Dog (Hiirthle). When i was taken the blood-pressure was high; 2 corresponds to a medium, 3 to a low, and 4 to a very low, blood- pressure; p is the primary elevation — this and the succeeding elevations between p and a are called systolic waves; the systolic waves are followed by a marked elevation d, which corre- sponds to the dicrotic wave. V v'W Fig- 43- — Blood - Pressure Tracing. The horizontal straight line inter- secting the curves is the line of mean pressure. When .the heart begins beating again, the pressure-curve rises, not by a continuous ascent, but by successive leaps, each corresponding to a beat of the heart, and each overtopping its predecessor, till the old line of minimum or of mean pressure is again reached. The mean arterial blood-pressure is the permanent pressure plus one-half of the average pulsatory oscillation. In a blood-pressure tracing the line of permanent pressure joins all the minima; the line of maximum pressure joins all the maxima ; the line of mean pressure is drawn between them in such a way that, of the area included between it and the blood-pressure curve, as much lies above as below it (Fig. 43). As has been said, a tracing taken with a mercury man- ometer gives approximately the mean blood-pressure. Each beat of the heart is represented on it by a single elevation of variable size, sometimes not amounting to more than one-twentieth of the height of the curve above the line of zero or atmospheric pressure, but some- times much larger. The oscillations due to the heart-beat are superposed upon much longer, and often, as registered in this way, larger waves, caused by the movements of respiration. So much having been said by way of definition, we have now to consider the amount of the mean arterial pressure, the variations which it under- goes, and the factors on which its maintenance depends. As to its amount, it will be sufficiently accurate to say that in the systemic arteries of warm-blooded animals in general (including birds), and of man in particular, the mean pressure does not, under ordinary conditions, descend much below 100 mm. of mercury, nor rise much above 200 mm. ; while in cold-blooded animals it seldom exceeds 50 mm., and may fall as low as 20 mm. It does not seem possible, at least with our present data, to further subdivide these two great groups; nor do we know precisely whether the distinction depends mainly on morphological or mainly on physio- logical differences, whether, that is to say, the warm-blooded animal has a higher blood-pressure than the cold-blooded chiefly because its vascular system (and especially its heart) is anatomically more perfect. or because its heart beats faster and works harder. It may be that it is for both of these reasons that the birds, which in certain other respects are more nearly related to the reptiles than to the mammals, mount, as regards the pressure of the blood, into the mammalian class, and that a manometer in the carotid of a goose will rise as high, or almost as high, as in the carotid of a horse, a sheep, or a dog, while the pressure in the aorta of a tortoise is no higher than in the aorta of a frog. But we know that the mere average rate of the heart has of itself com- paratively little influence on the blood-pressure within either group, for the heart of a rabbit beats, on the average, very much faster than the heart of a dog, and yet the arterial pressure in the dog is certainly at least as great as in the rabbit. Nor does the size of the body seem to have any definite relation to the mean pressure, even in animals of the same species ; and there is no reason to suppose that the pressure is materially less in the radial artery of a dwarf than in the radial artery of a giant. MECHANICS OF THE CIRCULATION IN THE VESSELS Measurement of the Blood-Pressure in Man. — In man the blood- pressure has been estimated by adjusting over an artery an instru- ment known as a sphygmomanometer or sphygmometer, which, in its most modern form, consists essentially of a hollow rubber pad or bag containing air, and connected with a metallic pressure gauge or a mercurial manometer. The simplest method is that devised by Riva-Rocci (Fig. 44). An armlet in the form of a broad rubber bag, supported externally by canvas or leather, is adjusted round the upper arm. The interior of the bag is connected with a mercury manometer, and also with a strong rubber bulb provided with a valve. By rhythmical compression of the bulb the pressure can be raised. Between the pressure bulb and the rest of the system is a thin rubber balloon, which by its distension renders the changes of pressure more gradual. The finger of the observer is placed over the radial artery, ^nfl th*» pressure is raised until the, pulse disappears. Then the pressure is allowed to fall gradual! vT~ and the manometer reading at the moment when the pulse first reappears in the radial gives the maximum or systolic pressure in the brachial artery. Instead of palpating the radial artery, a stetho- scope may be placed over the brachial just below the edge of the armlet, according to the method of Korotkoff , by which, in addition to the systolic, the determined^ This is much the best of all the methods of measuring the arterial pressures in man. The pressure is raised somewhat above that necessary to obliterate the pulse, and then allowed to fall slowly. At the moment when pulsations first begin to break through below the armlet, a succession of sharp taps, synchronous with the pulse, is heard. The tapping sound grows rapidly louder as the artery opens up more and more, then abruptly diminishes and changes its character, and gradually dis- appears. Several phases have been distinguished after the first maxi- mum. Everybody agrees that the pressure shown by the manometer thft gmmd is first heard is the svst<">1i>. jr^gg^r^ This corresponds Fig. 44. — Riva-Rocci Apparatus. a, armlet; b, manometer tube; c, bottle containing mercury, into which b dips ; d, thin rubber bulb ; e, thick rubber bulb for getting up pressure. , closely with the systolic pressure as determined by palpating the radial ; and it can be shown experimentally that at pressures in the armlet exceeding this the lumen of the brachial artery is actually obliterated and not merely narrowed to such a degree as to prevent the passage of the pulse wave, while still permitting the passage of some blood (see Practical Exercises, p. 213). The diastolic pressure is the, pressure at which a. sudden ^weakening arm-firming ol the sound occurs (beginning of the fourth phase), and not tne pressure at which the sound becomes altogether inaudible (Mac- William, etc.). The sound seems to be due to vibrations set up in the walls of the artery and the structures in con-tact with it, when it is suddenly opened by the pulse waves. According to Erlanger, the essen- tial thing is the ' water hammer ' action of the blood when, moving through the artery under compression in the armlet, it strikes the stag- nant blood in the uncompressed artery below and distends it. THE CIRCULATION OF THE BLOOD AND LYMPH Various instruments have been devised for determining the blood pressure from the changes in the oscillations communicated to the arm- let by the artery as the pressure in the armlet is allowed to fall or caused to rise. The sphygmomanometer of Erlanger (Fig. 45) is arranged to obtain graphic records of the pulse from which both the maximum and the minimum blood-pressures may be deduced. The mean pres- sure cannot be directly measured, but must lie much nearer to the minimum than to the maximum, since the line of mean pressure bi- sects the area enclosed by the pulse curve, and this area is broad at the base and narrow at the apex. The rubber bag is applied in the form of a cuff or armlet to the arm above the elbow over the brachial artery. It communicates with a mercury manometer, which gives the pressure exerted upon the arm. It is also connected with a rubber bulb, B, enclosed in a glass bulb, G, and through a stopcock with a syringe bulb, V, .provided with a valve. ^The space between" B and G communicates (i) with the tambour ; (2) with the exterior through the stopcock by the tube E, and also through a pin- point opening in the membrane of the tambour. While the armlet is being ad- justed the stopcock is * turned so that the rubber bag is in communication with the external air through F. The same is true of the space TS in the glass -bulb. The tam- bour is thus protected against undue strain during adjustment* The stopcock is now rotated so as to cut off the armlet from the exterior and to permit the entrance of air through F from V, which is used as a pump to raise the pressure, the space TS and the tambour being still in communication with the exterior. When the desired pressure has been reached, the stopcock is turned into an intermediate position, which cuts off both the armlet and the space TS from the exterior, and the pulse is then transmitted to the tambour and recorded on the drum. By certain adjustments of the Fig. 45. — Sphygmomanometer of Erlanger. MECHANICS OF THE CIRCULATION IN THE VESSELS 115 stopcock air can be allowed to escape more or less rapidly from the armlet. To determine the maximum or systolic blood-pressure, the air- pressure in the armlet is raised considerably (about 50 mm. Hg) above what it is expected to be. While the lever is writing on the drum the small oscillations due to the impact on the bag of the pulse-waves in the central portion of the obliterated artery, the pressure is gradually diminished by allowing air to escape. At the moment when the pressure upon the arm falls below the maximum blood-pressure, and the pulse-wave is first able to break through the brachial artery, the oscillations of the lever will more or less abruptly increase in amplitude. The pressure shown by the manometer at this point is the systolic blood-pressure. To obtain the minimum or diastolic pressure, the air- pressure in the armlet is raised somewhat (10 to 15 mm. Hg) above the pressure expected. The pressure is diminished by 5 mm. Hg at a time, records of the oscillations being taken on the drum. The manometer reading at the point at which the oscillations, after reaching the maxi- mum, begin abruptly to diminish, corresponds to the minimum blood- pressure. According to Brooks and Luckhardt, the criteria which are supposed to fix the systolic and diastolic pressures in these and similar methods yield results which are too high. The personal equation also seems to introduce a considerable error in the selection of the points on the trac- ings at which the characteristic changes are supposed to occur (Kilgore). In using the sphygmometer of Hill and Barnard (Fig. 46), the bag is inflated till the pulsation indicated by the gauge reaches a maxi- mum. The mean pressure shown at this point is assumed to be equal to.or some- what greater than, the dia- stolic pressure. The effect of muscular exer- cise upon the pressure is in- fluenced by the nature of the work. Such an effort as the lift- ing of a heavy weight causes a sudden and great increase, which is very transient. Thus, the average arterial pressure in a number of men was in before, 1 80 during, and no two to three minutes after the lift (McCurdy). The rise of pressure in this case is due largely to the marked diminution of the calibre of the bloodvessels mechanically produced by the strong and sustained contraction of the muscles. This increases the resistance to the passage of the blood along the arteries, while the veins are emptied by the pressure, and more blood thus reaches the right side of the heart. It is obvious that the heart and vessels may easily be exposed to an injurious strain during such efforts. In such an exercise as running, while the pressure mounts to some extent at first, as already mentioned, Fig. 46. — Sphygmometer of Hill and Barnard. It consists of a broad armlet, A, connected by a T-tube, B, with a pressure gauge, C, and a small compressing air-pump, D, fitted with a valve. u6 THE CIRCULATION OF THE BLOOD AND LYMPH the rise is not maintained, owing to the dilatation of the cutaneous vessels. In the anterior tibial artery of a boy whose leg was to be amputated, the blood-pressure, measured by means of a manometer connected directly with the artery, was found to vary from 100 to 1 60 mm., according to the position of the body and other circum- stances. In a woman sixty years old, in good health, the following readings were obtained with a sphygmomanometer : June 28 - 126 — 130 mm. of mercury. ,, 29 ----- 126 — 136 „ Aug. 3 132 — 144 ,,7 134— M0 „ 12 136 — 144 Such measurements on man show that the mean blood-pressure under similar conditions in one and the same artery, and in one and the same individual, may vary for a considerable time only within comparatively narrow limits. This relative constancy of the general arterial pressure is the result of a delicate adjustment between the work of the heart, the resistance of the vessels, and the volume of the circulating liquid. The quantity of the blood is tolerably steady in health, and con- siderable changes may be artificially produced in it (p. 191) without affecting the pressure in any great degree. On the other hand, the work of the heart and the peripheral resistance are highly variable and vastly influential. A narrowing of the arterioles throughout the body or in some extensive vascular tract increases the peripheral resistance ; and if the heart continues to beat as before, the pressure must rise. If the arterioles are widened, while the heart's action remains unchanged, the pressure must fall. In like manner an increase or a decrease in the activity of the heart, in the absence of any change in the peripheral resistance, will cause a rise or a fall in the blood-pressure. But if a slowing of the heart is accompanied by an increase in the peripheral resistance, or a dilatation of the arterioles by an increase in the activity of the heart, the one change may be partially or completely balanced by the other, and the pressure may vary within narrow limits or not at all. Not only is the mean pressure, as measured in a large artery, tolerably constant, but if recorded simultaneously in two arteries at different distances from the heart, it is seen to decrease very gradu- ally so long as the arteries remain large enough to hold a cannula. It is nearly as high, for instance, in the crural artery of a dog as in the carotid. It is easy to see that this must be so, for the resistance of the arteries between the point where the arterioles are given off and the heart is only a small fraction of the total resistance of the vascular path; and we have said (p. 84) that the lateral pressure at any cross-section of a system of tubes through which liquid is flow- ing is proportional to the resistance still to be overcome. This is also the reason why the pressure is always much lower in the pul- monary artery and right ventricle than in the aorta and left ventricle MECHANICS OF THE CIRCULATION IN THE VESSELS 117 (only one-fifth to one-sixth as great), for the total resistance of the vascular path through the lungs is much less than that of the systemic circuit. In dogs with natural respiration the pressure in the pulmonary artery was found to vary between 14 and 26 mm. of mercury, averaging about 20 mm. The Velocity-Pulse. — We have seen that the blood is propelled through the arteries in a series of waves that travel from the heart towards the periphery. The particles in the front of the pulse-wave are constantly changing, but since every section of the arterial tree is successively distended, every section contains more blood while the pulse-wave is passing over it than it contained immediately before. And since there is always a fairly free passage for this blood towards the periphery, there is a bodily transfer on the whole of a certain quantity with every wave. The translation of the blood, instead of being entirely intermittent, as it would be in a rigid tube or in an elastic system with a slow action of the central pump, is to some extent constantly going on ; for a portion of a blood- wave is always passing through every section of the arterial channel. Thus, we arrive at the same distinction as to the onward movement of the blood itself as we previously reached in regard to the blood-pressure, the distinction between the constant or permanent factor of the velocity and the periodic factor, which we may call the velocity-pulse. The Velocity of the Blood. — By the velocity or rate of flow of a river we should mean, if the flow were uniform throughout the whole cross- section, the rate of movement of any given portion or particle of the water. If we could identify a portion of the water, we could determine the velocity by measuring the distance travelled over by that portion in a given time. If the velocity was uniform over the channel, we could predict the actual time which would be required to traverse any fractional part of the measured distance. If, however, the velocity of the current changed from point to point, then we could only deduce from our observation the mean rate of the river for the measured dis- tance. To determine the actual rate for any given portion of this distance over which the rate was uniform, we should have to make a separate observation for this portion alone. But as soon as we pass from an ideal frictionless river to an actual stream, in which the water at the bottom and near the banks flows more slowly than that in the middle and on the surface, we are in every case restricted to the notion of mean velocity. We may distinguish between the velocity of different parts of the current, between that of the mid-stream and the side current, the bottom and the surface layers ; but when we consider the river as a whole, we take cognizance only of the mean or average velocity. And at any cross-section this may be defined as the volume of water passing per hour, or whatever the unit of time may be, divided by the cross-section of the current. It is evident that this does not enable us to determine the actual velocity of any given particle of the water at any given moment within a measured interval ; nor does it tell us whether or not the average velocity of the current has itself undergone variations within the period of observation. n8 We have dwelt upon this point because the measurement of the velocity of the blood, to which we must now turn, involves the same considerations. Within the smaller arteries, as the microscope shows us, and as we should in any case expect from what we know of fluid motion, the blood-current, apart from the periodical varia- tions in its velocity, due to the action of the heart, varies in speed from point to point of the same cross-section. The layer next the periphery of the vessel, the so-called peripheral plasma-layer or Poiseuille's space, moves more slowly than the central portion, the axial stream. In fact, we must suppose that in the large as well as in the small vessels the layer just in contact with the vessel-wall is at rest, while the stratum internal to this slides on it and has its velocity diminished by the friction. The next layer again slides on the last, but since this is already in motion, its velocity is not so much diminished, and so on. The velocity must therefore increase as we pass towards the axis of the bloodvessel, and reach its maxi- mum there (p. 193). Again, the velocity must be altered wherever an alteration occurs in the width of the bed, that is, in the total cross-section of the vascular system ; for since as much blood comes back in a given time to the right side of the heart as leaves the left side, the same quantity must pass in a given time through every cross-section of the circula- tion. Wherever the total section of the vascular tree increases, the blood-current must slacken; wherever it diminishes, the current must become more rapid. Now, the total section, increasing some- what as we pass from the heart along the branching arteries, under- goes an abrupt augmentation, and reaches its maximum in the capillary region. It suddenly diminishes again at the venous end of the capillary tract, and then more gradually as we pass heart- wards along the veins, but never becomes so small as in the arterial tract. We must, therefore, expect the mean velocity to be greatfig* in the large arteries, less in the veins, and least in the arterioles, capillaries, and venules. It must, of course, be remembered that the total section varies from time to time at any given distance from the heart. The capillary tract is especially variable in its area, and capillaries full of blood at one moment may be collapsed and empty at another, according to the changes of calibre and pressure in the arteries which feed and the veins which drain them. Although in strictness we are only at present concerned with the arteries, it will be well to consider here what a change of velocity at any part of the vascular channel really implies. To say that when the channel widens the velocity diminishes is not to explain the meaning of this diminution. A diminution of velocity implies a diminution of kinetic energy, and it is necessary to know what becomes of the energy that disappears. The stock of energy imparted by the contraction of the heart to a given mass of blood constantly diminishes as it passes round from the aorta to the right side of the heart, for friction is constantly being overcome and heat generated. This energy, ng as we have seen, exists in a moving liquid in two forms, potential and kinetic, the former being measured by the lateral pressure, the latter varying directly as the square of the velocity. Whenever the velocity, and therefore the kinetic energy, of a given mass of the blood is diminished without a corresponding increase in the potential energy, some of the total stock of energy must have been used up to overcome resistance (p. 84). In a uniform, rigid, horizontal tube, as has been already remarked, the velocity (and consequently the kinetic energy) is the same at every cross-section of the tube, while the potential energy, represented by the lateral pressure, diminishes regularly along the tube. When the calibre of the tube varies, it is different. Suppose, for instance, that the liquid passes from a narrower to a wider part, the velocity must diminish in the latter. The kinetic energy of visible motion which has disappeared must have left something in its room. Here there are three possibilities: (i) The kinetic energy that has disappeared may be just enough to overcome the extra friction in the wider part of the tube due to eddies and consequent change of direction of the lines of flow; in this case the potential energy of a given mass of the liquid will be the same at the beginning of the wider part as in the narrower part. The lost kinetic energy will have been transformed into heat. (2) The kinetic energy which has disappeared may be greater than is enough to overcome the extra resistance ; a portion of it must, therefore, have gone to increase the potential energy, and the lateral pressure will be greater in the wide than in the narrow part. (3) The lost kinetic energy may be less than enough to overcome the extra resistance; in this case both the lateral pressure and the velocity will be less in the wide than in the narrow part. It has been experimentally shown that when a narrow portion of a tube is succeeded by a considerably wider portion, and this again by a narrow part, case (2) holds; and the liquid may, under these conditions, actually flow from a place of lower to a place of higher lateral pressure. In the vascular system the conditions are not the same. The widening of the bed which takes place as we proceed in the direction of the arterial current is not due to the widening of a single trunk, but to the branching of the channel into smaller and smaller tubes. In the larger arteries the increase of resistance is so gradual that both the potential and the kinetic energy diminish only slowly, and the lateral pressure and velocity are not much less in the femoral artery than in the aorta or carotid. But in the arterioles the friction increases so greatly that although the velocity, and therefore the kinetic energy, in the capillary region is much less than in the arteries, the amount of kinetic energy lost is not upon the whole equivalent to the energy consumed in overcoming the extra resis- tance; the potential energy of the blood is also drawn upon, and the lateral pr.essure falls sharply in the capillary region, as well as the velocity. Where the capillaries open into the veins, the lateral pres- sure again sinks abruptly, while the velocity begins to increase, till in the largest veins it is probably about half as great as in the aorta. Where does the extra kinetic energy of the blood in the veins come from ? To say that the vascular channel again contracts as the blood passes from the capillaries into the veins, and that, since the 120 THE CIRCULATION OF THE BLOOD AND LYMPH same quantity must flow through every cross-section of the channel, the velocity must necessarily be greater in the narrower than in the wider part, does not answer the question. The greater portion of the kinetic energy of the arterial blood is, as we have seen, destroyed, or, rather, changed into an unavailable form, into heat, in the capil- lary region. The mean velocity of the blood in the capillaries is not more than -g-^ to ;I^ of the velocity in the aorta; the kinetic energy of a given mass of blood in the capillaries cannot therefore be more than (-2-5T7)2, or 4 0 * 0 0 of its kinetic energy in the aorta. In the veins, taking the velocity at half the arterial velocity, the kinetic energy of the mass would be one-fourth of that in the aorta, or at least 10,000 times as great as in the capillary region. This extra kinetic, energy comes partly from the transformation of some of the poten- tial energy of the blood. The resistance in the veins is very small compared with that in the capillaries; less of the potential energy represented by the lateral pressure at the end of the capillary tract is required to overcome this resistance, and some of it is converted into the kinetic energy of visible motion, the lateral pressure at the same time falling somewhat abruptly. Contributory sources of kinetic energy in the veins are the aspiration caused by the respira- tory movements and the pressure caused by muscular contraction in general, which, thanks to the valves, always aids the flow towards the heart. From these two sources new energy is supplied, to rein- force the remnant due to the cardiac systole (p. 133). Measurement of the Velocity of the Blood — i . Direct Observation. — (a) This method can be applied to transparent parts by observing the rate of flow of the corpuscles under the microscope. But it is only where the blood moves slowly, as in the capillaries, that the method is of use. (6) Part of the path of the blood through a large vessel may be artificially rendered transparent by the introduction of a glass tube, of approximately the same bore as the vessel (Volkmann). The tube is filled with salt solution, and the blood admitted by means of a stop- cock at the moment of observation. The time which the blood takes to pass from one end of the tube to the other is noted, and the length divided by the time gives the velocity of the blood in the tube. If the calibre of the tube is the same as that of the artery, this is also the velocity in the vessel ; but if the calibre is different, a correction would have to be made. The method is not a good one, for the reason, among others, that the long tube introduces an extra resistance. 2. Ludwig's Stromuhr. — This instrument measures the quantity of blood which passes in a given time through the vessel at the cross- section where it is inserted. It consists of a U-shaped tube, with the limbs widened into bulbs, but narrow at the free ends, which are con- nected with a metal disc. By rotating the instrument, these ends can be placed alternately in communication with a cannula in the central, and another in the peripheral, portion of a divided artery; or they can be placed so that none of the blood passes through the bulbs, but all goes by a short-cut. One limb of the instrument is filled with oil, and the other with defibrinated blood. The limb containing the oil is first put into communication with the central end, and that con- taining the blood with the peripheral end, of the artery. The blood MECHANICS OF THE CIRCULATION IN THE VESSELS 121 from the artery rushes in and displaces the oil into the other limb, the defibrinated blood passing on into the circulation. As soon as the blood has reached a certain height, indicated by a mark, the instrument is reversed and the oil is again displaced into the limb it originally occupied. This process is repeated again and again, the time from beginning to end of an experiment being carefully noted. The number of times the blood has filled a bulb in that period, the capacity of the bulb and the cross-section of the vessel being known, all the data required for calculating the velocity of the blood in the vessel have been obtained. Suppose, for example, that the cap- acity of the bulb up to the mark is 5 c.c., and that it is filled twelve times in a minute, the quantity flowing through the cross-section of the artery is I c.c., or 1,000 cub. mm., per second. Let the diameter of the vessel be 3 mm., then its sectional area is re x (- \ =— — -=7-06 sq. mm. The velocity is -- ^ = 11 mm Fig. 47.— StromuhrofLudwigand Dogiel. A, B, glass bulbs; a, a metal disc, to which A and G are attached, and which can be rotated on the disc 6 ; E, F, can- nulae attached to b, and con- nected with the peripheral and central ends of a divided blood- vessel. At the beginning of the experiment, A and the junction between A and B are filled with oil; B is filled with physiological salt solution or defibrinated blood: a being turned into the position shown in the figure, the blood passes through F and D into A , and the oil is forced into JB. As soon as the blood has reached the mark m, the disc a, with the bulbs, is rapidly ro- tated, so that C is now opposite F. The blood now passes into B, and the oil is again driven into A. When the oil has reached D, reversal is again made, and so on. per second. Various improvements in this method have been made, such as a graphic regis- tration of the reversals of the stromuhr. 3. A tube or box, in which swings a small pendulum, is inserted in the course Fig. 48.— Pitot's Tubes. of the vessel. The pendulum is deflected by the blood, and the amount of the deflection bears a relation to the ve- locity of the stream (Vierordt's hczmatachometer ; Chauveau and Lortet's much more perfect dromograph) (Fig. 49). 4. Pilot's Tubes. — If two vertical tubes, a and b, of the form shown in Fig. 48, be inserted into a horizontal tube in which liquid is flowing hi the direction of the arrow, the level will be higher in a than would be the case in an ordinary side-tube without an elbow ; in b it will be lower. For the moving liquid will exert a push on the column in a, and a pull 122 THE CIRCULATION OF THE BLOOD AND LYMPH on that in b. The amount of this push and pull will vary with the velocity, so that a change in the latter will correspond to an alteration in the difference of level in the two tubes. Instruments on this prin- ciple have been constructed by Marey and Cybulski, the former regis- tering the movements of the two columns of blood by connecting the tubes to tambours provided with writing levers, the latter by photo- graphy (Fig. 50). 5. The electrical method, described on p. 135, for the measurement of the circulation time, can also be applied to the estimation of the mean velocity of the blood between two cross-sections of the arterial path which are separated by a sufficient distance. For example, salt solution can be injected into the left ventricle or the be- ginning of the aorta, and the interval which it takes to reach a pair of electrodes in contact with, say, the femoral artery determined. Knowing the distance between the point of injection and the electrodes, we can then calculate the mean velocity. 6. In the calorimetnc method of measuring the quantity of blood which passes through such parts as the hands (or feet) in man, the flow is deduced from the quantity of heat given off by the part in a given time, and the difference be- tween the temperatures of the blood entering and leaving the part. The hands are immersed in a large bath of water (a few degrees below arterial blood temperature) for a sufficient time to permit any change of temperature of the parts due to the difference in temperature between them and the water to be established. The hands are then rapidly transferred to calorimeters previously filled with water at the same temperature as that of the bath. All the heat henceforth given off can be assumed to be due to the cooling of the blood passing through the hands, since the small amount of heat produced in the resting hands is negligible for this purpose. The temperature of the arterial blood at the wrist is taken as 0-5° C. below that of the rectum, this being the relation actually found in a normal man.* The tempera- ture of the venous blood leaving the hand is taken as that of the calorimeter, since it has been found that blood withdrawn from the hand veins by puncture, and collected with suitable precautions to prevent loss of heat as far as possible and to permit the calculation of the unavoidable loss, has a temperature only a negligible fraction of a degree above that of the bath in which the hand is im- mersed. The flow in grammes per minute is obtained from the formula Q= — rr — THV ~> where Q is the quantity of blood, H the number Wl \\- J. j S * The temperature of the arterial blood at the wrist was assumed to be the calorimeter temperature at which the calorimeter neither loses heat to the hand nor g-uns heat from it. If the heat production in the resting hand is negligible, this must correspond to the temperature of the entering blood. Fig. 49. — Chauveau's Dromograph. A, tube connected with blood- vessel; B, metal cylin- der in communica- tion with A. The upper end of B has a hole in the centre, which is covered by a mem- brane, m, through which a lever, C, passes; C has a small disc, p, at its end, which projects into the lumen of A, and is de- flected in the direction of the blood - stream through A. The de- flection is registered by a recording tambour in communication by the tube E with a tambour D, the flexible mem- brane of which is con- nected with the lever or pendulum C. MECHANICS OF THE CIRCULATION IN THE VESSELS 123 of small calories (gramme-calories) given off in m minutes, T the tem- perature of the blood entering the hand, T1 the temperature of the blood leaving the hand, and s the specific heat of blood (0-9). For purposes of comparison the volume of the hands is measured, and the blood-flow expressed in grammes per 100 c.c. of hand per minute. Further details are given in the Practical Exercises (p. 219). Fig. 51 shows one of the calorimeters on its adjustable stand. The collar of thick felt which fits closely around the wrist, and prevents loss of heat from the orifice through which the hand is inserted, is shown standing on the top of the calorimeter, as also the thermometer with the small sliding lens, or ' reader.' In Fig. 108, p. 220, the position of the subject with hands in the calorimeters is shown. Fig- 50- — Cybulski's Arrangement for recording Variation's in the Velocity of the Blood. A, tube connected with central, B with peripheral, end of divided bloodvessel. The blood stands higher in the tube C than in D. A beam of light passing through the meniscus in both tubes is focussed by the lens L on the travelling photographic plate E. The velocity at any moment is deduced from the height of the meniscus in the two tubes C and D. Fig. 51. — Calorimeter with stand for measuring blood-flow in hand. Of these methods, 3 and 4 are alone suited for the study of the velocity-pulse, that is, the change of velocity occurring with every beat of the heart. The curves obtanied by Chauveau's dromo- graph show a general agreement with blood-pressure tracings taken by a spring manometer, and with records of the external pulse obtained by a sphygmograph. There is a primary increase of velocity corresponding with the ventricular systole, and a secondary increase corresponding with the dicrotic wave (Fig. 54). Like all the other pulsatory phenomena, the velocity- pulse disappears in the capillaries, and is only present under exceptional circumstances in the veins. 1 24 THE CIRCULATION OF THE BLOOD AND LYMPH Fick, from a comparison of sphygmographic and plethysmographic tracings (p. 128), taken simultaneously from the radial artery and the hand, has demonstrated that in man the velocity-pulse exhibits Fig. 52. Fig. 53- Fig. 52. — The highest of the three curves is a plethysmographic record taken from the hand; the second curve is a sphygmogram taken simultaneously from the corresponding radial artery; the lowest (interrupted) curve is the curve of velocity deduced from a comparison of the first two (Fick). Fig- 53. — Simultaneous plethysmographic and sphygmographic tracings. the same general characters as in animals (Figs. 52 and 53). And v. Kries has confirmed Fick's conclusions by actual records .of the velocity-pulse obtained by means of an arrangement called a gas tachograph (Fig. 55). Fig. 54. — Simultaneous Tracings of the Velocity (Upper Curve) and Pressure (Lower Curve) (Lortet). The tracings were taken from the carotid artery of a horse. The curve of velocity was obtained by the dromograph. The dicrotic wavs is marked on it. The slightly curved ordinates drawn through the curves indicate corresponding points. This consists of a plethysmograph connected with the tube of a gas- burner. When the part enclosed in the plethysmograph expands, air issues from the connecting tube, and causes an increase in the height of the flame. When the part shrinks, air is drawn in from the flame, which is depressed. Since the speed of the blood in the veins may be considered constant during the time of an experiment, the rate at which MECHANICS OF THE' CIRCULATION IN THE VESSELS 125 the volume of the part alters at any moment is a measure of the pulsa- tory change of velocity in the arteries of the part. And by photo- graphing the movements of the flame on a travelling sensitive surface, the velocity-pulse is directly recorded. The mean velocity, like the mean blood-pressure, is more variable in the large arteries near the heart than in the smaller and more di stant arteries. Dogiel found in measurements taken with the stromuhr (a good instrument for the estimation of mean speed), within a period of two minutes, velocities ranging from over 200 mm. to under 100 mm. per second in the carotid of the rabbit, and from over 500 mm. to less than 250 mm. in the carotid of the dog. Chauveau, with the dromograph, found the velocity in the carotid of a horse to be 520 mm. per second during systole, 150 mm. during the pause, 220 mm. during the period of the dicrotic wave. It is probable, however, that, if these numbers are at all accurate for bloodvessels in the immediate neighbourhood of the heart, there must be a rapid diminution in the velocity even while the arteries are still of considerable calibre. For it has been found . by the electrical method that, in anaesthetized dogs at any rate, as is shown in the following table, the mean velocity between the origin of the aorta and the crural artery in the middle of the thigh is usually less than 100 mm. per second. Fig- 55- — Photographic Record of the Velocity-Pulse ob- tained by the Gas Tachograph (v. Kries). The upper curve is the photographic representation of the move- ments of the flame, and corresponds to the curve of velocity. No. of Experi- ment. Body, weight in Kilos. Distance between Point of Injection and Electrodes, in Millimetres. Average Time be- tween Injection and Arrival of the Salt Solution, in Seconds. Average Pulse-rate per Minute. Average Velocity per Second, in Milli- Average Distance traversed per Heart-beat, in Milli- I. 34'55 42O 4 '62 105 90-9 51-9 II. 17-5 495 57 69 86«8 75*4 III. 14-99 400 5'0 102 80-0 47-0 IV. 10-32 470 7-12 74-5 72-9 58-7 V. 7-165 330 7-83 46-3 42-1 54'5 (weak beat) i In I. the injecting cannula was in the descending part of the thoracic aorta, in V. at the very origin of the aorta, and in II., III., and IV. in the left ventricle. 126 THE CIRCULATION OF THE BLOOD AND LYMPH As to the speed of the blood in the arteries of man, our data are insufficient for more than a loose estimate. But it does not seem likely that the mean velocity of a particle of blood in moving from the heart to the femoral artery can exceed 150 mm. per second for the whole of its path. This would correspond to rather more than a third of a mile per hour. In the arch of the aorta the average speed may be twice as great. ' The rivers of the blood ' are, even at their fastest, no more rapid than a sluggish stream. A red cor- puscle, even if it continued to move with the velocity with which it set out through the aorta, would only cover about 15 miles in twenty- four hours, and would require five years to go round the world. The average flow through the hands of a healthy young man, as determined in eighteen experiments on different dates ranging over two years, at room temperatures varying from 19° to 27° C., was 12-8 grammes per 100 c.c. of hand per minute for the right hand, and 12-3 grammes for the left. Ten of the observations on this man are con- densed in the table. Blood-flow in Grammes Temperature of—- per 100 c.c. of Hand per Minute. Date. Room. ' Arterial Blood. Calorim. Right Left. November 30, 1910 - 20-2 36-6 28-0 I0'l 9'4 December 22, 1910 - 2I-I 36'8 29-7 I3'7 12-5 February I, 1911 22-8 36-9 30-3 12-6 12-7 March 17, 1911 2I-I 36-6 29-9 n-8 n-3 May 24, 1911 - 27-0 36-8 30-9 18-5 17-5 November 3, 1911 - 25-I 36-7 30-0 I2-I n-7 November 9, 1911 - 24-I 36-7 30-8 IS'? 12-5 November 15, 1911 - 25-2 36-8 30-8 I4'O 13-8 December n, 1911 - 24-5 36-6 3I-I 12-4 12 '0 March 26, 1913 24-5 36-6 31'5 14-7 I5'I Since the great function of the circulation in the skin is the regulation of the temperature of the body (see Chapter XII.), the blood-flow in the hands is, of course, much influenced by the external temperature. Thus, by far the greatest flow in the above table corresponds to the high room temperature of 27° C. With a given external temperature, the degree of humidity of the air also affects the flow. Under similar conditions of external temperature and daily routine, including diet, the hand flow in one and the same individual does not vary greatly when measured at about the same hour on different days. Different individuals, when tested under apparently similar conditions, show a greater range in the blood-flow. Some normal persons know and say that their hands are habitually cool or cold; others, like the man on whom the above results were obtained, that their hands are habitually warm. The former may be expected to show a relatively small, and the latter a relatively large, flow of blood through the hands. It is MECHANICS OF THE CIRCULATION IN THE VESSELS 127 possible that such habitual differences are associated with differences in the total heat value of the food consumed or in the proportions of the various food substances, especially of proteins (see p. 605), for it is well known that even persons engaged in the same work, and living under similar external conditions, may differ greatly in their dietetic habits, both as to quantity and quality of the food. And since the cutaneous circulation is by far the most important factor in the loss of heat from the body, the hearty eaters, other things being equal, may be expected to have the largest blood-flow through parts like the hands. The importance of the flow through the skin in the total hand flow is illustrated by the fact that the flow per unit of volume through the distal half of the hand, which, of course, has a large surface in proportion to its volume, is considerably greater than through the hand as a whole. For the forearm the flow per 100 c.c. is, in its turn, much less than in the hand (Hewlett) . In the foot the blood-flow, as estimated by the calorimetric method, is smaller per unit of volume of the part than in the hand, the ratio of foot flow to hand flow per 100 c.c. of the part usually lying in normal persons between i to 3 and i to 2. This is largely due to the proportionally greater proportion of skin in the hand, as well as to the smaller proportion of bone, which has not an active circulation. In the sitting position the following results were obtained for the flow in the feet on the person whose hand flows have been given above : Blood-flow in Grammes Temperature of— per 100 c.c. of Foot per Minute. Date. Room. Arterial Blood. Calorim.i Right. Left May 2, 1911 - 25-2 36-8 30-4 3'5 ^_ May 18, 1911 - • 26-4 36-9 31'4 3'5 — June 17,1911 - 21-8 37-o 30-6 4'9 4*2 March 26, 1913 24-5 36-5 31-4 3'9 4-1 The great variations in the vascularity of different organs and parts, as revealed by the examination of injected specimens or by inspection of the organs during life, indicate that there must be great differences in the blood-flow. Observations with the stromuhr in animals have shown that this is the case. The following list gives the number of c.c. of blood passing per minute through 100 grammes of organ, according to the results of Burton-Opitz, Tigerstedt, and other observers : Posterior extremity. « 5 Liver (venous) . 59 Skeletal muscle „ . 12 Liver (total) . . .84 Head .... 20 Brain . . ... . 136 Stomach .. 21 Kidney .. . 150 Liver (arterial) > 25 Adrenal . . . 500 Intestines . . 31 Thyroid gland . 560 Spleen ..» 58 The Volume-Pulse. — When the pulse-wave reaches a part it dis- tends its arteries, increases its volume, and gives" rise to what may be called the volume-pulse. 128 THE CIRCULATION OF THE BLOOD AND LYMPH This may be readily recorded by means of a pie thy s mo graph, an instrument consisting essentially of a chamber with rigid walls which enclose the organ, the intervening space being filled up with liquid (Fig. 56). The hiovements of the liquid are transmitted either through a tube filled with air to a recording tambour, or directly to a piston or float acting upon a writing lever. Special names have been given to plethysmographs adapted to particular organs; for example, Roy's oncometer for the kidney. The method has been successfully applied to the investigation of circulatory changes in man, a finger, a hand or an entire limb being enclosed in the plethysmo graph. With a fairly sensitive arrangement, every beat of the heart is represented on the tracing by a primary elevation and a dicrotic wave (Fig. 57). The general appearance of the curve is very similar to that of an ordinary pulse-tracing, though there are some differences of detail, especially in the time relations. A volume-pulse has been actually ob- served not only in limbs and por- tions of limbs, but also (in animals) in the spleen, kidney and brain and other organs, and in the orbit. Fig. 56. — Plethysmograph (Mosso). M, balanced test-tube, in communication with the glass vessel, D, which contains the arm, escape of water being prevented by the rubber cuff, A . When water passes from vessel D to M , or from M to D, M moves down or up, and its movements are recorded by the writing-point .V. M is steadied by the liquid in P, into which it dips. The so-called cardio-pneumatic movements also constitute a volume-pulse, although of complex origin. This name is given to the rhythmical changes of pressure accompanying the beat of the heart, which can be detected in the air of the respiratory passages when one nostril is connected with a recording tambour, or water manometer, the other nostril and the mouth being closed, and the respiration suspended in inspiration, with the glottis open. Or the MECHANICS OF THE CIRCULATION IN THE VESSELS 129 'mouth may be. connected with the recording apparatus, the nostrils being closed. One factor in the production of these movements may be the change of blood- volume in the soft tissues of the mouth, naso- pharynx, and perhaps also in the lower respiratory passages accom- panying the heart-beat. Another factor, and a more influential one, is the rhythmical alteration of pressure caused directly by the alter- nate systole and diastole of the heart in the air contained in the lung-tissue surrounding it, which acts as a kind of air plethysmo- graph. One interesting way in which the cardio-pneumatic move- ments may reveal themselves is by a variation with each beat of the heart in the intensity of a note prolonged in singing, especially after fatigue has set in. Upon the whole, the air-pressure falls during systole, owing to the expulsion of blood from the chest, and rises during diastole. The main cardio-pneumatic movement is, there- fore, a systolic inspiration and a diastolic expiration (Practical Exercises, p. 303). Doubtless the weight of an organ would also show a pulse correspond- ing to the beat of the heart, and so would the temperature^-at least, of the superficial parts. For the amount of heat given off by the blood Fig- 57- — Plethysmograph Tracing from Arm. The tracing was taken by means 01 a tambour connected with the plethysmograph. The dicrotic wave is distinctly marked. to the skin increases with its mean velocity, and, therefore, although the difference may not in general be measurable, more heat is pre- sumably given off during the systolic increase of velocity than during the diastolic slackening. And this, along with other considerations, suggests that, at any rate in certain situations and under certain con- ditions, there may even be a pulse of chemical change ; that is, a slight and as yet doubtless inappreciable ebb and flow of metabolism corre- sponding to the rhythm of the heart. The Circulation in the Capillaries. — From the arteries the blood passes into a network of narrow and thin- walled vessels, the capil- laries, which in their turn are connected with the finest rootlets of the veins. Physiologically, the arterioles and venules must for many purposes be included in the capillary tract, but the great anatomical difference — the presence of circularly-arranged muscular fibres in the arterioles, their absence in the capillaries — has its 9 130 physiological correlative. The calibre of the arterioles can be altered by contraction of these fibres under nervous influences; the calibre of the capillaries, although it varies passively with the blood- pressure, and is possibly to some extent affected by active con- traction of the endothelial cells, cannot be under the control of vaso- motor nerves acting on muscular fibres (but see p. 173). Harvey had deduced from his observations the existence of channels between the arteries and the veins. Malpighi was the first to observe the capillary blood-stream with the microscope, and thus to give ocular demonstration of the truth of Harvey's brilliant reasoning. He used the lungs, mesentery and bladder of the frog. The web of the frog, the tail of the tadpole, the wing of the bat, the mesentery of the rabbit and rat, and other transparent parts, have also been frequently employed for such investigations. From the apparent velocity of the corpuscles and the degree of magnification, Fig- 58- — Diagram to illustrate the Slope of Pressure along the Vascular System. A, arterial; C, capillary; V, venous tract. The interrupted line represents the line of mean pressure in the arteries, the wavy line indicating that the pressure varies with each heart-beat. The line passes below the abscissa axis (line of zero or atmospheric pressure) in the veins, indicating that at the end of the venous system the pressure becomes negative. it is easy to calculate the velocity of the capillary blood-stream. It has been estimated at from 0-2 to 0-8 mm. per second in different parts and different animals. The comparative slowness of the current and the disappearance of the pulse are the chief characteristics of the capillary circulation. The explanation we have already found in the great resistance of the narrow arterioles and the much-branched capillary vessels. Although the average diameter of a capillary is only about 10 u, (5 to 20 ft in different parts of the body), the number of branches is so prodigious that the total cross-section of the systemic capillary tract has been estimated at 500 to 700 times that of the aorta. Such estimates are, of course, by no means exact. The total cross-section of the vascular channel gradually widens as it passes away from the left ventricle. In the capillary region it undergoes a great and sudden increase. A part of this increase is to be attributed to the arterioles, which, although individually very narrow, have a total bed considerably greater than that of the arteries from which they spring. Where the arterioles pass into the capillaries proper, a further and a still greater and more abrupt increase in the bed occurs. At the venous end of this region the cross-section is again somewhat abruptly contracted, and then gradually lessens as the right side of the heart is approached; but the united sectional area of the large thoracic veins is greater than that of the aorta. Attempts have been made to measure the blood-pressure in the capillaries by weighting a small plate of glass laid on the back of one of the fingers behind the nail, until the capillaries are just emptied, as shown by the paling of the skin (v. Kries), or by observing the height of a column of liquid that just stops the circula- tion in a transparent part (Roy and Graham Brown). The last- named observers found that a pressure of 100 to 150 mm. of water (about 7 to ii mm. of Hg) was needed to bring the blood to a standstill in the capil- laries and veins of the frog's web; that is, about a third of the blood-pressure in the frog's aorta. The pres- sure in the capillaries at the root of the nail Fig. 59. — Relation of Blood- Pressure, Velocity, and Cross-Section. The curves P, V, and S represent the blood-pressure, velocity of the blood, and total cross- section respectively in the arteries A, capillaries C, and veins V. in man varies from 30 to 50 mm. of mercury, as estimated by the method of v. Kries. But the method is exposed to serious errors. The method of measuring the venous pressure described on p. 132 can also be applied to the capillaries, and is somewhat more satisfactory. Under certain conditions the pulse-wave may pass into the capillaries and appear beyond them as a venous pulse. Thus, we shall see that when the small arteries of the submaxillary gland are widened, and the vascular resistance lessened, by the stimulation of the chorda tympani nerve, the pulse passes through to the veins. And, normally, a pulse may be seen in the wide capillaries of the nail-bed — especially when they are partially emptied by pressure — as a flicker of pink that comes and goes with every beat of the heart. We have seen that the lateral pressure at any point of a uniform rigid tube through which water is flowing is proportional to the amount of resistance in the portion of the tube between this point and the outlet. In any system of tubes the sum of the potential and kinetic energy must diminish in the direction of the flow; and although the problem 132 THE CIRCULATION OF THE BLOOD AND LYMPH is complicated in the vascular system by the branching of the channel and the variation in the total cross-section, yet theory and experiment agree that in the larger arteries the lateral pressure diminishes but slowly from the heart to the periphery, the resistance being small com- pared with the resistance of the whole circuit. In the capillary region the vascular resistance abruptly increases; the velocity (and therefore the kinetic energy) abruptly diminishes, and the lateral pressure falls much more steeply between the beginning and the end of this region than between the heart and its commencement. In the veins only a small remnant of resistance remains to be overcome, and the lateral pressure must sink again rather suddenly about the end of the capillary tract. Fig. 59 shows by a rough diagram the manner in which the pressure, velocity and cross-section probably change from part to part of the vascular system. The Circulation in the Veins. — The slope of pressure, as we have just explained, must fall rather suddenly near the beginning and near the end of the capillary tract. It continues falling as we pass along the veins, till the heart is again reached. In the right heart, and in the thoracic portions of the great veins which enter it, the pressure may be negative — that is, less than the atmospheric pressure. And since nowhere in the venous system is the pressure more than a small fraction of that in the arteries, its measurement in the veins is correspondingly difficult, because any obstruction to the normal flow is apt to artificially raise the pressure. A man- ometer containing some lighter liquid than mercury, such as water or a solution of sodium citrate or magnesium sulphate, is usually employed, so that the difference of level may be as great as possible. In the sheep the pressure was found to be 3 mm. of mercury in the brachial, and about u mm. in the crural vein. Burton-Opitz obtained the following pressures in dogs (of about 15 kilos): left facial vein, 5-1 ; right external jugular, - o-ii ; central end of superior vena cava, — 2'8; femoral vein, 5-4; renal vein, 10-9; portal vein, 8' 9 mm. of mercury. Estimation of Venous Pressure in Man. — The venous pressure in man has been estimated by several observers with more or less satisfactory results. The best-known method is that of v. Recklinghausen. A circular rubber bag, with a central opening, is laid over the course of a vein, so that the vein can be observed through the opening, as in Fig. 60. The bag is smeared with glycerin. A glass plate is laid over the opening and held firmly, so that the vein and the surrounding skin are in a closed chamber. The bag is provided with a side-tube, which connects it with a pump and a water manometer. By means of the pump air is forced into the bag till the vein is just seen to collapse. The pressure indicated on the manometer at this moment is taken as the pressure in the vein. By means of a modification in this method, Eyster and Hooker have found that the pressure in the small veins of the arm or hand generally varies between 3 and 10 cm. of water. In conditions of congestion of the venous system the pressure may rise to 20 cm. of water (say 15 mm. of mercury) or more. In making this measurement it is necessary to take account of the MECHANICS OF THE CIRCULATION IN THE VESSELS 133 position of the vein, since the hydrostatic factor (p. igo) in the venous pressure is so important. Thus it is obvious that the pressure in the veins of the hand will be greater when it is hanging down than when it is raised to level of the heart or above it. Accordingly, the actual readings of the manometer must always be corrected for the vertical distance between the vein and the heart, the' height of a column of blood equal to this distance being deducted from or added to the manometer reading, according to whether the vein is below or above the heart level. For practical purposes the heart level is supposed to correspond to the lower end of the sternum (costal angle). For the measurement of the pressure in the right auricle, the follow- ing simple and elegant method has been given by Gaertner, following a suggestion of Frey: He raises the arm of the sitting patient, and observes a small vein on the back of the hand. At the moment when the vein collapses the elevation of the arm is stopped, and the vertical distance between the vein and the heart measured. This expressed in millimetres of blood (i.e., approximately of water) is the pressure in the auricle, since the veins of the arm constitute manometer tubes connected with the auricle. The venous pressure being so low, or, in other words, the potential energy which the systole of the heart imparts to the blood being so greatly exhausted /t before it reaches the veins, other in- fluences begin here appreciably to affect the blood- stream : 1. Contraction of the Muscles. — This compresess the neighbouring veins, and since the blood is compelled by the valves, if it moves at all, to move towards the heart, the venous circulation is in this way helped. 2. Aspiration of the Thorax. — In inspiration the intrathoracic pressure, and therefore the pressure in the great thoracic veins, is diminished, and blood is drawn from the more peripheral parts of the venous system into the right heart (p. 226). 3. Aspiration of the Heart. — When the heart after its contraction suddenly relaxes, the endocardiac pressure becomes negative, and blood is sucked into it, just as when the indiarubber ball of a syringe is compressed and then allowed to expand. But we cannot attribute any great importance to this; and, of course, it is only the relaxa- tion of the right ventricle which could directly affect the venous circulation. 4. Every change of position of the limbs, as in walking, aids the venous circulation (Braune), and this independently of the muscular contraction. When the thigh of a dead body is rotated outwards, and Fig. 60. — Diagram of Measurement of Venous Pressure (v. Recklinghausen). H, back of hand, with V, a vein; B, the rubber bag with central opening; T, tube leading from bag to manometer and pump ; G, glass plate. 134 THE CIRCULATION OF THE BLOOD AND LYMPH at the same time extended, a manometer connected with the femoral vein shows a negative pressure of 5 to 10 mm. of water. When the opposite movements are made, the pressure becomes positive. It follows from the number of casually-acting influences which affect the blood-flow in the veins that it cannot be very regular or constant. We have seen that in the great arteries there is a con- siderable variation of velocity and of pressure with every discharge of the ventricle, and although this variation is absent from the veins, since normally the pulse, due to the ventricular discharge, does not penetrate into them, the venous flow is, nevertheless, as a matter of fact, more irregular than the arterial. So that if it is difficult to give a useful definition of the term ' velocity of the blood ' in the case of the arteries, it is still more difficult to do so in the case of the veins. Where voluntary movement is prevented, one potent cause of variation in the venous flow is eliminated; and in curarized animals certain observers have found but little differ- ence between the mean velocity in the veins and in the corre- sponding arteries. Others have found the velocity in the veins considerably less, which is indeed what we should expect from the fact that the average cross-section of the venous system is greater than that of the arterial system. Burton-Opitz, by means of a stromuhr, obtained a mean velocity of 147 mm. per second in the external jugular vein of a 13-kilo dog. To sum up, we may conclude that, upon the whole, the blood passes with gradually-diminishing velocity from the left ventricle along the arteries; it is greatly and somewhat suddenly slowed in the broad and branching capillary bed; but the stream gathers force again as it becomes more and more narrowed in the venous channel, although it never acquires the speed which it has in the aorta. Venous Pulse. — To complete the account of the circulation in the veins, it may be recalled that, in addition to the venous pulse described on p. 131, which, as an occasional phenomenon, may travel through widened arterioles and capillaries from the arteries into the veins, and therefore in the direction of the blood-stream, a so-called venous pulse, travelling from the heart against the blood- stream and depending on variations of pressure in the right auricle, may be detected in the jugular veins in healthy persons, and more distinctly in certain disorders of the circulation, where indeed it may be evident at a greater distance from the heart — for example, over the liver as the so-called liver pulse. In animals a venous pulse of this nature has been demonstrated in the venae cavae, the jugular vein, and with a delicate manometer even in the large veins of the limbs. It moves with a speed of I to 3 metres a second (Morrow). It is most easily observed in the jugular veins in man, because of their proximity to the heart. We have already pointed out the significance of the study of this venous pulse for the analysis MECHANICS OF THE CIRCULATION IN THE VESSELS 135 of cardiac events (p. 100). A jugular venous pulse of a perfectly different origin is seen in cases of incompetence of the tricuspid valve. Here the chief elevation is synchronous with the ventricular systole, and is caused by the regurgitation of blood from the right ventricle through the auricle into the veins. The so-called ' com- municated venous pulse ' is simply due to the proximity of some large artery, especially when enclosed in a common sheath, whose pulsations are directly transmitted to the vein. The changes of pressure in the great veins accompanying the respiratory move- ments (p. 290) are also sometimes spoken of as a venous pulse, but they are produced in a different way — namely, by the rhythmical alteration in the intrathoracic pressure, which alternately favours and hinders the venous return to the heart. The Circulation-Time. — Hering was the first who attempted to measure the time required by the blood, or by a blood-corpuscle, to complete the circuit of the vascular system. He injected a solution of potassium ferrocyanide into a vein (generally the jugular), and collected blood at intervals from the corresponding vein of the opposite side. After the blood had clotted, he tested for the ferrocyanide by addition of ferric chloride to the serum. The first of the samples that gave the Prussian blue reaction corresponded to the time when the injected salt had just completed the circulation. This method was improved by Vierordt, who arranged a number of cups on a revolving disc below the vein from which the blood was to be taken. In these cups samples of the blood were received, and the rate of rotation of the disc being known, it was possible to measure the interval between the injection and appear- ance of the salt with considerable accuracy. Hermann made a further advance by allowing the blood to play upon a revolving drum covered with paper soaked in ferric chloride, and by using the less poisonous sodium ferrocyanide for injection. Even as thus modified, the method laboured under serious defects. It was not possible to make more than a single observation on one animal, at least without allowing a considerable interval for the elimina- tion of the ferrocyanide, and, further, the method was unsuited for the estimation of the circulation-time in individual organs. In both of these respects the more recently introduced electrical method presents considerable advantages; for by its aid we can not only obtain satis- factory measurements of the circulation-time in such organs as the lungs, liver, kidney, etc., but we can repeat them an indefinite number of times on the same animal. A cannula, connected with a burette (or a Mariotte's bottle, or a syringe), containing a solution of sodium chloride (usually a 1*5 to 2 per cent, solution), is tied into a vessel — say the jugular vein. Sup- pose that the time of the circulation from the jugular to the carotid is required — that is, practically the time of the lesser or pulmonary circu- lation. A small portion of one carotid artery is isolated, and laid on a pair of hook-shaped platinum electrodes,* covered, except on the concave side of the hook, with a layer of insulating varnish. To further secure insulation, a bit of very thin sheet -indiarubber is slipped between the artery and the tissues. By means of the electrodes the * The electrodes can easily be made by beating out one end of a piece of thick platinum wire to a breadth of 5 or 6 mm., and then bending the flattened part into a hook, or by bending pieces of stout platinum foil, 136 THE CIRCULATION OF THE BLOOD AND LYMPH piece of artery lying between them, with the blood that flows in it, is connected up as one of the resistances in a Wheatstone's bridge (p. 726). The secondary coil of a small inductorium, arranged for giving an inter- rupted current, and with a single Daniell or dry cell in its primary, is substituted for the battery, and a telephone for the galvanometer, according to Kohlrausch's well-known method for the measurement of the resistance of electrolytes. It is well to have the induction machine set up in a separate room and connected to the resistance-box by long wires, so that the noise of the Neefs hammer may be inaudible. The bridge is balanced by adjusting the resistances until the sound heard in the telephone is at its minimum intensity, the secondary coil being placed at such a distance from the primary that there is no sign of stimulation of muscles or nerves in the neighbourhood of the electrodes Fig. 61. — Measurement of the Pulmonary Circulation -Time in Rabbit by Injection of Methylene Blue. when the current is closed. A definite, small quantity of the salt solu- tion is now allowed to run into the vein by turning the stop-cock of the burette. It moves on with the velocity of the blood, and reaching the artery on the electrodes causes a diminution of its electrical resistance (p. 26). This disturbs the balance of the bridge, and the sound in the telephone becomes louder. The time from the beginning of the injec- tion to the alteration in the sound is the circulation - time between jugular and carotid. It can be read off by a stop-watch, or more accurately by an electric time-maker writing on a revolving drum (Fig. 62). Instead of the telephone a galvanometer may be used, the equal and oppositely directed induction shocks being replaced by a weak voltaic current, and the platinum by unpolarizable electrodes (p. 731). But this is less convenient. 137 The circulation-time of an organ like the kidney can be measured by adjusting a pair of electrodes under the renal artery and another under the renal vein, and reading off the interval required by the salt solution to pass from the point of injection first to the artery and then to the vein. The difference is the circulation -time through the kidney. For certain purposes, and particularly for measurements on small animals like the rabbit, or on organs whose vessels are too delicate to be placed on electrodes without the risk of serious interference with the circulation, another method may be employed with advantage. It depends on the injection of a pigment, like methylene blue, which at first overpowers the colour of the blood and shows through the walls of the bloodvessels, but is soon reduced to a colourless substance (Fig. 61). The details of the method are given in the Practical Exercises (p. 217.) It may be said in general terms that in one and the same animal the time of the lesser circulation is short as compared with the total circulation-time, relatively constant, and but little affected by changes of temperature. In animals of the same species it increases with the size, but more slowly, and rather in proportion to the increase of surface than to the increase of weight. Thus a dog weighing 2 kilogrammes had an average pulmonary circulation-time of 4*05 seconds, while that of a dog weighing ir8 kilos was 8*7 seconds, and that of a dog with a body-weight of 18*2 kilos only ip'4 seconds. It is probable that in a man the pulmonary circulation- time is not usually much less than 12 seconds, nor much more than 15 seconds. The circulation-time in the kidney, spleen and liver is relatively long and much more variable than that of the lungs, being easily affected by exposure and changes of temperature (increased by cold, diminished by warmth). In a dog of 13-3 kilos weight the average circulation-time of the spleen was 10-95 seconds; kidney, 13-3 seconds; lungs, 8-4 seconds. The circulation-time of the stomach and intestines is (in the rabbit) comparatively short, not exceeding very greatly that of the lungs, but it is lengthened by exposure. The circulation-time of the retina and that of the heart (coronary circulation) are the shortest of all. The total circulation-time is properly the time required for the whole of the blood to complete the round of the pulmonary and systemic circulation. But there are many routes open to any given particle of blood in making its systemic circuit. If it passes from the aorta through the coronary circulation it takes an exceedingly short route. If it passes through the intestines and liver, or through the kidney, or through the lower limbs, it takes a long route. So that to determine the total cir- culation-time by direct measurement we must know (i) the quantity of blood that passes on the average by each path in a given time, and (2) the average circulation-time of each path. If the average weight of blood in each organ be represented by w^, wz, w3, etc. ; and the average circulation-times by tlt tz, t3, etc. ; and t be the total systemic circulation. time ; then WJ-T , w^. , w^. , etc., will represent the quantity of blood passing through each organ in t seconds, since in the average circula- THE CIRCULATION OF THE BLOOD AND LYMPH tion-time of an organ the whole of the blood in it at the beginning of the period of observation will have been exchanged for fresh blood. But the whole of the blood in the body, which we may call W, passes once round the systemic circulation in t seconds. Therefore, w^-r + a>^7 +to3— , etc., =W. In this equation everything can be de. *1 '2 *3 termined by experiment except t, and therefore t can be calculated. Adding t to the pulmonary circulation -time, we arrive at the tota circulation -tune . Although our experimental data are as yet too meagre to make the calculation more than a rough approximation, it appears probable that in certain animals the total circulation-time is five or six times as great as the pulmonary circulation -time. If the same ratio holds good in man, the total circulation-time is unlikely to be much less than a minute . or much greater than a _ | 1 . minute and a quarter. We shall see directly that this estimate is confirmed by data de- rived from a different source. In the mean- time, we may use it II m provisionally to calcu- late the work done by the heart. Let us take for simplicity the total circulation - time as i minute in a 70 -kilo man, the quantity of blood as 4 £ kilos, * and the mean pressure in the aorta as 150 nun. of mercury. Up to the time when the semi- lunar valves are opened , the work done by the left ventricle is spent in raising the intra ven- tricular pressure till it is sufficient to over- come the pressure in the aorta. If a vertical tube were connected with the left ventricle, the blood would rise till the column was of the same weight as a column of mercury of equal section and 150 mm. high. This column of blood would be about 1-92 metres in height. If a reservoir were placed in communication with the tube at this height, a quantity of blood equal to that ejected from the ventricle would at each systole pass into the reservoir; and the work which the blood thus collected would be capable of doing, if it were allowed to fall to the level of the heart, v/ould be equal to the work expended by the heart in forcing it up. Thus, in i minute the work of the left ven- tricle would be equal to that done in raising 4^ kilos of blood to a height * The mean of the 5^ kilos given by most writers, and of the 3^ kilos ob- tained by Haldane and Smith (p. 56). _JLJL.fl .0-JLJLJLJlJLJl-n-JLJLJ B HJLJLJl Fig. 62. — Time of the Lesser Circulation. Cat anaes- thetized with Ether. Time-trace, seconds. The line above the time-trace was written by an electro- magnetic signal, the circuit of which was closed at the moment when injection of methylene blue into the jugular vein was begun, and opened at the moment when the change of colour in the carotid was observed. I, normal circulation-time; II, cir- culation-time after section of both vagi (much diminished) ; III, circulation-time during stimulation of the peripheral end of one vagus (much increased). MECHANICS OF THE CIRCULATION IN THE VESSELS 139 of i'92 metres — that is, about 8^64 kilogramme-metres; in 24 hours it would be, say, 12,450 kilogramme-metres. Taking the mean pressure in the pulmonary artery at one-third of the aortic pressure, we get for the daily work of the right ventricle about 4,150 kilogramme -metres. The work of the two ventricles is thus about 16,600 kilogramme-metres,* which is enough to raise a weight of nearly 4 pounds from the bottom of the deepest mine in the world to the top of its highest mountain, or to raise the man himself to ij times the height of the spire of Strasburg Cathedral, or twice the height of the loftiest ' skyscraper ' in New York. By friction in the bloodvessels this work is almost all changed into its equivalent of heat, nearly 40,000 gramme-calories (p. 688) . Further, since the contraction of the heart is always maximal (p. 154), and there is reason to believe that the quantity of blood ejected at a single systole by the left ventricle (being dependent upon the inflow from the pulmon- ary veins, and therefore upon the inflow into the right side of the heart from the systemic veins) varies widely, some of the mechanical effect of the contraction must be wasted when the quantity is less than the ventricle is capable of expelling. Output of the Heart. — If 4$ kilos of blood pass through the heart in i minute with the average pulse-rate of 72 per minute, the quantity ejected by either ventricle with every systole will be — — =62-5 grm., or a little less than 60 c.c. The output may be expressed in grammes or cubic centimetres per minute (the minute volume), or per second, or per beat. It has been measured in animals in several ways — e.g., by inserting a stromuhr (p. 121) on the course of the aorta, or by recording the variations in the volume of the heart, or, better, of the ventricles, by means of a plethysmograph (cardiometer of Henderson), in which the organ is enclosed. Another method, which does not entail the opening of the chest, is to allow a salt solution to run slowly, for a de- finite number of seconds, into the left ventricle through a tube passed into it from the carotid artery. A sample of the mixture of blood and salt solution is collected from a branch of the femoral artery, where its arrival is detected by the change of electrical resistance (p. 135). From the amount of salt solution which must be added to a normal sample of blood drawn before the injection to make its conductivity the same as that of the sample taken during the passage of the mixture, the quantity of blood with which the solution was mixed in the ventricle during the injection can be approximately determined. By this method it has been shown in a series of experiments on more than twenty dogs, ranging in weight from 5 to nearly 35 kilos, that the output of the left ventricle per kilo of body-weight per second diminishes as the size of the animal increases ; and the relation between body-weight and out- put is such that in a man weighing 70 kilos we can hardly suppose that the ventricle discharges, during bodily rest, more than 105 grm. of blood per second, or 87 grm. (80 c.c.) per heart-beat with a pulse-rate of 72. Putting this result along with that deduced from the circulation- time, we can pretty safely conclude that the average amount of blood thrown out by each ventricle at each beat is not more than 70 or 80 c.c. Zuntz, from the quantity of oxygen absorbed by the blood in the lungs in a definite short time, and the difference between the oxygen content of samples of the arterial and venous blood, has estimated the output per beat at 60 c.c. But according to him this may be doubled during * Since the blood on expulsion is moving with a certain velocity, an addi- tion might be made for its kinetic energy. But this would only increase the total work by a small fraction (about i per cent.). 140 THE CIRCULATION OF THE BLOOD AND LYMPH severe muscular work, when, as a matter of fact, by the aid of the Rontgen-rays or by percussion of the chest, the volume of the heart may be shown to be considerably increased. Tigerstedt, on the basis of stromuhr measurements in animals, puts the ventricular output per beat in man at 50 to 100 c.c. ; Plesch, on the basis of gasometric ob- servations on man, at 59 c.c. Recently Krogh, using a gasometric method based on the absorption of nitrous oxide gas in the lungs, found that the minute volume during rest may vary between wide limits (2-8 to 8'7 litres of blood per minute, corresponding, with a pulse-rate of 70, to 40 c.c. to 120 c.c. per beat). During muscular work there is a great and immediate Increase, up to, it may be, 21-6 litres per minute. These great variations in the output of the ventricle depend primarily upon variations in the rate of return of the blood to the heart by the veins. According to Henderson, however, such great variations in the output per beat as are postulated by the majority of physiologists who have worked at the subject do not occur, and the fundamental variable is the rate of the beat. In healthy persons in whom the pulse-rate is permanently much below the normal (p. 107) the output of the ventricle per beat must, of course, be correspondingly increased. In a man with a pulse -rate always below 40 during rest in the sitting position, the flow in the hands was found to be normal in amount, and all the signs of a normal delivery of blood from the left ventricle were present. Here the output per beat must have been twice the usual amount during rest. SECTION IV. — THE HEART-BEAT IN ITS PHYSIOLOGICAL RELATIONS. So far we have been considering the circulation as a purely physical problem. We have spoken of the action of the heart as that of a force-pump, and perhaps to a small extent that of a suction- pump too. We have spoken of the bloodvessels as a system of more or less elastic tubes through which the blood is propelled. We have spoken of the resistance which the blood experiences and the pressure which it exerts in this system of tubes, and we have considered the causes of this resistance, the interpretation of this pressure, and the physical changes in the vascular system that may lead to variations of both. But so far we have not at all, or only incidentally and very briefly, dealt with the physiological mechanism through which the physical changes are brought about. We have now to see that, although the heart is a pump, it is a living pump ; that, although the vascular system is an arrangement of tubes, these tubes are alive; and that both heart and vessels are kept constantly in the most delicate poise and balance by impulses passing from the central nervous system along the nerves. In many respects, and notably as regards the influence of nerves on it, we may look upon the heart as an expanded, thickened and rhythmically-contractile bloodvessel, so that an account of its innervation may fitly precede the description of vaso-motor action in general. THE HEART-BEAT IN ITS PHYSIOLOGICAL RELATIONS 141 The Relation of the Heart to the Nervous System. — A very simple experiment is sufficient to prove that the beat of the heart does not depend on its connection with the central nervous system, for an excised frog's heart may, under favourable conditions, of which the most important are a moderately low temperature, the presence of oxygen, and the prevention of evaporation, continue to beat for days. The mammalian heart also, after removal from the body, beats for a time, and indeed, if defibrinated blood be artificially circulated through the coronary vessels, for several or even many hours. But although this proves that the heart can beat when separated from the central nervous system, it does not prove that nervous influence is not essential to its action, for in the cardiac substance nervous elements, both cells and fibres, are to be found. The Intrinsic Nerves of the Heart. — In the heart of the frog numerous nerve-cells occur in the sinus venosus, especially near its junction with the right auricle (Remak's ganglion). A branch from each vagus, or rather from each vago-sympathetic nerve (for in the frog the vagus is joined a little below its exit from the skull by the sympathetic), enters the heart along the superior vena cava (pp. 157, 198). Running through the sinus, with whose ganglion-cells the true vagus fibres, or some of them, are believed to make physiological junction (p. 163), the nerves pursue their course to the auricular septum. Here they form an intricate plexus, studded with ganglion-cells. From the plexus nerve-fibres issue in two main bundles, which pass down the anterior and posterior borders of the septum to end in two clumps of nerve-cells (Bidder's ganglia), situated at the auriculo-ventricular groove. These ganglia in turn give off fine nerve-bundles to the ven- tricle, which form three plexuses — one under the pericardium, another under the endocardium, and a third in the muscular wall itself, or myo- cardium. From the last of these plexuses numerous non-medullated fibres run in among the muscular fibres and end in close relation with them. Similar plexuses of nerve-fibres exist in the mammalian ventricle. But while scattered ganglion-cells are found in the upper part of the ventricular wall, most observers have been unable to demonstrate any either in the mammal or the frog in the apical half. In the rat's heart, according to the careful observations of Schwartz, true ganglion-cells are confined to an area on the posterior surface of the auricles, lying always under the visceral pericardium. Other writers, however, have stated that ganglion-cells do exist in the apex both of the cat's and of the frog's heart. In connection with the whole question it must be borne in mind that in other organs improved histological methods have brought typical nerve-cells to light in situations where they were not suspected or were denied to exist, and. further, that all investigators are not agreed upon the histological criteria by which ganglion-cells are to be distinguished. Cause of the Rhythmical Beat of the Heart. — Scarcely any physio- logical question has excited greater interest for many "years than the mechanism of the heart-beat. Several properties of the cardiac tissue ought to be distinguished in discussing this question: (i) Its I42 THE CIRCULATION OF THE BLOOD AND LYMPH automatism — i.e., its power of beating in the absence of external stimuli; (2) its rhythmicity — i.e., its power of responding to con- tinuous stimulation by a series of rhythmically repeated contrac- tions; (3) its conductivity — i.e., its power of conducting the contrac- tion wave or the impulse to contraction once it has been set up ; and (4) the power of co-ordination, in virtue of which the various parts of the heart beat in a regular sequence. The excitability of the cardiac tissue — that is, its power of appro- priate response (namely, by contraction) to a suitable stimulus — does not particularly concern us here, since it is in no wise a property special to the heart. Only, as we shall see in the sequel, the time- relations of this excitability are of interest, for the existence of a refractory period — that is, an interval during which the cardiac muscle refuses to respond to excitation — throws light upon the rhythmicity of the heart-beat. The tonicity of the heart — i.e., its power of remaining contracted to a certain extent in the intervals between successive beats — is another property of great importance in certain aspects, but which only needs to be mentioned at present. Automatism of the Heart-Beat — Neurogenic and Myogenic Hypo- theses.— That the heart-beat is automatic is sufficiently shown by the fact that, as already mentioned, an excised and empty heart will go on beating for a time, for many hours or even for days in the case of cold-blooded animals. When blood, or even a suitable solution of such inorganic salts as exist in serum, is caused to circu- late through the coronary vessels of the excised heart of a warm- blooded animal, it also continues to contract for a long time. In trying to understand the real significance of the automatic beat of the heart, physiologists have endeavoured, first, to compare different portions of the heart as regards the degree in which they possess this property of automaticity ; and, second, to associate, if possible, one or other of the active tissues that compose the organ, muscle, and nervous tissue, with this characteristic property. It cannot be pretended that a final answer to this question is possible at present. Nor is the historical controversy which it has occasioned perhaps as important in itself as the space usually devoted to it in textbooks might imply. Yet it is probable that the series of fundamental facts in the physiology of the heart elicited in the long discussion can be best presented, even for the purposes of the elementary student, as they were originally brought forward in the form of pros and cons, of arguments for and against the neurogenic or the myogenic hypothesis. There is good evidence that as in the amphibian heart the contraction starts in the sinus venosus, so in the mam- malian heart it starts in the sinus tissue of the right auricle in the region of the sino-auricular node. Attempts have been made to demonstrate that the origination of the impulses which are after- wards conducted to all parts of the heart is normally confined to THE HEART-BEAT IN ITS PHYSIOLOGICAL RELATIONS 143 the node itself, and the sino-auricular node is by some authors denominated the pace-maker of the heart, the tissue which sets the pace for the rest of the organ and gives the time to auricles and ventricles alike. The experimental results, however, are by no means harmonious, some observers finding that destruction of the region of the node causes no change in the rate of the heart-beat, others that the beat is permanently slowed. But even were we in a position to sharply delimit a given region of the heart as the point at which the strong tendency to contraction inherent in the cardiac tissue as a whole first breaks into an actual beat, this would scarcely enable us to decide offhand where the cause of the automatism resides, in the muscular tissue or in the intrinsic nervous apparatus, because in nearly all animals hitherto investigated the muscular tissue, ganglion-cells, and nerve-fibres are inseparably intermingled. In Limulus, however, the horseshoe or king crab, the cardiac ganglion-cells are collected in a nerve-cord running longitudinally in the median line along the dorsal surface of the segmented heart, and Fig. 63. — The Heart ana che Heart Nerves of Limulus: Dorsal View (Carlson). (The heart is figured one-half the natural size of a large specimen.) aa, Anterior artery; la, lateral arteries; In, lateral nerves; mnc, median nerve-cord; os, ostia. sending off at intervals branches to two lateral cords, and also branches which enter the heart muscle directly (Fig. 63). When the median nerve-cord is§me way a portion of the heart, such as the apex of the ventricle, when stimu- lated in the quiescent condition by an interrupted current, responds by a rhythmical series of beats, and not by a tetanus. It is evident that the cardiac muscle, like ordinary striped muscle, is for some time after excitation incapable of responding to a fresh stimulus — i.e., there is a relractory period^ But this is immensely longer in cardiac than in skeletal muscle. When the phenomenon is analyzed, it is found that a stimulus falling into the heart muscle between the moment at which the contraction begins and the moment at which it reaches its maximum produces no effect — is, so to speak, ignored. When the stimulus is thrown in at any point between the maximum of the systole and the beginning of the next contraction, it causes what is called an extra contraction. The extra contraction is followed by a longer pause than usual — a so-called compensatory pause — which just restores the rhythm, so that the succeeding systole falls in the curve where it would have fallen had there been no extra contraction (Fig. 68). In man, extra systoles followed by compensatory pauses may occur under pathological conditions, giving rise to an important group of cardiac irregularities. These extra systoles may be either auricular or ventricular, the auricle or the ventricle contracting prematurely without waiting for the signal of the sinus rhythm. The analysis of pulse- tracings showing these irregularities has led to results of great physio- logical and clinical interest (Cushny, Mackenzie, etc.), but cannot be dwelt on here. When every second beat is an extra systole, generally weaker than the preceding and the succeeding normal beat, the condi- tion is called pulsus bigeminus. The weaker beat is always followed by a compensatory pause of greater duration than that preceding it. From the pulsus bigeminus must be distinguished that form of alterna- ting pulse termed pulsus alternans, in which every second beat is dimin- shed in size, but the intervals separating the beats are of uniform length. THE CIRCULATION OF THE BLOOD AND LYMPH The refractory period is shorter for strong than for weak stimuli, and is markedly diminished by raising the temperature of the heart. So that stimulation of the heated heart with a series of strong induction shocks may cause a tetaniform condition, if not a typical tetanus. The con- traction of the normally beating heart is really a simple contrac- tion, and not a tetanus. The electrical changes correspond to a single contraction (p. 833); and when the nerve of a nerve-muscle pre- paration is laid on the heart, the muscle responds to each beat by a simple twitch, and not by tet- 3. — Refractory Period and Compensatory Pause A frog's heart was stimulated at a point corre- sponding to the nick in the horizontal line below each curve. In i and 2 there was no response ; in 3 and 4 there was an extra contraction, succeeded by a compensatory pause. anus (p. 203). That the cardiac muscle itself, apart from the intrinsic nervous mechanism, shows the phenomenon of ' refractory state ' has been shown in the Limulus heart after extirpation of the ganglion (Carlson). Like ordinary skeletal muscle, the cardiac muscle is at first bene- fited by contraction, perhaps by an ' augmenting ' action of fatigue- products such as carbon dioxide (Lee), so that when the apex is stimulated at regular intervals each contraction is somewhat stronger than the preceding one. To this phenomenon the name of the staircase or ' treppe ' has been given, from the appearance of the tracings (p. 749). SECTION V. — THE NERVOUS REGULATION OF THE HEART (EXTRINSIC NERVOUS MECHANISM OF THE HEART). While, as we have seen, the essential cause of the rhythmical beat of the heart resides in the tissue of the heart itself, it is constantly affected by impulses that reach it from the central nervous system. These impulses are of two kinds, or, rather, produce two distinct effects: inhibition, shown by a diminution in the rate or force of the heart-beat, or in the ease with which the contraction is conducted over the heart-wall; and augmentation, or increase in the rate or THE NERVOUS REGULATION OF THE HEART '57 force of the beat or in the conductivity. Both the inhibitory and th.3 augmentor impulses arise in the medulla oblongata, and perhaps a narrow zone of the neighbouring portion of the cord ; and they can be artificially excited by stimulation in this region. They pursue their course to the heart by fibres which may in certain animals be • mingled together, but are anatomically distinct. We may, there- fore, divide the extrinsic or external nervous mechanisrr of the heart into a cardio-inhibitory centre with its efferent inhibitory nerve-fibres and a cardio-augmentor centre with its efferent acceler- ator or augmentor fibres. Both of those centres, as we shall see, have, in addition, extensive relations with afferent nerve-fibres from all parts of the body, including the heart itself. It was in the vagus of the frog that inhibitory nerves were first discovered by the brothers Weber seventy years ago, and even now our knowledge of the cardiac nervous mechanism is more com- plete in this animal than in any other. We shall, therefore, first describe the phenomena of inhibition and augmen- tation as we see them in the heart of the frog, and then pass on to the mammal. In the frog the inhibitory fibres leave the medulla oblongata in the vagus nerve. The augmentor fibres come off from the upper part of the spinal cord by a branch from the third nerve to the third sympa- thetic ganglion, and thence find their way along the sympathetic cord to its junction with the vagus, in which they run, mingled with the inhibitory fibres, down to the heart. When the vago-sympathetic in the frog or toad is cut, and its peripheral end stimulated, . the heart in the vast majority of cases is stopped or slowed, or its beat is distinctly weakened without, it may be, any marked slowing. In other words, the rate at which the heart was working before the stimulation is greatly diminished, or reduced to- zero. Such an effect, a diminution of the rate of working, we call Inhibition. What precise form the inhibition shall take, whether the stoppage shall be complete or partial, and if partial whether the beats shall be simply weakened without being slowed or both slowed and weakened, appears to depend partly upon the strength of the stimulus used and partly upon the state of the heart itself. Some hearts it may be impossible to stop with weak stimulation, although other signs of inhibition may be distinct; Fig. 69. — Diagram of Extrinsic Nerves of Frog's Heart (after Foster). Ill, 3rd spinal nerve; AV, annulus of Vieus- sens; X, roots of vagus; IX. glosso-pharyngeal nerve; VS. combined vagus and sympa- thetic; i. 2, and 3, the ist, 2nd, and 3rd sympathetic ganglia. The dark line indi- cates the course of the sym- pathetic fibres. The arrows show the direction of the aug- mentor impulses. 158 THE CIRCULATION OF THE BLOOD AND LYMPH while they are readily stopped by stronger stimulation. In other cases the strongest stimulation may not produce complete standstill. Again, the inhibitory effect produced on a heated heart by a given strength of stimulation of the vagus may be greater than that caused in a heart at the ordinary temperature or a cooled heart. This is especially evident on the auricular tracings when these are recorded separately from those of the ventricle. Even on the verge of heat standstill of the heart inhibition is easily obtained (Fig. 71). Some writers have assumed that the different inhibitory effects produced by the vagus are due to the existence in it of separate groups of fibres, some affecting only the rate of the contraction, others its strength, Fig. 70. — Tracing from Frog's Heart. A, auricular, V, ventricular tracing. Sinus stimulated (primary coil 70 mm. from secondary). Heart at temperature 11-2° C. Complete standstill. The time-tracing between the curves marks intervals of two seconds. others still the conductivity of the muscular tissue and its excita- bility. This theory has enriched the vocabulary of physiology with a number of sonorous terms derived from the Greek, but has not otherwise established itself, although it has been ugfftil in 'emphasizing the fact that the inhibitory nerves can inflnenrp ^jie heail "beat hi several distinct wavs. ~ but there are other points of importance to be noted in regard to this inhibition : (i) It does not begin for a little time after stimulation has begun. In other words, there is a distinct latent period; and the length of this latent period is related to the phase of the heart's THE NERVOUS REGULATION OF THE HEART 159 contraction at which the stimulus is thrown in, and to the rate at which the heart is beating. As a general rule, the heart makes at least one beat before it stops.^-" " . - "(2)" The inhibition does not continue indefinitely, even if stimula- tion of the nerve is kept up. Sooner or later, and usually, in fact, afterari interval of a few seconds, the heart begins again to beat iTTt .^, ccO?* haTbeencpmrjletelvstoppeo^ or to quicKen its beat if it has only b"een slowed, or JxTstrengthen it n the inhibition has onlyj^e^ker^edjthe j±s,oid rate oi ^ _ so, but very often there follows a longer or shorter period during which the heart works at a greater rate than it did before the inhibi- tion, and this greater rate of working may be manifested by increased Fig. 71. — Activity of Vagus on Verge of Heat Standstill. Auricular and ventricular contractions of toad's heart recorded. Heart at 34'5° C., v 50, stimulation of vagus (distance of coils 50 mm.). The ventricle was already in heat standstill; the auricle was at once inhibited. Then follows secondary augmentation (due to the sympathetic fibres), during which the ventricle also resumes beating. An interval of a minute elapsed between the first and second parts of the tracing, during which the heart remained at 34'5° C. The auricle, was almost in standstill (contractions can still be seen on the curve with a lens), when the vagus was again stimulated at v 50 with the same distance between the coils. Complete inhibition followed by secondary augmentation. frequency of beat, or increased strength of beat, or by both. When the temperature of the heart is low, increased frequency; when it is high, increased strength, is generally seen during this period of secondary augmentation. * The cause of this secondary augmentation, and of the primary augmentation sometimes seen in fresh prepara- tions and often in hearts that have been long exposed (Fig. 73), excited much speculation before it was known that sympathetic fibres existed in the vagus. There is no longer any doubt that it is due to the stimulation of these accelerator or, as it is better to call * Augmentation is termed ' secondary ' when it is preceded by inhibition, primary ' when it is not so preceded 160 THE CIRCULATION OF THE BLOOD AND LYMPH them (since mere acceleration is not the only consequence of their stimulation), augmentor fibres in the mixed nerve. Fr»r lr\ pYrjfci- tion of the roots of the vagus proper within the skull, and therefore above the junction of the sympathetic fibres, causes no secondary augmentation, or very little, and the inhibition lasts far longer than when the mixed trunk is stimulated. (2) Excitation of the upper or cephalic end of the sympathetic cord before it has joined the vagus causes, after a relatively long latent period, marked augmenta- tion. And if the contractions of the heart are registered, the tracing bears a close resemblance to the curve of secondary augmen- tation following excitation of the mixed nerve on the other side with an equally strong stimulus and for an equal time. (3) When the vago-sympathetic is stimulated weakly there is little or no Fig. 72. — Frog's Heart: Vagus stimulated. Temperature of heart 8° C.; 78 mm. between the coils. Diminution in force of auricle and ventricle, but not com • plete standstill. Time-tracing shows two-second intervals. secondary augmentation. Now, it is known that the augmentor fibres require a comparatively strong stimulus to cause any effect when they are separately excited, whereas a weak stimulus will excite the inhibitory fibres. The question arises at this point, why it is that, when the inhibitory and augmentor fibres are stimulated together in the mixed nerve (and the same is true when the sympathetic on one side and the vagus on the other are stimulated at the same time), the inhibitory effect always comes first, when there is any inhibitory effect, while the augmentation always has to follow. The answer has sometimes been given, that the latent period of the augmentor fibres is longer than that of the inhibitory fibres. But although this is certainly the case, the answer is insuffi- cient. For the period of postponement may be much greater than the THE NERVOUS REGULATION OF THE HEART 161 latent period of the sympathetic fibres when stimulated by themselves. The inhibition apparently runs its course without being affected by the simultaneous augmcntor effect, which, lying latent until the end of the inhibition, then bursts out and completes its own curve. It is not like the passing of two waves through each other, but rather like the stopping of one wave until the other has passed by. It seems as if augmentation cannot develop itself in the presence of inhibition — at least, until the latter is nearly spent. Like a musical-box devised to play a series of melodies in a fixed order, and from which a particular tune cannot be obtained till those preceding it have been run through, the heart, in some way or other, is arranged, in the presence of competing impulses from its extrinsic nerves, to play the tune of inhibition before the tune Fig. 733. — Frog's Heart. A, auricular, V, ventricular tracing. Ventricle beating very feebly. Vagus stimulated (60 mm. between coils). Marked augmentation of ventricular beat. of augmentation. In the frog, at any rate, the two processes can hardly be considered as antagonistic, in the sense that a definite amount of augmentor excitation can overcome a definite amount of inhibitory excitation. Nor is it the case that, when the heart is played upon at the same time by impulses of both kinds, it pits them against each other and strikes the balance accurately between them. It is possible, how- ever, that when the inhibitory fibres are very weakly, and the augmentor fibres very strongly stimulated, the amount of inhibition may be some- what diminished. In mammals, on the other hand, a true antagonism seems to exist; and stimulation of the inhibitory nerves is less effective when the augmentors are excited at the same time. The cardiac nerves affect not only the rate and force of the contraction, but also the con- 162 THE CIRCULATION OF THE BLOOD AND LYMPH ductivity of the heart. Thus in the frog's heart during stimulation of the vagus, the contraction passes more slowly, and during stimulation of the sympathetic more quickly, from auricles to ventricle. In mammals (and in what follows we shall restrict ourselves chiefly to the dog, cat, and rabbit, as it is in these animals that the subject has been most carefully studied) the inhibitory fibres run down the vagus in the neck and reach the heart by its cardiac branches. They are derived from the bulbar roots of the spinal accessory, whose inner branch joins the vagus. The augmentor fibres leave the spinal cord in the anterior roots of the second and thira thoracic nerves, and possibly to some extent by the fourth and fifth. Through the corresponding Fig. 736- — Miniature Myocardiograph (actual size) (Wiggers). C, light. aluminium. segment capsule covered by tightly stretched rubber dam, upon which is cemented an aluminium plate pivoting upon the chord side of the capsule. The plate carries a light arm, A, with an eyelet at its end. A similar arm, A1, is rigidly fastened to the body of the capsule. The arms can be bent so that the distance between the eyelets varies from 3 to 25 millimetres, and are connected by stitches through the eyelets to points on the auricular surface. The approximation of the points causes a negative pressure in the cap- sule, and in the recording capsule (see descrip- tion of Fig. 31, p. 93) connected with it by the tube T. Hence, the down-stroke of the^curve represents contraction, the up-stroke relaxation. The apparatus is supported by a very', light spring, S. This enables it to follow varying degrees of distension and movement of the auricle without affecting the curve of contrac- tion. The small mass (less than 2 gm.)'and high vibration frequency of the instrument insure a more faithful record of the con- traction than with older forms. white rami communicantes they reach the sympathetic cord, and running up through the stellate ganglion (first thoracic), and the annulu* of Vieussens, which surrounds the subclavian artery, to the inferior cervical ganglion, they pass off to the heart by separate ' acce- lerator ' branches, taking origin either from the annulus or from the inferior cervical ganglion. Some augmentor fibres are often, if not always, present in the dog's vago-sympathetic in the neck. It is especially easy to demonstrate their presence five or six days after section of the nerve, when the excitability of the inhibitory fibres has disappeared. In the dog the vagus and cervical sympathetic are, in the great majority of cases, contained in a strong common sheath, and pass THE NERVOUS REGULATION OF THE HEART 163 together through the inferior cervical ganglion. Upon opening this sheath they may with care be separated, the fibres running in distinct strands, and not mixed together as in the vago-sympathetic oi the frog. For some distance below the superior cervical ganglion the cervical sympathetic is not connected with the vagus, and here the nerves may be separately stimulated without any artificial isolation. In the rabbit and some other mammals, including man, the vagus and sympathetic run a separate course in the neck. In the mammal, the inhibitory fibres have a smaller direct action on the ventricle than in the frog. It indeed beats more slowly when the auricle is slowed, but this is only because in the normally beating heart the ventricle takes the time from the auricle. The strength of the ventricular contractions may be not at all diminished, even when the auricle is beating very feebly during inhibition. When the auricle is completely stopped, which does not occur so readily as in the frog, the ventricle also stops for a short time, but soon begins to beat again with an independent rhythm of its own. In the frog the ventricle is directly affected by stimulation of the vagus, and the force of its beats is diminished independently of the inhibitory effects in the auricles (Practical Exercises, pp. 198, 203). It has been shown by delicate optical methods of recording the contraction of small units of the auricular musculature (designated as the ' fractionate ' contraction by Wiggers) by the method illus- trated in Fig. 736, that the diminution in the size of the contrac- tion produced by stimulation of the vagus is essentially due to depression of the contractility of the muscle, and not, or at any rate not primarily, to a diminution in its excitability. In other words, the strength of stimulus needed to elicit a contraction of the muscle when under the influence of vagus excitation, in the early part of the inhibition at least, is not increased, whereas the size of the contraction evoked by a given stimulus is diminished (Fig. 74) . It is not doubted that the excitation of the vagus does reduce the excitability of the auricular muscle, but this reduction seems not to occur so early as the reduction in the amplitude of the beat, and cannot therefore be responsible for it. The duration of each (fractionate) contraction is not altered by vagus excitation. It can likewise be shown that the depression of contractility is not secondary to the depression of conductivity produced by the vagus. Finally, it is not dependent upon the slowing of the rhythm which accompanies the diminution in the contraction of the naturally beating auricle. For when the auricle is compelled to beat with an artificial rhythm by applying to it a series of regularly spaced electrical stimuli at a more rapid rate than the normal rhythm, stimulation of the vagus still causes reduction in the amplitude of the contraction without change in the rate (Fig. 74). One other point is worthy of note. Excitation of the vagus causes an increase in the size of the first, or occasionally of the first two subsequent I64 THE CIRCULATION OF THE BLOOD AND LYMPH contractions of the auricular muscle when the rhythm is slowed, but not otherwise. It is probable that this is due to the beneficial influence of the longer period of rest associated with the diminution in frequency of the beat before the inhibitory action has had time to cause depression in the amplitude (Wiggers). Fig. 74. — Influence of the Vagus on the Contraction of the Dog's Right Auricle The two upper curves are simultaneous records of the contraction of very short portions of the auricle (so-called fractionate contraction) taken from two regions, one near the sinus node P, and one far from it, D. The contraction begins and ends slightly sooner in the proximal than in the distal region. The lowest curve shows the depressant effect of the vagus excitation mani- fested when an artificial rhythm (a series of electrical break shocks), which is not altered by stimulation of the vagus, is substituted for the normal rhythm sustained by the ' pacemaker.' The inhibitory fibres, then, influence the heart particularly through the auricles; they are par excellence auricular nerves. On the other hand, the accelerantes in all mammals which have been investigated not only extend to the ventricles, but are even mainly distributed to them. They are emphatically ventricular fibres, and THE NERVOUS REGULATION OF THE HEART 165 in accordance with its greater mass the left ventricle receives more fibres than the right. Stimulation of the accelerator nerves in the dog causes an increase in the force of both the auricular and ventricular con- traction, and as a rule, in addition, some increase in the rate of the beat. As to the nature of the physiological linkage between the cardiac nerves and the muscular tissue of the heart we know but little. Ganglion-cells lie on the course of the vagus fibres after they have entered the heart, and although the view has been advocated that they are simply stations where the inhibitory impulses pass from medullated to non-medullated fibre s, and where possibly other Fig. 75- — L ood-Pres-u 2 Tracings: Rabbit. Vagus stimulated at i. Stimulus siruuger in B than in A (Hiirthle's spring manometer). anatomical changes and rearrangements occur, they may be inter- mediate mechanisms which essentially modify the physiological impulses falling into them. It has been stated that in the dog the right vagus controls chiefly the rate of the heart, and the left vagus chiefly the conduction from auricles to ventricles, and the suggestion has been made that this is because the right vagus has a special relation to the sino-auricular node, in which impulses are supposed to arise, and the left vagus a special relation to the auriculo-ven- tricular node, the upper end of the A-V bundle, the main conduction system (Cohn and Lewis). The nervi accelerantes are already non-medullated before they 166 THE CIRCULATION OF THE BLOOD AND LYMPH reach the heart. The fact that the action of the accelerantes can be restored by perfusing the heart with a nutrient solution at a much longer interval after somatic death than the action of the vagus strengthens the suggestion that ganglion-cells are interposed on the inhibitory though not on the augmentor path, without, however, proving of itself that such a difference exists In one experiment the heart of an anthropoid ape was revived wnen thiee successive periods — viz., four and a half, twenty-eight and a half, and fifty-three hours respectively — had elapsed after the death of the animal, although during the last period the heart had been twice frozen hard. The vagus was shown to be still capable of causing some inhibition six hours after death, and the accelerans some augmentation as late as fifty-three hours after death (Hering). In the discussions over the relation of the extrinsic to the intrinsic cardiac nervous apparatus appeal has frequently been made to the action of certain poisons on the heart. Thus, after nicotine stimulation of the vago-sympathetic causes no inhibition of the frog's heart ; it may cause augmentation. But stimulation of the junction of the sinus and auricle still causes inhibition. Atropine not only abolishes the inhibi- tory effect of stimulation of the vagus trunk, but also that of stimula- tion of the junction of sinus and auricle. Muscarine causes diastolic arrest in a heart already poisoned with nicotine, but not in a heart under the influence of atropine. And a heart brought to a standstill by muscarine can be made to beat again by the application of atropine, although not by nicotine. These facts may be explained as follows : Nicotine paralyzes, not the very ends of the vagus, but the ganglia through which its fibres pass. Stimulation of the sinus, which is practically stimulation of the vagus fibres between the ganglion-cells and the muscular fibres, is therefore effective, although stimulation of the nerve-trunk is not (Langley). On the other hand, the atropine group paralyzes the nerve-endings themselves, or interferes with the reception of the inhibitory impulses by acting on a so-called receptive substance in the muscle (p. 182), so that neither stimulation of the sinus nor of the nerve-trunk can cause inhibition. Muscarine, on the contrary, stimulates the vagus fibres between the nerve-cells and the muscle, or the actual nerve -end ings, or exerts an inhibitory action on the muscle itself through the appropriate receptive substance, and thus keeps the heart in a state of permanent inhibition, which is removed when atropine cuts out the nerve-endings, or combines with the receptive substance. It is quite in accordance with this that muscarine has no effect on a heart whose vagus nerves, as occasionally happens, have no inhibitory power. Pilocarpine has very much the same action as muscarine. Stannius' Experiment.— Another series of phenomena, intimately related to our present subject, have excited, since they were first made known by Stannius, an enormous amount of discussion. The chief facts of this classical experiment we have already mentioned (p. 144), and they are also described in the Practical Exercises (p. 194). They are easy to verify, but difficult to interpret. The most probable explana- tion of the standstill caused by the first ligature is that the lower portion THE NERVOUS REGULATION OF THE HEART 167 of the heart, when cut off from the sinus in which the beat normally originates, needs some time for the development of its automatic power to the point at which an independent rhythm can be maintained. The effects following the second Stannius ligature seem to depend upon the power of the ventricle to develop and maintain an independent rhythm, but the contractions are supposed by some to be started by stimulation of the muscular tissue in the auriculo- ventricular groove by the ligature. Nature of Inhibition and Augmentation.— So far we have been dis- cussing the phenomena of inhibition and augmentation as ultimate facts. We have not attempted to go behind them, nor to ask what it is that really happens when inhibitory impulses fall into a heart, which from the first days of embryonic life has gone on beating with a regular rhythm, and in the space of a second or two bring it to a standstill. The question cannot fail to press itself upon the mincl of anyone who has ever witnessed this most beautiful of physiological experiments; Fig. 76. — Frog's Heart. Sympathetic stimulated (30 mm. between the coils). Temperature 12°. Marked increase in force. Only auricular tracing rcpro. duced. Time-trace, two-second intervals. but as yet there is no answer except ingenious speculations. The most plausible of these is the trophic theory of Gaskell, who sees in the vagus a nerve which so acts upon the chemical changes going on in the heart as to give them a trophic, or anabolic, or constructive turn, and thus to lessen for the time the destructive changes underlying the muscular contraction. The augmentor nerves, on the other hand, are supposed to exert a katabolic influence, and to favour these destructive changes. And while, according to Gaskell, the natural consequence of inhibition is a stage of increased efficiency and working power when the inhibition has passed away, the natural complement of augmentation is a tem- porary exhaustion. It is very risky, however, to rely, as Gaskell did, upon a supposed change of sign in the electrical effects during vagus stimulation, and the only chemical test to which the theory has been subjected, the comparison of the oxygen consumption of the heart during and in the absence of inhibition, is adverse to it. The amount of oxygen used up relatively to the functional activity of the heart as measured by the product of the frequency of the beat and the maximal increase of pres- sure caused by it, is not increased by stimulation of the vagus (Rohde). 163 THE CIRCULATION OF THE BLOOD AND LYMPH Whatever the exact mechanism of augmentation may be, there is no basis for the statement that the cardio-augmentor nerves have an action on the heart so fundamentally different from the action of motor nerves on skeletal muscle that they cannot originate contractions in a heart entirely at rest. Excitation of the cardio-augmentor nerves can cause rhythmical contractions in the perfectly quiescent heart of molluscs, and a sudden and prolonged outburst of beats of great force in the frog's heart, which has been brought to a standstill by cautiously heating it to 40° to 43° C. (Practical .Exercises, p. 194) fo'r a minute or two, or to a considerably lower temperature, for a longer time (Fig. 77). A similar effect can be obtained on the quiescent mammalian heart by stimulation of the nervi accelerantes. 28" 5 S30 M 1 1 1 1 m 1 1 ii i 1 1 1 1 1 1 n M i 1 1 1 1 1 1 1 1 1 n 1 1 1 1 1 n rr jqg. 77. — Effect of Stimulation of Frog's Cardiac Sympathetic during Complete Standstill of the Heart at 28-5° C. Upper tracing, auricle ; lower, ventricle. To be read from right to left. Time-trace, two-second intervals. The Normal Excitation of the Cardiac Nervous Mechanism.— We have now to inquire how this elaborate nervous mechanism is normally set into action. And we may say at once, that striking as are the effects of experimental stimulation of the vagus trunk or the nervi accelerantes in their course, it is only under exceptional cir- cumstances that the efferent nerve-fibres, at any rate before they have entered the heart, can be directly excited in the intact body. In certain cases the pressure of a tumour or an aneurism on the nerve-trunks, or, in the case of the accelerators the progress of a pathological change in the sympathetic ganglia through which the THE NERVOUS REGULATION OF THE HEART 169 nerve-trunks, or, in the case of the accelerators, the progress of a pathological change in the sympathetic ganglia through which the fibres pass, has been thought to bring about by direct stimulation a slowing or a quickening of the pulse. In some individuals the vagus has been excited by compressing it against a bony tumour in the neck; and by compressing the nerve against the vertebral column it is possible to cause inhibition in many normal persons, although it ought to be stated that the experiment is not free from danger. But it is from the cardie-inhibitory and cardio-augmcntor centres in the medulla oblongata that the impulses which regulate the activity of the heart are normally discharged. Inhibitory im- pulses are constantly passing out from the medulla, for section of both vagi causes almost invariably an increase in the rate of the heart, at least in mammals, although the increase is less conspicuous in animals like the rabbit, whose normal pulse-rate is high, than in animals like the dog, whose pulse-rate is comparatively low. Section of one vagus usually causes only a comparatively slight increase, for the other is able of itself to control the heart. It is not certainly known whether the augmentor centre in like manner discharges a continuous stream of impulses, or is only roused to occasional activity by special stimuli. For the results of section of the nervi acceler- antes, or the extirpation of the inferior cervical and stellate ganglia, are dubious and conflicting. But if it does exert a tonic influence on the heart, this is feebler than the tone of the inhibitory centre. As to the nature of this inhibitory tone, and the manner in which it is maintained, we know but little. It may be that the chemical changes in the nerve-cells of the inhibitory centre lead of themselves to the discharge of impulses along the inhibitory nerves. But there is some evidence that, in the complete absence of stimulation from without, the activity of the centre would languish, and perhaps be ultimately extinguished. For when the greater number of the afferent impulses have been cut off from the medulla oblongata by a transverse section carried through its lower border, division of the vagi produces little effect on the rate of the heart. Also, when the upper cervical cord and the brain are resuscitated after a period of anaemia, the return of cardio-inhibitory tone is tardy in comparison with the return of the truly automatic function of respiration, and does not seem to precede the opening up of the afferent paths to the cardio-inhibitory centre. Indeed, reflex inhibition may be produced at a time when the inhibitory centre has regained none of its tone. The suggestion is that the normal tone of the centre is largely dependent upon reflex impulses. Be this as it may, we know that the activity of the inhibitory centre is profoundly influenced — and that both in the direction of an increase and of a diminution — by impulses that fall into it through afferent nerves and by stimuli directly applied to it. And we may assume that the same is true of the augmentor centre. The common statement that stimulation I7<> THE CIRCULATION OF THE BLOOD AND LYMPH of the central end of one vagus, the other being intact, produces distinct inhibition does not hold for all mammals. In dogs this is sometimes the case, but often (under anaesthesia, at any rate) there is little or no inhibition, or even augmentation. In etherized cats, on the other hand, some inhibition is always seen. Of all the afferent fibres of the vagus, the pulmonary fibres produce the most marked reflex inhibition. The cardiac fibres are much less effective. These pulmonary nerves also influence the respiratory and vaso- motor centres. The respiration is temporarily arrested, and the blood-pressure falls through the dilatation of the small arteries when they are excited. It is of interest in connection with the subject of death during the administration of anaesthetics, that the afferent vagus fibres coming from the alveoli of the lungs can be chemically stimulated when irritant vapours, such as chloroform, hydrochloric acid, ammonia, bromine, or formaldehyde are inhaled through a tracheal cannula, causing reflex arrest of the heart and of the respira- tory movements and a fall of blood-pressure through vaso-dilatation (Brodie). At a certain stage in chloroform anaesthesia, before it has become very deep, comparatively trifling causes may bring about great and sudden changes in the pulse-rate, owing to the abnormal mobility of the vagus centre (MacWilliam). The depressor nerve, a branch of the vagus, which is easily found in the rabbit as a slender nerve running close to the sympathetic in the neck, and a little to its inner side, but in the dog is usually blended with the vago-sympathetic, falls into the same category ^•ith the vagus itself as regards its reflex action on the heart, to which it bears an important relation. In all mammals some of its fibres end in the wall of the aorta, but some of them may run down over the heart to the ventricle. Stimulation of its peripheral end has no effect, for the fibres in it which influence the circulation are afferent, not efferent. But excitation of its central end causes a marked fall of blood- pressure (p. 185), accompanied by, but not essentially due to, a distinct slowing of the heart. If the animal is not anaesthetized, there may be signs of pain, and for this reason the depressor has sometimes been spoken of, somewhat loosely, as the sensory nerve of the heart. The abdominal sympathetic (of the frog) also contains afferent fibres, through which reflex inhibition of the heart can be produced when they are excited mechanically by a rapid succession of light strokes on the abdomen with the handle of a scalpel. . On the other hand, when the central end of an ordinary peripheral ne/ve like the sciatic or brachial is excited, the common effect is pure augmentation (Fig. 78), which sometimes develops itself with even greater suddenness than when the accelerator nerves are directly stimulated. Occasionally, however, the augmentation is abruptly followed by a typical vagus action. Here the reflex inhibitory effect seems to break in upon and cut short the reflex augmertor effect. THE NERVOUS REGULATION OF THE HEART 171 These examples show that certain afferent nerves are especially related to the cardio-inhibitory, and others to the car dio- augment or, centre, or at least that the central connections of some nerves are such that inhibition is the usual effect of their reflex excitation, while the opposite is the case with other nerves. But it is im- probable that the effect of a stream of afferent impulses reaching the cardiac centres by any given nerve is determined solely by anatomical relations. The intensity and the nature of the stimulus seem also to have something to do with the result. For when ordinary sensory nerves are weakly stimulated, augmentation is said to be more common than inhibition, and the opposite when they are strongly stimulated. And while a chemical stimulus, like the inhaled vapour of chloroform or ammonia, causes in the rabbit Fig. 78. — Myocardiographic Tracing of Cat's Ventricle. The signal line shows the point at which the central end of the brachial nerve was stimulated during resuscitation of the animal after a period of cerebral anaemia. Some augmenta- tion of the ventricular beat is seen. The notches in the ventricular tracing are doe to the artificial respiration. Time-trace, seconds. reflex inhibition of the heart through the fibres of the trigeminus that confer common sensation on the mucous membrane of the nose, the mechanical excitation of the sensory nerves of the pharynx and oesophagus when water is slowly sipped causes acceleration.* The stimulation of the nerves of special sense is followed sometimes by the one effect and sometimes by the other. To complete the catalogue of the nervous channels by which impulses may reach the cardiac centres in the medulla, we may add that there must be an extensive connection between them and the cerebral cortex, since every passing emotion leaves its trace upon the curve of cardiac action. The so-called ' reflex cardiac death/ which is an occasional consequence of intense psychical influences (anxiety, fright, etc.), * In 78 healthy students the average pulse-rate (in the sitting position \ was increased from 73 to 85 per minute by sipping water. 172 THE CIRCULATION OF THE BLOOD AND LYMPH may be due to the prolonged excitation of the cardio-inhibitory centre, as well as to the disturbance of other centres in the bulb by the cortical storm. It is a remarkable fact, too, and one that can only be explained by such a connection, that although in the vast majority of individuals the will has no influence whatever on the rate or force of the heart, except, perhaps, indirectly through the respiration, some persons have the power, by a voluntary effort, of markedly accelerating the pulse. In one case of this kind it was noticed that perspiration broke out on the hands and other parts of the body when the heart was voluntarily accelerated. A rise of blood-pressure due to constriction of the vessels has also been observed. The effort cannot be kept up for more than a short time, and the pulse-rate quickly goes back to normal. It has been recently shown that this peculiar power is more common than has been supposed, and that where it is present in rudiment it can be cultivated, although it is a dangerous acquisition. As an example of the direct action on a cardiac centre of a changed chemical composition of the blood, we may cite the inhibition produced by injection of bile into a vein and revealed in the slow pulse of many cases of jaundice ; and as an instance of the direct action of a physical change, the slowing of the heart as the blood-pressure rises (p. 188) in asphyxia or on clamping the aorta. The variation in the pulse-rate associated with changes in the position of the body, to which we have already referred (p. 107), is brought about by direct stimulation of the in- hibitory centre by the increase of blood-pressure in the medulla oblongata when a person who has been standing assumes the supine, or even the sitting, posture. But it is also due in part to changes in the amount of muscular contraction, since muscular exercise causes acceleration of the heart either reflexly, through afferent muscular nerves, or by a direct effect of waste products of the metabolism of the muscles on the cardiac centres in the bulb or on the heart itself (p. 280). Theoretically, quickening of the heart might be caused either by a diminution in the inhibitory tone or by an increase in the activity of the augmentor centre; and slowing of the heart might be due either to a diminution in the augmentor tone, if such exists, or to an increase in the activity of the inhibitory centre. So that it is not always easy to interpret such results as we have quoted above. But it would appear that under ordinary conditions the rate of the heart is mainly regulated by the inhibitory centre, which, within a considerable range, can produce variations in either direction. The augmentor mechanism is perhaps merely auxiliary to the inhibitory, being called into action only in emergencies. THE NERVOUS REGULATION OF THE BLOODVESSELS 173 SECTION VI. — THE NERVOUS REGULATION OF THE BLOODVESSELS (VASo-MoTOR NERVES;. Just as the muscular walls of the heart are governed by two sets of nerve-fibres, a set which keeps down the rate of working and a set which may increase it, the muscular walls of the vessels are under the control of nerves which have the power of diminishing their calibre (v aso- constrictor) , and of nerves which have the power of increasing it (vaso-dilator). All nerves that affect the calibre of the vessels, whether vaso-constrictor or vaso-dilator, are included under the general name vaso-motor. These vaso-motor nerves, like the augmentor and inhibitory fibres of the heart, are connected with a centre or centres, which in turn are in relation with numerous afferent nerves. It is convenient to distinguish the afferent nerves which cause on the whole a vaso-constriction and a consequent increase of arterial pressure as pressor nerves, and those which cause on the whole vaso-dilatation, with fall of pressure, as depressor nerves, reserving the terms vaso-constrictor and vaso-dilator for the efferent portions of the reflex arcs. It is through this reflex mechanism that the bloodvessels are mainly influenced, although the endings of the vaso-motor nerves in the smooth muscular fibres or the muscular fibres themselves are sometimes directly affected by sub- stances circulating in the blood. Proteoses, for instance, cause by peripheral action dilatation of the vessels and a fall of blood-pressure (p. 215); suprarenal extract, or its active principle, adrenalin, or epinephrin, constriction, with a rise of pressure (pp. 216, 655). Apo- codeine paralyzes the vaso-motor nerve-endings after a preliminary stimulation, and now adrenalin causes no constriction. Chryso- toxin, an active principle of ergot, causes a marked rise of blood- pressure by stimulating the sympathetic ganglion-cells or the pre- ganglionic fibres of the vaso-constrictor path. Vaso-motor nerves control chiefly the small arteries. They have no direct influence on the capillaries.* Nor has the existence of an effective vaso-motor regulation of the calibre of the veins, except in the portal system, been proved up to this time by any clear and unambiguous experi- ment, although there are grounds on which it has been surmised that the nervous system does influence the ' tone ' of the whole venous tract. These grounds will be mentioned in the proper place. * It is usually taught that the capillaries, being devoid of muscular fibres in their walls, are not supplied with vaso-motor fibres, and that the only kind of active contraction of which they are capable is due to a process analogous to the turgescence of vegetable cells, the thickness of the wall being increased at the expense of the lumen, while the total cross-section of the vessel remains unchanged. It has been asserted, however, that a true contraction, in which both the total section and the lumen are diminished, may be caused in the capillaries of the nictitating membrane of the frog either by direct stimulation or by excitation of vaso-motor fibres in the sympathetic (Steinach and Kahn). 174 THE CIRCULATION OF THE BLOOD AND LYMPH Meanwhile, before describing the distribution of the best-known tracts of vaso-motor fibres and defining the position of the vaso- motor centres, we must glance at the methods by which our know- ledge has been attained. (1) In translucent parts inspection is sufficient. Paling of the part indicates constriction; flushing, • dilatation of the small vessels. This method has been much used, sometimes in conjunction with (2), in such parts as the balls of the toes of dogs or cats, the ear of the rabbit, the conjunctiva, the mucous membrane of the mouth and gums, the web of the frog, the wing of the bat, the intestines, etc. (2) Observation of changes in the temperature of parts. This method has been chiefly employed in investigating the vaso-motor nerves of the limbs, the thermometer bulb being fixed between the toes. In such peripheral parts the temperature of the blood is normally less than that of the blood in the internal organs, because the opportunities of cooling are greater. The effect of a freer circulation of blood (dilatation of the arteries) is to raise the temperature ; of a more restricted circulation (constriction of the arteries), to lower it. (3) Measurement of the blood-pressure. If we measure the arterial blood-pressure at one point, and find that stimulation of certain nerves increases it without affecting the action of the heart, we can conclude that upon the whole the tone of the small vessels has been increased. But we cannot tell in what region or regions the increase has taken place ; nor can we tell whether it has not been accompanied by diminution of tone in other tracts. But if we measure simultaneously the blood-pressure in the chief artery and chief vein of a part such as a limb, we can tell from the changes caused by section or stimulation of nerves whether, and in what sense, the tone of the small vessels within this area has been altered. For example, if we found that the lateral pressure in the artery was diminished, while at the same time it was increased in the vein, we should know that the ' resistance ' between artery and vein had been lessened, and that the blood now found its way more readily from the artery into the vein. If, on the other hand, the venous pressure was diminished, and the arterial pressure simultaneously increased, we should have to conclude that the vascular resistance in the part was greater than before. If the pressure both in artery and vein was increased, we could not come to any conclusion as to local changes of resistance with- out knowing how the general blood-pressure had varied. (4) The measurement of the velocity of the blood in the vessels of the part. This may be done by the stromuhr or dromograph, or by allowing the blood to escape from a small vein and measuring the outflow in a given time, or, without opening the vessels, by estimating the circulation-time (p. 135). When changes in the general arterial pressure are eliminated, slowing of the blood-stream through a part corresponds to increase of vascular resistance in it; increase in the rate of flow implies diminished vascular resistance. Sometimes the red colour of the blood issuing from a cut vein, and the visible pulse in the stream, indicate with certainty that the vessels of the organ have been dilated. (5) Alterations in the volume of an organ or limb are often taken as indications of changes in the calibre of the small vessels in it. We have already seen how these alterations are recorded by means of a plethysmo graph (p. 128). The brain is enclosed in the skull as in a natural plethysmograph, and changes in its volume may be registered by connecting a recording apparatus with a trephine hole. THE NERVOUS REGULATION OF THE BLOODVESSELS 175 (6) For the separation of the effects of stimulation of vaso-constrictor and vaso-dilator fibres when they are mingled together, as is the case in many nerves, advantage is taken of certain differences between them. For example, the vaso-constrictors lose their excitability sooner than the vaso-dilators when cut off from the nerve-cells to which they belong. So that if a nerve is divided, and some days allowed to elapse before stimulation, only the dilators will be excited. The vaso-dilators are more sensitive to weak stimuli repeated at long intervals than to strong and frequent stimuli, and the opposite is true of the constrictors. When a nerve containing both kinds of fibres is heated, the excitability of the vaso-constrictors is increased in a greater degree than that of the dilators; when the nerve is cooled, the dilators preserve their excita- bility at a temperature at which the constrictors have ceased to respond to stimulation (Fig. 79). The Chief Vaso-Motor Nerves. — The first discovery of vaso-motor nerves was made in the cervical sympathetic. When this nerve is Fig. 79- — Plethysmograms: Hind-Limb of Cat (after Bowditch and Warren). To be read from right to left. On the left hand is shown the effect of slow stimulation ot the sciatic (i per second); on the right hand the effect of rapid stimulation (64 per second). In the first case the limb swelled owing to excitation of the vaso-dilators; in the second, it shrank tl/:ough excitation of the vaso-constrictors. cut, the corresponding side of the head, and especially the ear, become greatly injected owing to the dilatation of the vessels. This experiment can be very readily performed on the rabbit, and the changes are most' easily followed in an albino. The ear on the side of the cut nerve is redder and hotter than the other; the main arteries and veins are swollen with blood, and many vessels formerly invisible come into view. The slow rhythmical changes of calibre, which in the normal rabbit are very characteristically seen in the middle artery of the ear, disappear for a time after section of the sympathetic, although they ultimately again become, visible. Stimulation of the cephalic end of the cut sympathetic causes a marked constriction of the vessels and a fall of temperature on the same side of the head. From these facts we know that the cervical 176 THE CIRCULATION Op THE BLOOD AND LYMPH sympathetic in mammals contains vasoconstrictor fibres for the side of the head and ear, and that these fibres are constantly in action. Certain parts of the eye, and the salivary glands, larynx, oesophagus, and thyroid gland, are also supplied with vaso- motor (constrictor) nerves from the cervical sympathetic. It has been asserted that the cervical sympathetic contains some of the vasoconstrictor fibres that supply the corresponding half of the brain and its membranes, but this has been disputed, and some ob- servers deny that the vessels 01 the brain have any vaso-motor nerves. Non-medullated nerve-fibies, however, may be seen in and around the walls of the cerebral and spinal bloodvessels, and it is difficult to believe that these have not a vaso-motor function, although this has not as yet been clearly demonstrated by experimental methods. It has sometimes been argued that we ought not to expect the brain to be supplied with vaso-motors. For it is enclosed in a rigid box, and the quantity of blood in it can be increased or diminished only to the slight extent to which the cerebro-spinal liquid can be displaced into the vertebral canal. Important changes in the cerebral blood-supply are therefore brought about, it is said, not by a widening or narrowing of the cerebral vessels, but by an alteration in the velocity of the blood in them as a result of a rise or fall of the systemic arterial pressure. This argument, however, leaves out of account the consideration that in general the brain does not function as a whole, but that certain parts of it may often become active and relatively hyperaemic, while other parts are inactive and relatively anaemic, and that important changes in the distribution of the blood in the encephalon may be effected, although the total mass of blood in the organ undergoes little or no alteration. It is, of course, true that it is not the absolute quantity of blood in an organ which is a function of its activity, but the rate at which it is renewed. And it is theoretically possible that an organ at rest should contain as much blood as when it is active, or even more. But such cases, if they exist, are certainly rare. The fact that adrenalin generally constricts the vessels of a perfused brain (Wiggers) is in favour of the existence of vaso-motors. The retina, which from the stand- point of development is a portion of the brain, is undoubtedly supplied with vaso-constrictor fibres which run in the cervical sympathetic. That the cervical sympathetic contains some dilator fibres is proved by the fact that stimulation of the cephalic end in the dog causes flushing of the mucous membrane of the mouth on the same side. Further, after division of the nerve on one side in the rabbit it may be observed that when the animal is excited.the vessels of the ear whose nerve is intact may become still more dilated than those whose constrictor fibres have been paralyzed. The only explana- tion is that vaso- dilators are being excited from the central nervous system. The vaso-motor fibres of the head run up in the cervical sympa- thetic, and then pass into various cerebral nerves, of which the fifth or trigeminus is the most important. The trigeminus nerve contains vaso-constrictor nerves for various parts of the eye (conjunctiva, sclerotic, iris), and for the mucous THE NERVOUS REGULATION OF THE BLOODVESSELS Ifj membrane of the nose and gums, and section of it is followed by dilatation of the vessels of these regions. The lingual branch of the trigeminus supplies vaso-motor fibres to the tongue, and ap- parently both vaso-constrictor and vaso-dilator. In some animals — the rabbit, for instance — the ear derives part of its vaso-motor supply through the great auricular nerve, a branch of the third cervical nerve, which they reach as grey rami from the stellate ganglion. Another great vaso-motor tract, the most influential in the body, is contained in the splanchnic nerves, which govern the vessels of many of the abdominal organs. Section of theSvi nerves causes an immediate and sharp fall of arterial pressure. The intestinal vessels are dilated and overfilled with blood. As a necessary consequence of their immense capacity, the rest of the vascular system is under- filled, and the blood-pressure falls accordingly. Stimulation of the peripheral end of the splanchnic nerves causes a great rise of blood- pressure, owing to the constriction of vessels in the intestinal area. We therefore conclude that in the splanchnics there are vaso-motor fibres of the constrictor type, and that impulses are constantly passing down them to maintain the normal tone of the vascular tract which they command. When the splanchnic nerves are stimulated, the adrenal glands are so affected that adrenalin passes out by the veins into the blood-stream. It is clear that if the quan tity thus liberated were sufficiently large and its liberation suffi- ciently prompt it might play a part in the rise of pressure (p. 661) which follows stimulation of the nerves, whether they are excited directly or in the normal course of events reflexly. But it has not been demonstrated that this is an effective factor. Dilator fibres (for the intestines and the kidney, for example) have also been discovered in the splanchnic nerves, although the constrictors predominate, and special methods have to be employed for the detection of the dilators. The same is true of the nerves of the extremities, which certainly contain vaso-dilator fibres in addition to vaso-constrictors, although the difficulty of demonstrating the presence of the former is fully as great as it is in the splanchnics. For the investigation is com- plicated by the fact that such nerves as the sciatic supply with vaso-motor fibres two leading tissues — skin and muscle; and these are not necessarily affected in the same direction or to the same extent by stimulation of their vaso-motor fibres. The vaso-con- strictors under ordinary conditions preponderate, so that section of the sciatic or the brachial is generally followed by flushing of the balls of the toes and rise of temperature of the feet, stimulation by paling and fall of temperature. By taking advantage, however, of the unequal excitability of dilators and constrictors in a degenerating nerve, and of the differences between the two kinds of fibres in their i?3 THE CIRCULATION OF THE BLOOD AND LYMPH reaction to electrical stimuli (p. 175), it has been shown that vaso- dilators are also present, and come to the front when the conditions are rendered favourable for them and unfavourable for the constrictors. Vaso-motor fibres for the fore-limb (dog) issue from the cord in the anterior roots of the third to the eleventh dorsal nerves, and for the hind-limb in the anterior roots of the eleventh dorsal to the third lumbar. Stimulation of most of these roots causes constriction of the vessels, but stimulation of the eleventh dorsal may cause dilatation (Bayliss and Bradford). The Vaso-Motor Nerves of Muscle. — When the motor nerve of the thin mylo-hyoid muscle of the frog, which can be observed under the micro- scope, is cut, and the peripheral end stimulated, the vessels are seen to dilate distinctly, and this effect is not abolished when contraction of the muscle is prevented by a dose of curara insufficient to paralyze the vaso-motor nerves. This indicates the existence in the nerve of vaso- dilator fibres. But we must be cautious in transferring this result to ordinary skeletal muscle, for the mylo-hyoid is more closely allied to the muscles of the tongue than, for example, fo the muscles of the limbs, and in the mammal the tongue muscles are known to be better supplied with vaso-dilator fibres than the limb muscles. The average flow of blood through a mammalian muscle is indeed increased during volun- tary contraction, and during rhythmically repeated artificial tetaniza- tion of its motor nerve. The outflow of blood from the main vein of the levator labii superioris, one of the muscles used in feeding in the horse, was found to be in one experiment nearly eight times, in another about seven times, and in a third three and a half times as great during voluntary work with it (in chewing) as in rest. But as no increase in the blood-flow through the skeletal muscles of a completely curarized mammal during excitation of their nerves has ever been satisfactorily demonstrated, we must conclude that they are very scantily provided with vaso-dilator fibres or not at all. It is uncertain whether they are supplied with vaso-constrictors. The undoubted increase in the blood- flow in contraction may therefore be connected in some way with the mcjchanical or chemical changes in the muscular fibres themselves. It has been suggested that the muscular vessels are widened by the direct action of the acid products of the active muscle, since very dilute acids (lactic acid, e.g.] cause general dilatation of the small vessels. -A similar explanation has been extended to the dilatation of the vessels of the brain during cerebral activity by some of those who deny the existence of vaso-motor nerves for that organ, but here the evidence is by no means satisfactory. The vagus has been stated to contain vaso-constrictor, and the annulus of Vieussens vaso-dilator, fibres for the coronary arteries of the heart. But this question is far from being settled. Adrenalin causes dilatation and not constriction of the coronary vessels. There is some reason to believe that the metabolic products liberated in the heart-muscle, e.g., carbon dioxide, govern the changes in the calibre of the coronary arterioles. A close relationship exists between the output of carbon dioxide and the rate of flow through the coronary circulation. In asphyxia the flow through the coronary vessels is notably increased; indeed, it is at its maximum just before the heart fails altogether, as if an effort were being made to keep the heart going to the last by making up to it in the quantity of the blood supplied what it lacks in quality. As this increased flow is seen in the isolated heart-lung preparation, it has been concluded that metabolites produced in the cardiac muscle itself cause an increased coronary flow when increased demands are made on the heart, a local regulative THE NERVOUS REGULATION OF THE BLOODVESSELS i-tg mechanism being thus constituted. There is some evidence that carbon dioxide is not the most potent of these substances. Vaso-Motor Nerves of the Lungs. — There has been much discussion as to the course , and even as to the existence , of vaso-motor fibres for the lungs. The problem is perhaps the most difficult in the whole range of vaso-motor topography, for the pulmonary circulation is so related to other vascular tracts, that changes produced in the vessels of distant organs by the stimulation or section of nerves may affect the quantity of blood received by the right side of the heart, and therefore the quantity propelled through the lungs and the pressure in the pulmonary artery. And changes in the systemic arterial pressure may favour or hinder the discharge of the left ventricle, and therefore affect the pres- sure in the left auricle and the pulmonary veins. Nevertheless, evidence has been obtained from a number of sources that the lungs are supplied with vaso-constrictor fibres. Plumier, perfusing isolated, ' surviving ' lungs with blood under constant pressure and measuring the outflow, showed that adrenalin and also stimulation of the annulus of Vieussens caused great diminution in the flow — that is, constriction of the vessels. Wiggers also obtained constriction with adrenalin. Fiihner and Starling, working with a preparation including the heart as well as the lungs, found that adrenalin caused a rise of pressure in the pulmonary artery coupled with a fall of pressure in the left auricle, which could only be due to constriction of the vessels of the lungs. It is assumed that adrenalin produces vaso-constriction only ia vessels supplied with vaso-constrictor nerves (p. 655), and that in organs where this substance does not cause vaso-constriction no such fibres are present. In mam- mals the vaso-constrictor fibres seem to pass out from the upper half of the dorsal spinal cord, and some of them can be detected nearer their destination in the annulus of Vieussens. The vago-sympathetic of the tortoise contains vaso-constrictors for the lung of the same side (Krogh). Vaso-Dilator Fibres. — In most of the peripheral nerves these are mingled with vaso-constrictors; but in certain situations, for an anatomical reason that will be mentioned presently, nerves exist in which the only vaso-motor fibres are of the dilator type. Of these, the most conspicuous examples are the chorda tympani and the nervi erigentes or pelvic nerves; and, indeed, it was in the chorda that vaso-dilators were first discovered by Bernard. The chorda tympani contains vaso-dilator and secretory fibres for the sub- maxillary and sublingual salivary glands. With the secretory fibres we have at present nothing to do; and the whole subject will have to be returned to, and more fully discussed in Chapter VI. But a most marked vascular change is produced by stimulation of the peripheral end of the divided chorda tympani nerve. The glands flush red; more blood is evidently passing through their vessels. Allowed to escape from a divided vein, the blood is seen to be of a bright arterial colour and shows a distinct pulse. The small arteries have been dilated by the action of the vaso-motor fibres in the nerve. The resistance being thus reduced, the blood passes in a fuller and more rapid stream through the capillaries into the veins, and on the way there is not time for it to become completely venous. These vaso-dilator fibres are not in constant action, for section of the l8o THE CIRCULATION OF THE BLOOD AND LVMPH nerve, as a rule, produces little or no change. Vaso-constrictor fibres pass to the salivary glands from the cervical sympathetic along the arteries, and stimulation of that nerve causes narrowing of the vessels and diminution of the blood-flow, sometimes almost to complete stoppage. The nervi erigentes are the nerves through which erection of the penis is caused. When they are divided there is no effect, but stimulation of the peripheral end causes dilatation of the vessels of the erectile tissue of the organ, which becomes overfilled with blood. During stimulation of these nerves, the quantity of blood flowing from the cut dorsal vein of the penis may be fifteen times greater than in the absence of stimulation. It spurts out in a strong stream, and is bi filter than ordinary venous blood (Eckhard). Stimulation of the peripheral end of the nervus pudendus causes constriction of the vessels of the penis, so that it contains vaso- constrictor fibres which are the antagonists of the nervi erigentes. Vaso-Motor Nerves of Veins. — Like arteries, veins have plexuses of nerve-fibres in their walls, and contract in response to various stimuli. In some cases — e.g., in the wing of the bat — rhythmical contractions of the veins are strikingly displayed, but they do not depend on the central nervous system, as they persist after section of the brachial nerves. The existence of vaso-constrictor fibres for the venules given off in the liver by the portal vein is indicated by the fact that adrenalin diminishes the blood flow through the organ even when the hepatic artery has been tied (Burton-Opitz ; Macleod and Pearce, etc.). Stimulation of the distal end of the hepatic plexus causes similar effects. The fibres issue from the spinal cord by the anterior roots of the third to the eleventh dorsal nerves, but chiefly in the fifth to the ninth dorsal. The arterioles arising from the hepatic artery have their own vaso-motor supply, which is more com- plete than that of the portal vessels. When the liver is enclosed in a plethysmograph, and the central end of an ordinary sensory nerve, like the sciatic, excited, reflex vaso-constriction takes place in the portal area, the volume of the organ diminishes, and the blood-pres- sure rises in the portal vein (Francois-Franck). The vena portae and its branches are in the physiological sense arteries rather than veins, since they break up into capillaries, and it was to be expected that the regulation of the blood-flow in them would be carried out in the same way as in ordinary arteries, namely, by means of vaso-motor nerves. But we must not, without special proof, extend the results obtained in the portal system to ordinary veins. A certain amount of evidence, however, exists that even such veins as those of the extremities are supplied, though scantily, with vaso-constrictor (veno-motor) fibres. After ligation of the crural artery or aorta, stimulation of the peripheral end of the sciatic has been seen to cause contraction of short portions of the superficial veins of the leg. THE NERVOUS REGULATION OF THE BLOODVESSELS 181 Finally, adrenalin (epinephrin) causes constriction of rings of ' surviving ' veins just as of artery rings, although in correspondence with the smaller amount of muscular tissue in the former the con- traction is not so strong. As adrenalin is assumed to act only upon muscle supplied by sympathetic nerve-fibres (p. 655), this would seem to indicate the existence of such a supply for veins. The question is an important one in connection with the regulation of the filling, and therefore of the discharge, of the heart (Henderson), but the experimental data are as yet too meagre to justify further discussion of the matter here. Course of the Vaso-Motor Nerves. — In the dog the vaso-constrictors pass out as fine medullated fibres (1-8 to 3-6 p. in diameter) in the anterior roots of the second dorsal to about the second lumbar nerves. They proceed by the white rami communicantes to the lateral sym- pathetic ganglia, where, or in more distal ganglia such as the inferior mesenteric, they lose their medulla, and their axis-cylinder processes (p. 851) break up into fibrils that come into close relation with the nerve-cells of the ganglia. These ganglion-cells in their turn send off axis-cylinder processes, which, enveloped by a neurilemma, pass as non-medullated fibres by various routes to their final destination, the unstriped muscular fibres of the bloodvessels. Their course to the head has been already described. To the limbs they are distributed in the great nerves (brachial plexus, sciatic, etc.), which they reach from the sympathetic ganglia by the grey rami communicantes. The outflow of vaso-dilator fibies is not restricted to the same portion of the cord from which the outflow of constrictor fibres takes place. Their existence is indeed most easily demonstrated in nerves springing from those regions of the cerebro-spinal axis from which vaso-constrictor fibres do not arise, and where, therefore, we have not to contend with the difficulty of interpreting mixed effects. Vaso-dilators for the external generative organs and the mucous membrane of the lower end of the rectum pass out as small medullated fibres of the anterior roots of certain of the sacral nerves (mainly the second and third in the cat) into the pelvic nerve (nervus erigens). They end in relation with ganglion-cells in the neighbourhood of the organs which they supply. The seventh and ninth cranial nerves carry vaso-dilator fibres which are distributed by way of the lingual and other branches of the fifth to the salivary glands, the tongue, the mucous membrane of the floor of the mouth, and part of the soft palate. Those in the lingual, passing through the chorda tympani, end in gangi ion-cells near the submaxillary and sublingual glands, and the axons of these cells continue the path to the vessels of the glands. It is supposed that the vaso-dilators dis- tributed in other branches of the fifth also have ganglion-cells on their course. In fact, there is good evidence that every efferent vaso-motor fibre is interrupted by one, and only by one, ganglion-cell between the cord and the bloodvessels. The statement has been made that for certain regions of the body, especially the skin of the limbs, the vaso- dilator nerves are contained, not in the anterior, but in the posterior roots. And these, it is claimed, are not aberrant efferent fibres which have strayed in the course of development into the wrong roots, but true posterior root-fibres whose cells of origin lie in the spinal ganglia, and which conduct efferent vaso-dilator impulses in the wrong direction, so to speak, from the cord to the periphery — ' antidromic ' impulses (Bayliss). 1 82 THE CIRCULATION OF THE BLOOD AND LYMPH Effect of Nicotine on Nerve-Cells. — A method which has been found most fruitful in studying the relations of sympathetic ganglion-cells to the vaso-motor fibres, as well as to the pilo-motor* and secretory fibres which in certain situations are so intricately mingled with them, must here be mentioned. It depends upon the fact that when a suitable dose of nicotine (10 milligrammes in a cat) is injected into a vein, or a solution is painted on a ganglion with a brush, the passage of nerve- impulses through the ganglion is blocked for a time (Langley). The nerve-fibres peripheral to the ganglion are not affected. The question whether efferent fibres are connected with nerve-cells between a given point and their peripheral distribution can, therefore, be answered by observing whether any effect of stimulation is abolished by nicotine. If, for instance, the excitation of a nerve caused constriction of certain bloodvessels before, and has no effect after, the application of nicotine to a ganglion, its vaso-constrictor fibres, or some of them, must be con- nected with nerve-cells in that ganglion. Langley has brought forward evidence that many of the bodies which are commonly supposed to act upon nerve-endings (as nicotine, curara, atropine, pilocarpine, adrenalin, etc.) really act upon ' receptive ' substances of the cells in connection with which the nerve-fibres end. These receptive substances are con- ceived to be capable of being specifically affected by chemical bodies and by nervous stimuli, and in their turn to be capable of influencing the metabolism of the main cell substance on which its function depends. The receptive substances thus form beyond the histological link of the nerve-ending a kind of chemical link between the nerve-fibre and the cell which it supplies. We have thus traced the vaso-motor nerves from the cerebro- spinal axis to the bloodvessels which they control ; it still remains to define the portion of the central nervous system to which these scattered threads are related, which holds them in its hand and acts upon them as the needs of the organism may require. Vaso-Motur Centres. — Now, experiment has shown that there is one very definite region of the spinal bulb which has a most intimate relation to the vaso-motor nerves. If while the blood-pressure in the carotid is being registered, say, in a curarized rabbit, the central end of a peripheral nerve like the sciatic is stimulated, the pressure rises so long as the bulb is intact, this rise being largely due to the reflex constriction of the vessels in the splanchnic area. If a series of transverse sections be made through the brain, the rise of pressure caused by stimulation of the sciatic is not affected till the upper limit of the bulb is almost reached. If the slicing is still carried downwards, the blood-pressure sinks, and the rise following stimu- lation of the sciatic becomes less and less. When the medulla has been cut away to a certain level, only an insignificant rise or none at all can be obtained. The portion of the medulla the removal of which exerts an influence on the blood-pressure, and its increase by reflex stimulation, extends from a level 4 to 5 mm. above the point of the calamus scriptorius to within I to 2 mm. of the corpora quadrigemina. Stimulation of the medulla causes a rise, destruc- * Pilo-motor nerves supply the smooth arrector pili muscles, whose contrac- tion causes the hair to ' stand on end.' THE NERVOUS REGULATION OF THE BLOODVESSELS 183 tion of this portion of it a severe fall, of general blood-pressure. There is evidently in this region a nervous ' centre ' so intimately related, if not to all the vaso-motor nerves, at least to such very important tracts as to deserve the name of a vaso-motor centre. Experiment has shown that this is much the most influential centre, and it is usually called the chief or general vaso-motor centre. Some writers prefer to speak of it as the vaso-constrictor centre, since it is undoubtedly connected with most or all of the vaso-constrictor paths, and has not been shown to be similarly connected with the vaso-dilator paths. There is, indeed, not the same solid evidence for the existence of a general vaso-dilator centre in the bulb as for the existence of the general vaso-constrictor centre. Yet there are facts which indicate that the bulbar vaso-motor centre or centres, when reflexly stimulated, can, and often do, respond not merely by an increase or a remission of vaso-constrictor tone, but by a simul- taneous inhibition of vaso-constrictor fibres and excitation of vaso- dilators leading to a fall of pressure, or by a simultaneous inhibition of vaso-dilators and excitation of vaso-constrictors leading to a rise of pressure. The spinal cells of origin of the pre-ganglionic segments of the vaso-constrictor paths constitute subordinate centres which either normally support a certain degree of vascular tone, or come to do so after the chief vaso-motor centre has been cut off. Thus, in the frog it is possible to go on destroying more and more of the cord from above downwards, and still to obtain reflex vaso- motor effects, as seen in the vessels of the web, by stimulating the central end of the sciatic nerve. Although these effects indeed diminish in amount as the destruction of the cord proceeds, yet a distinct change can be caused when only a small portion of the cord remains intact. Similarly, in the mammal evidence has been obtained of the existence of ' centres ' at various levels of the cord, capable of acting eventually, if not at once, as vaso-constrictor centres after the loss of the controlling influence of the bulb. The best example of a vaso-dilator centre is that situated in the lumbar cord, which controls the erection of the penis. After total section of the cord at the upper limit of the lumbar region, erection, which is known to be due to a reflex dilatation of the arteries of the organ through the nervi eri- gentes, can still be caused (in dogs) by mechanical stimulation of the glans penis, so long as the afferent fibres of the reflex arc con- tained in the nervus pudendus are intact. Destruction of the lumbar cord abolishes the effect. It is impossible to avoid the conclusion that a vaso-dilator or erection centre, which is in relation on the one hand with the nervi erigentes, and on the other with the nervus pudendus, exists in the lower portion of the spinal cord. Vaso- motor centres for the hind-limbs have also been located in the i84 THE CIRCULATION OF THE BLOOD AND LYMPH same region. When the brain, the bulb, and the upper portion of the cord have been eliminated by ligation of all the arteries from which blood can possibly reach them, a sufficient vascular pressure persists to permit the circulation to go on in the lo\ver portion of the body for hours. And while section — or freezing (Fig. 80) — of the cord in the lower cervical region causes a marked fall of pressure, this r, not permanent if the animal is allowed to survive. Forty-one days after total section of the cord at the seventh cervical segment in a dog an arterial pressure of 130 mm. of mercury was found. A mechanism for the maintenance of vascular tone exists even beyond the limits of the central nervous system. For when the lower portion of the cord is completely destroyed, the dilatation of the vessels of the hind-limbs, which is at first so conspicuous, passes away after a time, the functions of vaso-motor centres having perhaps been assumed by the sympathetic ganglia (Goltz and Ewald). When the lumbo-sacral sympathetic chain is extirpated, Fig. 80. — Effect on Blood- Pressure of Freezing Spinal Cord (Pike). At i the first or second dorsal segment of a dog's cord was frozen with liquid air; at 2 and 3 central end of sciatic stimulated without effect on pressure (respectively one and a half and three minutes after freezing of cord). (Four-fifths of original size.) there is a further loss of vascular tone in the affected region. But even this is not irremediable. After a time recovery again occurs, although it may be more partial and tardy than before. This may take place either through the intervention of still more peripheral ganglia, or through the development of a certain tonus by the muscular fibres of the vessels when abandoned to themselves. As to the nature of the tone of the general vaso-motor centre, the same question may be asked which has been already discussed for the cardio-inhibitory centre. Is it reflex, or does it depend upon direct excitation of the centre by some constituent of the blood or lymph, or some substance produced in the centre itself ? The best answer which can at present be made is that a constant central excitation by the carbon dioxide formed in the centre or circulating in the blood is a not unimportant factor in the maintenance of the vaso-motor tone. A marked diminution in the carbon dioxide tension of the blood, a condition which is termed ' acapnia,' may indeed contribute to the severe fall of blood-pressure associated with THE NERVOUS REGULATION OF THE BLOODVESSELS 185 surgical shock (Henderson). In addition to the direct influence of carbon dioxide, and possibly of other substances, the arrival of afferent impulses at the centre seems to play a part in maintaining that continual discharge of efferent impulses along the vaso-motor nerves which constitutes its tone. In this regard, the vaso-motor centre occupies an intermediate position between the respiratory centre, the most purely automatic, and the cardio-inhibitory centre, the most purely reflex of the three great bulbar mechanisms. Of the anatomical relations of the nerve-cells that make up the bulbar and spinal vaso-motor centres, little more is known than may be deduced from the physiological facts we have been reciting. It has been surmised on histological grounds that certain cells of small size scattered up and down the thoracic and upper lumbar regions of the cord in the lateral horn (intermedio-lateral tract), and perhaps cropping out also in the bulb, are vaso-motor cells. There is good evidence that the pre-ganglionic sympathetic fibres, including the vaso-motor fibres which we have already discovered emerging from the cord in the spinal roots, are connected with these cells. And, indeed, there is reason to believe that the connection is made without the intervention of any other nerve-cells, and that the axis-cylinders of these vaso-motor fibres are the axis-cylinder processes of the vaso-motor cells. So that the simplest efferent path along which vaso-motor impulses can pass may be considered as built up of two neurons, one with its cell-body in the cord, and the other in a sympathetic ganglion. Less is known of the elements which constitute the bulbar centre and of their connections. But since it would appear that the spinal vaso-motor centres are under the control of the chief centre in the bulb, it is necessary to suppose that the axis-cylinder processes of some of the cells of the bulbar centre come into relation with the spinal vaso-motor cells, and that impulses passing, let us say, from the bulb to the vessels of the leg, would have to traverse three neurons (p. 852). Vaso-Motor Reflexes. — We have already seen that the cardiac centres are constantly influenced by afferent impulses, and that in the direction either of augmentation or inhibition. The vaso-motor centre in the bulb is equally sensitive to such impulses. They reach it for the most part along the same nerves, and by increasing or diminishing its tone cause sometimes constriction and sometimes dilatation of the vessels, the result depending partly upon the ana- tomical connection of the afferent fibres, but apparently in part also upon the state of the centre. Of the afferent nerves that cause vaso-dilatation, the most im- portant is the depressor, whose reflex inhibitory action on the heart has been already described. The fall in the arterial pressure is due chiefly, not to the inhibition of the heart, but to inhibition of the vaso-constrictor tone of the bulbar vaso-motor centre, combined with stimulation of vaso-dilator nerves, and consequent general dila- tation of the arterioles throughout the body. That' the depressor action involves excitation of vaso-dilators follows from the fact that vaso-dilatation occurs in the limbs on stimulation of the depressor after their vaso-constrictor nerves have been cut. Stimulation of the central end of the depressor may also cause dilatation of the 1 86 THE CIRCULATION OF THE BLOOD AND LYMPH vessels of the submaxillary gland even on the opposite side, whether the sympathetic has been divided or not, so long as the chorda tympani is intact, and this dilatation is not accompanied by a flow of saliva. Stimulation of the depressor produces its usual result after section of the vagi. It has been suggested that the function of the nerve is to act as an automatic check upon the blood-pressure in the interest both of the heart and the vessels, its terminations in the aorta or the ventricular wall being mechanically stimulated when the pressure tends to rise towards the danger limit. In rare cases, efferent inhibitory fibres for the heart have been found in the depres- sor of the rabbit. Many of the peripheral ' nerves contain fibres whose stimulation is fol- lowed by dilatation of the Fig. 81. — Diagram of Depressor in Rabbit. X, vagus ; SL, superior laryngeal ; D, de- pressor. Arrows show course of impulses. Fig.82. — Blood- Pressure Tracing: Rabbit. Central end of depressor stimulated at i ; stimulation stopped at 2. Time- trace, seconds. bloodvessels in special regions, usually the areas to which they are themselves distributed, accompanied by constriction of distant and, it may be, more extensive vascular tracts. Thus, the usual local effect of stimulating the afferent fibres of the lowest three thoracic nerves, in whose anterior roots run the vaso-motor fibres for the kidney, is a dilatation of the renal vessels (Bradford) , and the usual local effect of stimulating the infra-orbital or supra-orbital nerve a dilatation of the external maxillary artery. But the general effect in both cases is vaso-constriction in other regions of the body, which more than compensates the local dilatation, so that the arterial blood- pressure rises. It is not difficult to see that both of these changes render it easier for the part to obtain an increased supply of blood. Sometirres the reciprocal relation between vaso-dilatation in one part of the body and vaso-constriction in another is only apparent. For example, stimulation of the cut end of the sciatic causes, as we have already seen, extensive vaso-constriction and a notable rise in the blood- pressure. The constriction certainly involves the splanchnic area; but THE NERVOUS REGULATION OF THE BLOODVESSELS i87 superficial parts, as the lips, may be seen to be flushed with blood. In asphyxia, when the vaso-motor centres are directly stimulated by the venous blood, this apparent antagonism is still better marked : the cutaneous vessels are widely dilated and engorged, the face is livid, but the abdominal organs are pale and bloodless (Heidenhain) . The blood-pressure rises rapidly, reaches a maximum, and then gradually falls as the vaso-motor centre becomes paralyzed (Figs. 84 and 85). It has been shown that in both cases vaso-constriction ot the skin is really produced as well as vaso-constriction of the internal organs, but the increased blood-pressure mechanically overcomes the constriction of the cutaneous vessels. The kind of stimulus seems to have something to do with the direction of the reflex vaso-motor change. For while electrical stimulation of every muscular nerve, even of the very finest twigs that can be isolated and laid on electrodes, provokes always, whether the shocks follow each other rapidly or slowly, a rise of general Fig. 83. — Pressor Effect of Stimulation cf Central End of Vagus in a Cat during Resuscitation after Cerebral Anaemia. The depressions in the signal line ABC indicate the duration of three successive excitations of equal strength, sixty-five, seventy-three, and seventy-nine minutes respectively after restoration of the circulation. The pressor effect increases as resuscitation proceeds. Later on the original depressor effect was again obtained. The upper tracing is that of the artificial respiration. (Two-thirds original size.) blood-pressure, mechanical stimulation of a muscle, as by kneading or massage, causes a fall. The condition of the afferent fibres also exerts an influence. For example, excitation of the central end of a sciatic nerve that has been cooled is followed by vaso-dilatation and fall of pressure, the opposite of the ordinary result. These and similar facts have led to the idea that most afferent nerves contain two kinds of fibres, whose stimulation can affect the activity of the vaso-motor centres — ' reflex vaso-constrictor,' or ' pressor ' fibres, and ' reflex vaso-dilator/ or ' depressor ' fibres. The branch of the vagus, however, to which the name ' depressor * has been specially given is usually described as the only peripheral nerve'the excitation of which is in all circumstances followed by a general diminution of arterial pressure. But this is not strictly correct, for at an early period in the resuscitation of the brain after anaemia excitation of 188 THE CIRCULATION OF THE BLOOD AND LYMPH the rabbit's ' depressor ' causes a slight rise of pressure not followed by any fall. This, perhaps, indicates the presence in the ' depressor ' of a small number of pressor fibres, which are resuscitated sooner than the depressor fibres proper. The same phenomenon, only more marked, may be seen when the central end of the cat's vagus, containing the depressor fibres, is excited at intervals during resus- citation (Fig. 83). Or the result may depend upon a change in the response of the altered vaso-motor centres to impulses reaching them along the depressor fibres. If specific ' depressor ' fibres exist in other nerves, they are so mingled with ' pressor ' fibres that their action is masked when both are stimulated together. The state of the vaso-motor centre is unquestionably a factor which has some importance in determining the result of reflex vaso-motor stimula- tion. For instance, in an animal deeply anaesthetized with chloro- form or chloral, excitation of pressor fibres (in an ordinary sensory Fig. 84. — Rise of Blood- Pressure in Asphyxia : Rabbit. Respiration stopped at i. Interval between 2 and 3 (not reproduced) 44 seconds, during which the blood - pressure steadily rose. At 4, respiration resumed. Time-trace, seconds. nerve) causes, not a rise, but a fall of blood-pressure; while in an animal fully under the influence of strychnine stimulation of the depressor nerve causes not a fall, but a rise. The vaso-motor reflexes in man can be conveniently studied by the calorimetric method described on p. 221. One of the most important of the vaso-motor reactions is that by which the vessels of the skin respond to the temperature of the environment so as to regulate the loss of heat from the body (p. 699). When one hand, e.g. the left, is immersed in cold water (say at about 8° C.), the blood-flow in the right is at once reduced owing to reflex vaso- constriction. Other parts of the body are also affected, but not so readily as the contra-lateral hand, since the segments of the cord into which the afferent fibres from a given skin area run are at the same time the segments from which the efferent vaso-motor fibres for the symmetrically-placed area on the opposite side of the body arise. The reflex diminution in the flow persists for a time which THE NERVOUS REGULATION OF THE BLOODVESSELS t8g varies with the individual, the external temperature, and other circumstances, and then as a rule rather suddenly the vaso-con- striction gives way and the flow begins again to increase, even while the left hand is still kept in the cold water. When the left hand is transferred from the cold to warm water (at 43° or 44° C.), the first effect is a transient diminution in the blood-flow in the right hand. This soon gives place to an increase (vaso- dilatation). As an ex- ample, the following table gives the condensed results of three experiments on two young men. Experiments II. and III. were on the same man at an interval of three days. Experi- ment. Temperature of — Duration of Observation in Minutes. Flow in Grms. per too c.c. of Right Hand per Minute. Left Hand in — Room. Arterial Blood. Calorim. (28-9} (3i-i 16 12-2 _ I. J 29-0 1 I 29-0 f 36-6 J3I-7 J3I-2 4 6 6-9 9-9 Cold water. Cold water still. U9'iJ 132-3 II 11-6 Warm water. [24--9 13 IO-I — 24-2 3i-3 13 5-o Cold water. II. - 23-9 I 23-9 U3-9; 36-0 131-4 31-5 V3I-7 2 3 7 3-4 8-4 15-0 Warm water. Warm water still. Warm water still. (24*1 (3i-6 15 12-4 — 24-4 32-1 5 5-9 Cold water. ill. l 24'4 [ 36-5 S32-2 5 10-7 Cold water still. 24-4 32-4 3 7-9 Warm water. 124-5; V32-6 7 17-6 Warm water still. Such facts enable us to some extent to understand the manner in which the distribution of the blood is adjusted to the requirements of the different parts of the body, so that to a certain degree of approximation no organ has too much, and none too little. The blood-supply of the organs is always shifting with the calls upon them. Now, it is the actively-digesting stomach and the actively- secreting glands of the alimentary tract which must be fed with a full stream of blood, to supply waste and to carry away absorbed nutriment. Again, it is the working muscles of the legs or of the arms that need the chief blood-supply. But wherever the call may be, the vaso-motor mechanism is able, in health, to answer it by bringing about a widening of the small arteries of the part which needs more blood, and a compensatory narrowing of the vessels of other parts whose needs are not so great. It is also through the vaso-motor system, and especially by the action of that portion of it which governs the abdominal vessels, and too THE CIRCULATION OF THE BLOOD AND LYMPH of the nerves that regulate the work of the heart, that in animals to which the upright position is normal (monkey) and in man the influence of changes of posture on the circulation is almost com- pletely compensated.* The pressure in the upper part of the human brachial artery has been measured with a sphygmoman- ometer, first in the horizontal and then immediately afterwards in the standing posture, and in health it has been found to remain practically unchanged (Hill). But if the person was overworked or out of sorts, the compensation was less complete. It is well known that in debilitated persons, especially if long confined to bed, the sudden assumption of the upright position may cause vertigo, and even syncope, the normal compensatory mechanism being deranged. In such animals as the rabbit this compensation is totally inefficient. When a domesticated rabbit, which has been kept in a hutch, is suspended vertically with the feet down, the blood drains into the abdominal vessels, syncope speedily ensues, and in a period that ranges from less than a quarter to three-quarters of an hour the animal dies in the convulsions of acute cerebral anaemia (Salathe, Hill). The head- down position has no ill-effects. In wild rabbits, whose abdominal wall is more tense and elastic, these fatal symp- toms are not easily produced, and the same is true of cats and dogs. But in all animals, when the compensation is destroyed, as in paralysis of the vaso-motor centre by chloroform, the circulation may be profoundly influenced by the position of the body : elevation of the head may lead to cerebral anaemia, syncope, and even death ; elevation of the legs, and particularly the abdomen, may restore the sinking pulse by filling the heart and the vessels of the brain. If a chloralized dog be fastened on a board which can be rotated about a horizontal axis passing under the neck, the blood-pressure in the carotid artery falls greatly when the animal is made to assume the vertical position with the head up, and either rises a little or remains practically unchanged when the head is made to hang down. So great may the fall of pressure be in the former position that death may occur if it be long maintained (Practical Exercises, p. 213). * Two factors may be distinguished in the blood-pressure, the hydrostatic and the hydrodynamic elements. The hydrostatic portion of the pressure is due to the weight of the column of blood acting on the vessel; the hydro- dynamic portion of the pressure is due to the work of the heart. If a dog be securely fastened to a holder arranged in such a way that the animal can be placed vertically, with the head up or down, and the mean blood-pressure in the crural artery be measured in the two positions, there will be a considerable difference. For when the legs are uppermost the heart has to overcome the weight of the column of blood rising above it to the crural artery; when the head is uppermost the action of the heart is reinforced by the weight of the blood. And if no change were produced in the action of the heart, or in the general resistance of the vascular path, by the change of position, this differ- ence would be equal to the pressure of a column of blood twice as high as the straight-line distance between the cannula and the point of the arterial system at which the pressure is the same with head up as with head down (indifferent point). THE XERVOUS REGULATION OF THE BLOODVESSELS 191 Finally, it is in virtue of the amaxing power of accommodation possessed by the vascular system, as controlled by the vaso-motor and cardiac nerves, that so long as these are not disabled the total quantity of blood may be greatly diminished or greatly increased, without endangering life, or even causing more than a transient alteration in the arterial pressure. It is not until at least a quarter of the blood has been withdrawn that there is any notable effect on the pressure, for the loss is quickly compensated by an increase in the activity of the heart and a constriction of the small arteries. An animal may recover after losing considerably more than half its blood.* Conversely, the volume of the circulating liquid may be doubled by the injection of blood or physiological salt solution without causing death, and increased by 50 per cent, without any marked increase in the pressure. The excess is promptly stoweH 1 n 1 1 » 1 1 1 1 n i 1 1 1 1 1 1 1 F-g. 85. — Blood- Pressure Tracing from a Dog poisoned with Alcohol. Therespiratory centre beingparalyzed, respiration stopped, and the typical rise of blood-pressure in asphyxia took place. The pressure had again fallen, and total paralysis of the vaso-motor centre was near at hand, when at A the animal made a single respiratory movement. The quantity of oxgyen thus taken in was enough to restore the vaso-motor centre, and the blood-pressure again rose. repeated five or six times. (Three- fourths original size.) away in the dilated vessels, especially those of the splanchnic area; the water passes rapidly into the lymph, and is then more gradually eliminated by the kidneys. From these facts we can deduce the practical lesson, that blood- letting, unless fairly copious, is useless as a means of lowering the general arterial pressure, while it need not be feared that transfusion of a considerable quantity of blood, or of salt solution, in cases of severe haemorrhage, will dangerously increase the pressure. And from the physiological point of view the term ' haemorrhage ' includes more than it does in its ordinary sense. For as dirt to the sani- tarian is ' matter in the wrong place,' haemorrhage to the physiolo- gist is blood in the wrong place. Not a drop of blood may be lost from the body, and yet death may occur from haemorrhage into the pleural or the abdominal cavity, into the stomach or intestines. Not only so, but a man may bleed to death into his own blood- * It is not usually possible to obtain quite two-thirds of the total blood by bleeding a dog from a large artery. In seven dogs bled from the carotid, the ratio of the weight of the blood obtained to the body-weight was i : 24-7. i : zi"j. i : 22-7. i : 20-6, i : 18-6. i : 16, i : 13-5. In the last case, the blood clotte I with abnormal slo .vnes ;. and the animal died in a few minutes. IQi THE CIRCULATION OF TH£ BLOOD A\'D LYMPH vessels ; in surgical (also sometimes called traumatic or vascular) shock, it would appear that the blood which ought to be circulating through the brain, heart, and lungs may stagnate in the dilated veins. The essential nature of ' vascular ' shock, which should be dis- tinguished from the spinal shock described in Chapter XVI. is so little understood, in spite of the large amount of work devoted to the subject, that it would not be profitable to discuss it here. It may be remarked, however, that there is no reason to suppose that ex- haustion orlossof sensitiveness of the vaso-mo tor centre is concerned. While an undue proportion of the blood is accumulated in the great veins, the arterioles and capillaries of the splanchnic area, like those of the skin, are contracted and contain less blood than normal (Lyon, Janeway and Jackson, etc.). SECTION VII. — THE LYMPHATIC CIRCULATION. As has already been stated, some of the constituents of the blood, instead of passing back to the heart from the capillaries along the veins, find their way by a much more tedious route along the lymphatics. The blood capillaries are everywhere in very intimate relation with lymph capillaries, which, completely lined with epithelioid cells, lie in irregular spaces in the connective-tissue that everywhere accompanies and supports the bloodvessels. The constituents of the blood-plasma are filtered through, or secreted by the capillary walls into these lymph spaces, and mingling there with waste products discharged by the cells of the tissues, form the liquid known as tissue liquid or tissue lymph. From the tissue liquid the lymph capillaries take up the constituents of the ' lymphatic ' lymph, which then passes into larger lymphatic vessels, with lymphatic glands at intervals on their course. These fall into still larger trunks, and finally the greater part of the lymph reaches the blood again by the thoracic duct, which opens into the venous system at the junction of the left subclavian and internal jugular veins. The lymph from the right side of the head and neck, the right extremity, and the right side of the thorax, with its viscera, is collected by the right lymphatic duct, which opens at the junction of the right sub- clavian and internal jugular veins. The openings of both ducts are guarded by semilunar valves, which prevent the reflux of blood from the veins. Serous cavities like the pleural sacs, although differing from ordinary lymph spaces, are connected through small opening', called stomata, with lymphatic vessels. The rate of flow of the lymph in the thoracic duct is very small com- pared with that of the blood in the arteries — only about 4 mm. per second, according to one observer. Nevertheless, a substance injected into the blood can be detected in the lymph of the duct in four to seven minutes (Tschirwinsky) . The factors which contribute to the main- tenance of the lymph flow are : (1) The pressure under which it passes from the blood capillaries into the lymph spaces and from the lymph spaces into the lymph capillaries. The pressure in the thoracic duct of a horse may be as high as 12 mm. of mercury; in the dog it may be less than i mm. The difference is probably due, in part at least, to a difference in the experimental con- ditions, dogs being usually anaesthetized for such measurements, horses not. The pressure in the lymph capillaries must, of course, be higher than in the thoracic duct — how much higher we do not know. (2) The contraction of muscles increases the pressure of the lymph by compressing the channels in which it is contained, and the valves 193 with which the lymphatics are even more richly provided than the veins, hinder a backward and favour an onward flow. The contractions of the intestines, and especially of the villi, aid the movement of the chyle. By the contraction of the diaphragm, substances may be sucked from the peritoneal cavity into the lymphatics of its central tendon, through the stomata in the serous layer with which its lower surface is clad. It is even possible by passive movements of the diaphragm in a dead rabbit to inject its lymphatics with a coloured liquid placed on its peritoneal surface. Passive movements of the limbs and massage of the muscles are also known to hasten the sluggish current of the lymph, and are sometimes employed with this object in the treatment of disease. (3) The movements of respiration aid the flow. At every inspiration the pressure in the great veins near the heart becomes negative, and lymph is sucked into them (p. 226). (4) In some animals rhythmically - contracting muscular sacs or hearts exist on the course of the lymphatic circulation. The frog has two pairs, an anterior and a posterior, of these lymph hearts, which pulsate, although not with any great regularity, at an average rate ol sixty to seventy beats a minute, and are governed by motor and inhibi- tory centres situated in the spinal cord. The beat is not directly ini- tiated from the cord, but the tonic influence of the cord is necessary in order that the lymph hearts may continue to beat (Tschermak). Such hearts are also found in reptiles. It is possible that in animals without localized lymph hearts the smooth muscle, which is so conspicuous an element in the walls of the lymphatic vessels, may aid the flow by rhythmical contractions. PRACTICAL EXERCISES ON CHAPTER III. i. Microscopic Examination of the Circulating Blood.— (i) Take a tadpole and lay it on a glass slide. Cover the tail with a large cover- slip, and examine it with the low power (Leitz, oc. III., obj-. 3). Generally the tail will stick so closely to the slide, and the animal will move so little, that a sufficiently good view of the circulation can be obtained. If there is any trouble, destroy the brain with a needle. Observe the current of the blood in the arteries, capillaries and veins. An artery may be easily distinguished from a vein by looking for a place at which the vessel bifurcates. In veins the blood flows in the two branches of the fork towards the point of bifurcation, in arteries away from it. Sketch a part of a field. To Pith a Frog. — Wrap the animal in a towel, bend the head forwards with the index-finger of one hand, feel with the other for the depression at the junction of the head and backbone, and push a narrow-bladed knife right down in the middle line. The spinal cord will thus be divided with little bleeding. Now push into the cavity of the skull a piece of pointed lucifer match. The brain will thus be destroyed. The spinal cord can be destroyed by passing a blunt needle down inside the vertebral canal. (2) Take a frog and pith its brain only, inserting a match to prevent bleeding. Pin the frog on a plate of cork into one end of which a glass slide has been fastened with sealing-wax. Lay the web of one of the hind-legs on the glass and gently separate two of the toes, if necessary by threads attached to them and secured to the cork plate. Put the plate on the microscope -stage and fasten by the clips (see pp. 15, 118). (3) After the normal circulation has been studied thoroughly put a very small drop of tincture of cantharides on the portion of the web 13 194 THE CIRCULATION OF THE BLOOD AND LVMPti which is in the field of the microscope, using a fine pipette. Observe the process of inflammation, including stasis and diapedesis (p. 61). 2. Anatomy of the Frog's Heart. — Expose the heart of a pithed frog by pinching up the skin over the abdomen in the middle line, dividing it with scissors up to the lower jaw, and then cutting through the abdominal muscles and the bony pectoral girdle. The external ab- dominal vein, which will be observed on reflecting the skin, can be easily avoided. The heart ^yill now be seen enclosed in a thin mem- brane, the pericardium, which should be grasped with fine-pointed forceps and freely divided. Connecting the posterior surface of the heart and the pericardium is a slender band of connective tissue, the fraenum. A silk ligature may be passed round this with a threaded curved needle, or curved fine-pointed forceps; and tied, and then the fraenum may be divided posterior to the ligature. The anatomical arrangement of the various parts of the heart should now be studied. Note the single ventricle with the bulbus arteriosus, the two auricles, and the sinus venosus, turning the heart over to see the latter by means of the ligature. Observe the whitish crescent at the junction of the sinus venosus and the right auricle (Fig. 86). 3. The Beat of the Heart. — Note that the auricles beat first, and then the ventricle. The ventricle becomes smaller and paler during its systole, and blushes red during diastole. Count the number of beats of the heart in a minute. Now excise the heart, lifting it by means of the ligature, and tak- ing care to cut wide of the sinus venosus. Place the heart in a small porce- lain capsule on a little blotting - paper moist - ened with physiological salt solution.* Observe that it goes on beating. Put a little ice or snow in contact with the heart and count the number of beats in a minute. The rate is greatly dimin- ished. Now remove the ice and blotting-paper, cover the heart with the salt solution, and heat, noting the temperature with a thermometer. Observe that the heart beats faster and faster as the temperature rises. At 40° to 43° C. it stops beating in diastole (heat standstill). Now at once pour off the heated liquid, and run in some cold salt solution. The heart will begin to beat again. 4. Cut off the apex of the ventricle a little below the auriculo- ventricular groove. The auricles, with the attached portions of the ventricle, go on beating. The apex does not contract spontaneously, but can be made to beat by stimulating it mechanically (by pricking it with a needle) or electrically. Divide the still contracting portion of the heart by a longitudinal incision. The two halves go on beating. 5. Heart Tracings. — (i) Fasten a myograph-plate (Fig. 87) on a stand. Take a long light lever, consisting of a straw or a piece of * For frog's tissues this should be 0-7 to 0-75 per cent, sodium chloride solution, for mammalian tissues a little stronger (about 0-9 per aent.). Fig. 86. — Frog's Heart with Stannius* ligatures in Position (Cyon). Anterior surface of heart shown on the left, posterior surface on the right, a, right auricle; 6, left auricle; c, ventricle; d, bulbus arte- riosus; e, f, aortae; g, sinus venosus. PRACTICAL EXERCISES 195 Fig. 87. — Arrangement for obtaining a Heart Tracing from a Frog. thin chip, armed at one end with a writing-point of parchment-paper, supported near the other end by a horizontal axis, and pierced not far from the axis by a needle carrying on its point a small piece of cork or a ball of sealing-wax. A counterpoise is adjusted on the short arm of the lever in the form of a small leaden weight. Cover a drum with glazed paper and smoke it. The paper must be put on so tightly that it will not slip. To smoke the drum, hold it by the spindle in both hands over a fish-tail burner, depress the drum in the flame, and rotate rapidly. The speed of the drum can be varied by putting in or taking out a small vane. Arrange an electro-magnetic time- marker for writing seconds (Fig. 88). Pith a frog" (brain only), expose the heart, and put under it a cover-slip to give it support. Pin the frog on the myograph- plate, and adjust the foot of the lever so that it rests on the ven- tricle or the auriculo-ventricular junction. Bring the writing-point of the lever and that of the time- marker vertically under each other on the surface of the drum. Set off the drum at the slow speed (say, a centimetre a second) . When the lever rests on the auriculo-ventricular junction, the part of the tracing corresponding to the contraction of the heart will be broken into two portions, representing the systole of the auri- cles and ventricle re- spectively. Cut the paper off the drum with a knife (keeping the back of the knife to the drum to avoid scoring it) and carry it to the varnishing- trough, holding the tracing by the ends with both hands, smoked side up. Im- merse the middle of it in the varnish, draw first one end and then the other through the varnish, let it drip for a minute into the trough, and fasten it up with a pin to dry. Fig. 88. — Electro - Magnetic Time - Marker connected with Metronome. The pendulum of the metro- nome carries a wire which closes the circuit when it dips into either of the mercury cups, Hg. (2) Heart Tracing, with Simultaneous Re- cord of Auricular and Ventricular Contractions. — (a) For this purpose two levers may be arranged, one resting on the auricle, the other on the ven- tricle, the writing-points being placed in the same vertical straight line on the drum. A convenient form of apparatus is shown in Fig. 89. (b) Gaskeirs Method (a modification of). — Attach a silk ligature to the very apex of the ventricle. Divide the frsenum, cut the aorta 196 THE CIRCULATION OF THE BLOOD AND LYMPH across close to the bulbus, pinch up a tiny portion of the auricle and ligature it. Remove the intestines, liver, lungs, etc., care being taken in cutting away the liver not to injure the sinus. Then remove the lower jaw, and cut away the whole of the body except the head, part of the oesophagus, and the tissue connecting it with the heart. Fix the head in a clamp sliding on an ordinary stand. The heart is held at the auriculo-ventricular junction in a Gaskell's clamp supported on a separate stand. The thread connected with the ventricle is brought round a pulley and attached to a lever above the heart. The auricle is connected with another lever. The writing-points of the two levers are arranged in a vertical line on the drum. The small pulley must be oiled from time to time to lessen the friction (Fig. 90). If tortoises or turtles are available, the much larger heart of these animals may be used for Experiments 5 (2) (a) and (&). The animal having been killed by cutting off its head, the ventral portion of the carapace is detachc d by the saw. The pericardium can now be slit open, and the pads of the levers arranged on auricles and ventricle fad fo rest an Aurtcte / . Pad to rest on Ventrtcte Fig. 89. — Apparatus for obtaining a Simultaneous Tracing of Auricular and Ventricular Contractions. respectively, as in Experiment 5 (2) (a), without further disturbing the heart. Or the heart may be removed, together with the upper portion of the body, the pericardium opened, and the liver cut away.1 The aortic trunk is then divided, and the portion of it attached to the heart grasped by a small forceps clamp. Fine silk ligatures are: attached to the apex of the ventricle and the top of the right auricle.. The vagus nerves are exposed in the neck, ligated, and divided. The upper portion of the body is supported on a stand. The forceps grasp- ing the aorta is fixed in an ordinary holder, and the threads are attached to the levers, as in Experiment 5 (2) (b). With the vagi, Experiment 7 may be performed. It must be remem- bered that the activity of the two vagi is unequal in the tortoise, the right being the more active. 6. Dissection of the Vagus and Cardiac Sympathetic Nerves in the Frog. — (i) Put the tissues in the region of the neck on the stretch by passing into the gullet a narrow test-tube or a thick glass rod moistened with water, and by pinning apart the anterior limbs. Expose the heart PRACTICAL EXERCISES 197 by cutting through the pectoral girdle in the way described in 2 (p. 194). On clearing away a little connective tissue and muscle with a seeker, three large nerves will come into view. The upper is the glosso- pharyngeal, the lower the hypoglossal; the vagus crosses diagonally between them (Fig. 91). Above the vagus trunk, running parallel to it, and separated from it by a thin muscle and a bloodvessel (the carotid artery), lies its laryngeal branch. The vagus should be traced up to the gang- lion situated on it near its exit from the skull. (2) Then cut away the lower jaw, dividing and reflecting the membrane covering the roof of the mouth. At the junction of the skull and the back- bone will be seen on each side the levator anguli scapulae muscle (Fig. 92). Remove this muscle care- fully with fine forceps. Clear away a little con- nective tissue lying just over the upper cervical vertebrae, and the sym- pathetic chain, with its ganglia, will be seen. Pass a fine silk thread beneath the sympathetic about the level of the large brachial nerve, by means of a sewing-needle which has been slightly bent in a flame and fastened in a handle. Tie the ligature, divide the sympathetic be- low it, and isolate it care- fully with fine scissors up to its junction with the vagus ganglion. Batteries — To set up a Daniett Cell. — Fill the por- ous pot (Fig. 230, p. 724), previously well soaked in water, with dilute sulph- uric acid (i part of com- mercial acid to 10 or 15 parts of water) to within i£ inches of the brim, and place in it the piece of amalgamated zinc. If the zinc is not properly amalgamated, leave it in the pot fo'r a minute or two to clean its surface. Then lift it out, pour over it a little mercury, and rub the mercury thoroughly over it with a cloth. Put the pot into the outer vessel, which contains the copper plate, and is filled with a saturated solution of sulphate of copper, with some undissolved B — B Fig. 90. — Arrangement for recording Auricular and Ventricular Contractions (and studying the Influence of Temperature of the Heart). C, clamp holding the heart at the auriculo-ven- tricular groove; P, pulley round which a thread attached to the apex of the ventricle passes to the lever L'; L, lever connected with auricle. (The rest of the arrangement is for studying the influence of temperature on the heart and its nerves, G being a vessel filled with physiological salt solution in which the heart is immersed; R, an inflow tube from a reservoir containing salt solution at the temperature required; O', an out- flow tube by which G may be emptied into the beaker B'; O, a tube passing to the beaker B to prevent overflow from G; T, a thermometer.) 198 THE CIRCULATION OF THE BLOOD AND LYMPH crystals to keep it saturated. After using the Daniell, it must always be taken down. The outer pot is left with the copper plate and the sulphate solution in it. The zinc is washed and brushed bright. The sulphuric acid is poured into the stock bottle, and the porous pot put into a large jar of water to soak. The Bichromate Cell contains only one liquid — a mixture of i part of sulphuric acid with 4 parts of a 10 per cent, solution of potassium bichromate. In this is placed one, or in some forms two, carbon plates and a plate of amalgamated zinc. After using the battery, take the zinc out of the liquid. The Leclanche battery consists of a porous pot filled with a mixture of manganese dioxide and carbon packed around a carbon plate, which forms the positive pole. The pot stands in an outer jar of glass filled with a saturated solution of ammonium chloride, into which dips an amalgamated zinc rod, which constitutes the negative pole. Various forms of dry batteries can be conveniently used for running indue tion- coils or time-markers, but are not rrvi LqryngeaL branch of ' adapted for yielding constant cur- rents of long duration. 7. Stimulation of the Vagus in the Frog. — Make the same arrange- ments as in 5 (i) (p. 195), but in addition set up an induction machine arranged for an inter- rupted current (Fig. 93), with a Daniell, a bichromate, a Leclanche, or a dry cell in the primary circuit, which should also include a simple key. Insert a short-circuiting key in the secondary circuit. Attach the electrodes to the short-circuit- ing key, push the secondary coil up towards the primary until the shocks are distinctly felt on the tongue when the Neef 's hammer is set going and the short-circuiting key opened. Pith the brain of a frog, expose the heart, dissect out the vagus on one side, ligature it as high up as possible, and divide above the ligature. Fasten the electrodes on the cork plate by means of an indiarubber band, and lay the vagus on them. Set the drum off (at slow speed). After a dozen heart -beats have been recorded, stimulate the vagus for two or three seconds by opening the short- circuiting key. If the nerve is active, the heart will be slowed, weakened, or stopped. In the last case the lever will trace an unbroken straight line ; but even if the stimulation is continued the beats will again begin. 8. Stimulation of the Junction of the Sinus and Auricles. — After a sufficient number of the observations described in 7 have been taken with varying time and strength of stimulation, take the writing-points off the drum, apply the electrodes directly to the crescent at the junc- tion of the sinus venosus with the right auricle, and stimulate. The heart will be affected very much in the same way as by stimulation of the vagus, except that during the actual stimulation its beats may be quickened and the inhibition may only begin after the electrodes have been removed (Fig. 70, p. 158)'. Fig. 91. — The Relations of the Vagus in the Frog. 9- Effect of Muscarine (or Pilocarpine), and Atropine. — Paint on the sinus venosus with a small camel's-hair brush a very dilute solution oi muscarine (or of pilocarpine). The heart will soon be seen to beat more slowly, and will ultimately stop in diastole. Now apply a dilute solution of sulphate of atropine to the sinus. The heart will again begin to beat. Stimulation of the vagus will now cause no inhibition of the heart, because its endings have been paralyzed by atropine. (Muscarine or pilocarpine has also been applied to the heart, but it could be shown by a separate experiment that atropine by itself has the same effect on the vagus endings — p. 16 .) 10. Stannius' Experiment. — Pith a frog. Expose the heart in the way described under 2 (p. 194). Ligature the frsenum with a fine silk thread, and use the thread to manipulate the heart. With a curved needle pass a moistened silk thread between the aorta and the superior vena cava, and tie it round the junction of the sinus and right auricle (Fig. 86). The auricles and ventricle stop beating as soon as the ligature is tightened. The sinus venosus goes on beat- ing. Now separate the ven- tricle from the rest of the heart by an incision through the auriculo-ventricular groove, or tie a second ligature in the groove. The ventricle begins to beat again, the auricle re- maining quiescent in diastole (p. 166). Occasionally both auricle and ventricle, or only the auricle, may begin to beat. n. Stimulation of Cardiac Sympathetic Fibres in the Frog — (i) In the vago-sympathetic after the inhibitory fibres have been cut out by atropine. — Arrange everything as in 7 (p. 198). Assure yourself, by £AS Fig. 92. — Relation of the Sympathetic to the Vagus in the Frog (after Gaskell). Sym, sympathetic chain ; G, ganglion of the vagus; VS, vago-sympathetic; GP, glosso-pharyngeal nerve; LAS, levator anguli scapulae muscle ; SA, subclavian artery; A, descending aorta; V, vertebral column; OC, occipital part of skull; 1-4, spinal nerves. stimulating the vagus, that it inhibits the heart, and take a tracing during stimulation. Then paint a dilute solution of atropine on the sinus. Stimulation of the vagus, which is really the vago-sympathetic (see Fig. 92), will now cause, not inhibition, but augmentation (increase in rate or force, or both), since the endings of the inhibitory fibres have been paralyzed by atropine. The strength of the stimulating current required to bring out a typical augmentor effect is greater than that needed to stimulate the inhibitory fibres. Take a tracing to show augmentation produced by stimulating the nerve. (2) By direct stimulation of the cervical sympathetic. — Make the same arrangements as in n (i), but, instead of isolating the vagus, dissect out the sympathetic on one side in the manner described in -6 (2) (p. 197), and do not apply atropine to the heart. Lay the upper (cephalic) end of the sympathetic on very fine and well-insulated electrodes, and stimulate (Fig. 76, p. 167). (To insulate electrodes the points may be covered with melted paraffin. When the paraffin has cooled, a narrow 200 THE CIRCULATION OF THE BLOOD AND LYMPH groove, just sufficient to lay bare the wires on the upper side, is made in it, and the nerve is laid in this groove.) Experiments 7, n (i) and n (2) will be rendered more exact by connecting a second electro-magnetic signal with a Pohl's commutator without cross-wires (Fig. 94), in such a way that the circuit is inter- rupted at the instant when stimulation begins. 12. The Action of Inorganic Salts on Heart- Muscle. — Expose and remove the heart of a tortoise or turtle (p. 196). Cut off the apical two- thirds of the ventricle by an incision parallel to the auriculo-ventricular groove. By a second parallel cut remove a ring of tissue 2 or 3 milli- metres wide from the upper end of this portion of the ventricle. Divide the ring at opposite ends of a diameter, so as to form two strips. Tie a fine silk thread to each end of one strip. Attach one of the threads to the short limb of a glass rod bent at right angles, so that it can be immersed at will in a beaker. The other end of the rod is fixed in a holder sliding on a stand. Attach the second thread to the short arm Fig- ?3- — Arrangement of Induction Machine for Tetanus. B, battery; K, simple key ; P, primary coil ; S, secondary coil ; A, C, binding screws to be connected with battery 46r single shocks; F, G, binding screws for tetanizing current; N, Neef's hammery D, short-circuiting key in secondary; E, electrodes. D and E are drawn to a much larger scale than the rest of the figure. of a counterpoised lever arranged to write on a slowly-moving drum. If the strip is still beating, wait till the contractions have ceased ; then (1) Immerse the strip in a beaker filled with o'y per cent, solution of sodium chloride. After a time it begins to beat rhythmically. The contractions become rapidly stronger, and then after a while diminish, and gradually cease. The tone or tonus of the strip is diminished by the solution. (2) Arrange the other strip in the same way, and immerse it in a solution of calcium chloride (about I per cent.) isotonic with the sodium chloride solution used in (i). If the strip is contracting, the contrac- tions will cease. Rhythmical contractions will not appear as in the sodium chloride solution. The tone of the strip may be increased. (3) Remove most of the calcium chloride solution from the beaker, and fill it up with 0-7 per cent, sodium chloride solution. The rhythmi- cal contractions will appear after a longer or shorter latent period, and will be stronger and last for a longer time than in the sodium chloride solution alone. (4) Immerse a fresh strip in a solution containing sodium chloride PRACTICAL EXERCISES aol (o-7 per cent.), calcium chloride (o-o25 per cent.), and potassium chloride (0-03 per cent.) (a modified Ringer's solution). A longer series of rhythmical contractions will be obtained than in either (i) or (3). That this is not due to the potassium chloride acting alone can be shown by immersing a strip in a solution of potassium chloride (about o '9 per cent.) isotonic with the sodium chloride solution used in (i). No contractions will be caused. 13. The Action of the Mammalian Heart. — Inject under the skin of a dog (preferably a small one) i c.c. of a 2 per cent, solution of morphine hydrochlorate for every kilo of body-weight. As soon as the morphine has taken effect (in 15 to 30 minutes, but better after an hour), fasten the animal back down on a holder (as in Fig. 135, p. 301), pushing the mouth-pin behind the canine teeth and screwing the nut home.* In the meantime select a tracheal cannulaf of suitable size, and get ready instruments for dissection — one or two pairs of artery-forceps, a pair of artery-clamps (bulldog pattern), two or three glass cannuloc of 'B' Fig- 94- — Arrangement for recording the Beginning and End of Stimulation. C. Pohl's commutator without cross-wires; B, battery in circuit of primary coil P; B', battery in circuit of electro-magnetic signal T; K, simple key in primary circuit; S, secondary coil. When the bridge of the commutator is tilted into the position shown in the figure, the primary circuit is closed and the circuit of the signal broken. various sizes for bloodvessels, ten strong waxed ligatures, sponges, hot water, a towel or two, and a pair of bellows to be connected with the tracheal cannula when the chest is opened. Arrange an induction - * A simple but efficient and convenient holder for a dog may be easily constructed as follows: Take a board of the length required (2^ to 5 feet, according to the size of the dog) . At one end fasten two short upright wooden pins, with a clear space of 4 to 6 inches between them. These are pierced from side to side with four or five holes at different heights. An iron pin passes behind the canine teeth of the animal through two corresponding holes in the uprights, and the muzzle is tied over this by a cord which secures the head. For a large dog an upper pair of holes is used, for a small dog a lower pair. The feet are fastened by cords to staples inserted into the sides of the board, the fore-legs being drawn tailwards for all operations on the neck or head, headwards for operations on the thorax. A rabbit-holder can be made in exactly the same way. f A tracheal cannula is easily made by heating a piece of glass tubing, about 6 inches long, a short distance from one end, and drawing it out slightly so as to form a ' neck.' The tubing is then bent about its middle to an obtuse angle, and the end next the neck is ground obliquely on a stone. The diameter of the cannula should be about the same as that of the trachea, into which it is to be inserted by its oblique end. 202 THE CIRCULATION OF THE BLOOD AND LYMPH coil and electrodes for a tetanizing current (Fig. 93, p. 300). With scissors curved on the flat clip away the hair from the front of the neck. Put the hair carefully away, and remove all the loose hairs with a wet sponge so that they may not get into the wounds. Give ether, or pour into the stomach by a tube 5 c.c. of a 0*5 per cent, solution of chloroform, in 10 per cent, alcohol per kilo of body-weight, diluted before administration with 3 or 4 volumes of water (Grehant's method). To put a Cannula in the Trachea. — The hair having been clipped in the middle line of the neck and the skin shaved, a mesial incision is to be made, beginning a little below the cricoid cartilage, which can be felt with the finger. The trachea is then cleared from its attach- ments by forceps or a blunt needle, and two strong ligatures are passed beneath it. A single loop is placed on each of these, but is not drawn tight. Raising the trachea by means of the upper ligature, the student makes a longitudinal incision through two or three of the cartilaginous rings, inserts the cannula, and ties the lower ligature firmly around its neck. The upper ligature can now be withdrawn. Clip off the hair on each side of the sternum. Make an incision on each side through the skin and down to the costal cartilages about 2 inches from the edge of the breast-bone, and long enough to expose about four costal cartilages (say, 3rd to 6th). With a curved needle pass waxed ligatures round the cartilages, and tie firmly to compress the intercostal vessels. The bellows should now, or earlier if any symptoms of impeded respiration have appeared, be connected with one end of the horizontal limb of a glass T-piece, the other end of which is similarly connected with the tracheal cannula. The stem of the T-piece is provided with a short piece of rubber tubing, which, when artificial respiration is being carried on, Is to be alternately closed and opened — closed during inflation of the lungs, and opened when the air is to be allowed to escape from them. Or a screw-clamp may be adjusted on the piece of tubing so that the opening is sufficiently narrow to permit the lungs to be properly inflated when the bellows are compressed, and yet sufficiently wide to permit easy escape of the air and collapse of the lungs at the end of each inflation. Ether may, when necessary, be administered, by inserting between the T-piece and the tube from the bellows an ether bottle with two tubes passing through the cork to within an inch or two of the ether. If the cannula has a side-opening, as is usually the case v/ith metal cannulae, the T-piece may be dispensed with. One student should take sole charge of the artificial respiration, which ought to be begun as soon as the chest has been opened, and continued at the rate of about twenty inflations per minute. The costal cartilages are rapidly cut through with strong scissors just on the sternal side of the ligatures, the artificial respira- tion being suspended for an instant, as each cut is made, to avoid wounding the lungs. The sternum is divided at its lower end and turned up. If there is much bleeding a ligature should be tied round its tipper end. With a curved needle a ligature is passed below the internal mammary arteries as they approach the sternum. That bone may now be removed, and the heart, enclosed in the pericardium, comes into view. A thread is passed with a suture-needle through each side ol the pericardium, which is then stitched to the chest-wall and opened. (a) Note the various portions of the heart, right and left ventricles, right and left auricles, with the auricular app.ndices. Feel the heait with the hand, and observe that the right ventricle is softer and has thinner walls than the left, and that the auricles are softer than the ventricles. Note how all the parts of the heart harden in the hand during systole and soften during diastole (pp. 86, 90). PRACTICAL EXERCISES 203 D (b) Dissect out the vago-sympathetic on one side in the neck of the dog. The guide to the nerve is the carotid artery. These two struc- tures and the internal jugu- lar vein lie side by side in a common sheath. Feel for the artery a little ex- ternal to the trachea, cut down on it, open the sheath , isolate the vago - sympa- thetic for about an inch, pass two ligatures under it, tie them, and divide be- tween the ligatures. The peripheral and central ends of the nerve may now be successively stimulated. Stimulation of the peri- pheral end causes slowing of the heart, or stoppage in diastole. Feel that it softens when it stops. It soon begins to beat again. Stimulation of the central end of the vago-sympa- thetic may or may not cause inhibition . If it does, expose the other vago- sympathetic, divide it, and repeat the stimulation of the central end . There will now be no inhibition of the heart. Incidentally it may be seen that stimulation of the central end of the vago - sympathetic causes strong, though, of course, withopened chest.abortive, respiratory movements. (c) Pith a frog (brain and cord), dissect out the sciatic nerve on one side up to the sacral plexus. Cut off the whole leg. Drop the cut end of the nerve on the heart, and hold the prep- aration so that the nerve touches the heart also by its longitudinal surf ace. At each cardiac beat the nerve is stimulated by the action current (p. 833), and the muscles of the leg contract. (d) Raise the board so that the head of the animal is down and the hind-feet up, and note whether there is any effect on the action P\U Fig. 95- — Myocardiograph of Adami and Roy (modified by Cushny and Matthews). AB, a perpendicular rod descending from a universal joint, which is not shown in the figure; CD, a brass sheath, moving easily on the rod, and bearing on its upper end an ivory pulley, and at its lower end a horizontal bar, which is inter- rupted by a plate of hard rubber, I. The per pendicular rod EF moves on the horizontal bai by the hinge-joint, J. EF is hooked at one end for attachment to the heart, and bored at the other for a thread which, passing over the pulley at C, passes through the universal joint and moves a writing lever not shown in the figure. CD is prevented from moving up AB by a ring of brass, G, which is screwed to AB, but is not attached to CD ; the hook F can therefore move to and from AB, and can rotate round it, while it cannot move up or down. The hooks F and B are insulated from each other by the hard rubber, I. H is a binding post through which, and through another connected with A, induction shocks may be sent at will into the portion of the heart lying between the hooks. 204 THE CIRCULATION OF THE BLOOD AND LYMPH and filling of the heart. Repeat the observation with head up and feet down. (e) Compress the aorta with the fingers, and observe the effect on the degice of dilatation of the various cavities of the heart. Repeat the experiment with the inferior vena cava, and compare the results. (/) Smoke a drum. Insert the hooks of the myocardiograph (Fig. 95) into the ventricle, taking care not to penetrate deeply into the -wall. Arrange the lever to write on the drum. While a tracing is being taken stimulate the peripheral end of the vagus. Unhook the cardiograph. (g) Stop the artificial respiration, and observe the changes which take place in the auricles and ventricles, comparing particularly the right side of the heart with the left. Before the heart has stopped beating, re- commence the artificial respiration. (h) Connect a cylinder of oxygen with a good-sized rubber catheter, 'C Fig. 96. — Arrangement to illustrate Action of Cardiac Valves in the Heart of an Ox (Gad). C, glass window in left auricle; D, window in aorta; E, tube inserted through apex of heart into left ventricle and connected with pump P; A, side tube on E, through which wires are connected with a'tiny incandescent lamp in the ventricle; W, water in bottle B; T, T', tubes. and pass the catheter down the trachea! cannula or through a separate opening in the trachea. Allow a small stream of oxygen to flow into the lungs. Artificial respiration is now unnecessary. The lungs remain at rest, yet the blood is sufficiently oxygenated, and the heart goes on beating. The myocardiographic tracing thus goes on undis- turbed by respiratory movements. (i) Stop the oxygen, and resume the artificial respiration. Make a PRACTICAL EXERCISES 205 Small penetrating wound with a scalpel in the left ventricle. Observe the course of the haemorrhage, and note especially the difference in systole and diastole. (/) Lay the electrodes on the heart, and stimulate it with a strong interrupted current. The character of the contraction soon becomes profoundly altered. Shallow, irregular contractions flicker over the surface, with a kind of simmering movement sugges- tive of a boiling pot (delirium cordis, fibrillar contraction). Now kill the ani- mal by stopping the artificial respiration. Observe how l9ng the heart continues to beat, and which of its divisions stops last. (k) Make a dissection of the cervical sympathetic up to the superior cervical ganglion, and down through the inferior cervical ganglion to the stellate or first thoracic ganglion. Make out the annulus of Vieussens and the .cardiac sympa- thetic (accelerator) branches going off from the annulus or the inferior cervical ganglion to the cardiac plexus. 14. Perfusion of the Isolated Mam- malian Heart. — The heart of a dog em- ployed for some other experiment may be used. Or a rabbit may be killed by a blow on the back of the head, and the heart at once removed. The aorta should not be cut off too short. Tie a cannula into the aorta and attach it to a T-piece connected by rubber tubes, which must be perfectly clean, with two bottles, one containing Ringer's solution (pp. 66, 201), preferably that made with dextrose, the other containing defibrin- ated blood diluted with Ringer's solu- tion. The defibrinated blood should be strained so as to remove any small pieces of fibrin. The bottles are supported on a high stand, so that the level of the bottles above the heart can be altered, and the pressure of the perfusion liquid thus varied. Perfusion may be begun with Ringer, to wash out any remaining blood and obviate the possible formation of clots in the small vessels. Oxygen is allowed to bubble through the Ringer's solution, but this is not necessary for the blood, since, if shaken up, it will retain far more oxygen than the Ringer's solu- tion. The temperature of the liquids should be at about 40° C. when nearing the heart. ,\ : : ~ ~:~ - - a -"-""- ri~; - -C b 1~ - ~~ ;,--=-'£ lit 9- /. N_ Fig. 97- — Mammalian Heart Per. fusion Apparatus (Gunn). a, . Liebig .condenser, cut off as shown; b, inlet for the warm water; d, thermometer almost filling up the lumen of the thin glass tube c ; e, cork ; /, cannula for aorta fitted with a collar of rubber tubing, g, in the end of the tube c ; h, Y-tube connected with two reservoirs, one contain- ing Ringer's solution, the other any other liquid which is to be perfused. can ~ ........... ____ o _________ easily insured by interposing a worm immersed in a heated bath or other heating arrangement between the cannula and the T-tube, and lor the study of its movements by inspection the heart itself can be place in a glass vessel immersed in the bath. tract! When records of the contract! 2ob THE CIRCULATION OF THE BLOOD AND LYMPH are to be obtained, threads are attached to the auricle and to the apex of the ventricl:. The heart is suspended by fastening the cannula in a holder on a stand, and the threads, after passing over pulleys to give them a convenient direction, are attached to writing-levers. As the heart cannot now be easily kept immersed in the bath, it is suspended in the air, and can be kept warm by the following simple arrangement: A copper pipe about 4 inches long is slit on one side, and on the opposite side is screwed or riveted to a copper rod, under which is hung a spirit-lamp. The lamp is adjusted at such a point on the rod that when the copper tube is placed around the heart the heat conducted along the rod keeps the air around the heart at about body-temperature. The perfusion liquid before it enters the heart may be heated thus: A Liebig's condenser is cut through the middle, and the large end closed by a paraffined cork. A glass tube is run down from the top through this cork, and the aorta is attached directly to this, so that the heart is very near the condenser. This tube is mostly filled up by a thermometer, so that the perfusion liquid passes through it in a thin stream which is easily heated By the water in the condenser, which contains a second ther- mometer. This water is kept constantly flowing through the condenser from a heated bath. The T-piece connecting with the perfusion bottles is attached to the upper end of the glass tube to which the heart is attached (Gunn and Cushny). 15. Action of the Valves of the Heart.— (i) Study the action of the valves of the ox- heart, connected with the pump P and bottle B in the artificial scheme, as shown in Fig. 96. The cavity of the heart is illuminated by means of a small electric lamp, the wires of which pass in at Fig. 98.— Diagram of Valves of the Heart. The valves are supposed to be viewed from above, the auricles having been partially removed. A, aorta with semilunar valve; B, pulmonary artery and valve; C, tricuspid, and D, mitral valve; E, right, andF, left coronary artery ; G, wall of right, and H, of left auricle, I, wall of right, and J, of left ventricle. A. When the piston of the pump is pushed down, water is forced through the aorta D along the tube T into the bottle, and flows back again into the left auricle by the tube T'. During each stroke of the pump the auriculo-ventricular valve is seen through the glass disc inserted into C to close, and the semilunar valve is seen through the glass in D to open. When the piston is raised, the semilunar valve is seen to be closed and the auriculo-ventricular valve to be opened. For comparison, a human heart with a valvular lesion might be used. (2) With the sheep's or dog's heart provided, perform the following experiments : (a) Open the pericardium and notice how it is reflected around the great vessels at the base of the heart. Distinguish the pulmonary artery, the aorta, the superior and inferior venae cavae, and the pul- monary veins. The trachea and portions of the lungs may also be attached. If so, remove them carefully without injuring the heart. PRACTICAL EXERCISES 207 (b) Take two wide glass tubes, drawn slightly into a netk at one end. One of the tubes should be about 10 cm. long, and the other about 50 cm. Tie the short tube A firmly by its neck into the superior vena cava- the long tube B into the pulmonary artery. Ligature the inferior vena cava. Connect A by a small piece of rubber tubing with a funnel supported in a ring on a stand. Pour water into the, funnel till the right side of the heart is full. It will escape from the left azygos vein, which must be tied. Put on any additional ligatures that may be needed to render the heart water-tight. Support B in the vertical position by a clamp. Fill the funnel with water, and it will rise in B to the same level as in the funnel. Now compress the right ventricle with the hand, and the water will rise higher in B. Relax the pressure and notice that the water remains at the higher level in B, being pre- vented by the semilunar valves from flowing back into the ventricle. By alternately compressing the ventricle and allowing it to relax, water can be pumped into B till it escapes from its upper end, and if this is so curved that the water falls into the funnel, a ' circulation ' which imitates that of the blood can be established. Note that during the pumping the sinuses of Valsalva, behind the semilunar valves at the origin of the pulmonary artery, become prominent. (c) Take out B and tear out one of the segments of the semilunar valve. Replace B, and notice that, while compression of the ventricle has the same effect as before, the water no longer keeps its level on relaxation, but regurgitates into the ventricle. This illustrates the condition known as insufficiency or incompetence of the valves. But if the injury is not too extensive, it is still possible, by more vigorously and more rapidly compressing the heart, to pump water into the funnel. This illustrates the establishment of compensation in cases of valvular lesion. (d) Now remove both tubes. Tie the pulmonary artery. Cut away the greater part of the right auricle. Pour water into the auriculo- ventricular orifice, and notice that the segments of the tricuspid valve are floated up so as to close the orifice. Invert the heart, and the ventricle will remain full of water. Open the right ventricle carefully, and study the papillary muscles and the chordae tendineai, noting that the latter are inserted into the lower surface of the -Segments of the tricuspid valve, as well as into their free edges. (e) Repeat (b), (c), and (d) on the left side of the heart, tying tube B into the aorta as far from the heart as possible, and A into the left auricle. (/) Separate the aorta from the left ventricle, cutting wide of its origin so as not to injure the semilunar valves, and tie a short wide tube into its distal end. Fill the tube with water, and notice that the valves support it. Cut open the aorta just between two adjacent segments of the valve, and notice the pockets behind the segments, and how they are related to each other, and connected to the wall of the vessel. 1 6. Sounds of the Heart. — (a) In a fellow-student notice the position of the cardiac impulse, the chest being well exposed. Use both a binaural and a single-tube stethoscope. Place the chest-piece of the stethoscope over the impulse, and make out the two sounds and the pause, (b) With the hand over the radial or brachial artery, try to determine whether the beat of the pulse is felt in the period of the sounds or of the pause. . (c) Listen with the stethoscope over the junction of the second right costal cartilage with the "sternum, and compare the relative intensity of the two sounds as heard here with their relative intensity as heard over the cardiac impulse. 17. Cardiogram. — Smoke a drum, and arrange a recording tambour and a time-marker beating half or quarter seconds to write on it (Fig. 88, 2o8 THE CIRCULATION OF THE BLOOD AND LYMPH p. 195). Apply the button of a cardiograph (Fig. 27, p. 90) over your own cardiac impulse, and fasten it round the body by the bands attached to the instrument. Connect the cardiograph by an indiarubber tube with a recording tambour (Fig. 99). Set the drum off at a fast speed, take a tracing, and varnish it. Compare with Fig. 28 (p. 91), and if the tracing is sufficiently typical, as is often not the case with human cardiograms, measure out the time -value of the various events in the cardiac revolution. Fig. 99.— Marey's Tambour. For the cardiograph, a small glass funnel, or thistle-tube, the stem of which is connected with the recording tambour, may be substituted, the broad end of the funnel being pressed over the apex-beat. 1 8. Sphygmographic Tracings. — Attach a Marey's sphygmograph (Fig. 37, p. 103) to the arm. Fasten a smoked paper on the plate D. Apply the pad C of the sphygmograph to the wrist over the point where the pulse of the radial artery can be most distinctly felt. Ad just the pressure by moving the screw G. The writing-point of the lever E will rise and fall with every pulse-beat. When everything is satisfactorily arranged, set off the clockwork which Fig. 100. — Dudgeon's Sphygmograph. moves the plate D, and a pulse tracing will be obtained. Study the changes which can be produced in the pulse curve — (a) by altering the position of the body (sitting, standing, and lying down); (6) by exercise (Fig. loi) ; (c) by inhalation of 2 drops of amyl nitrite poured on a hand- kerchief by the demonstrator (Fig. 102); (d) by raising the arm above the head and letting it hang at the side ; (e) by compression of the brachial artery at the bend of the elbow; (/) by altering the pressure of the pad Varnish the tracings after marking on them the conditions under which they were obtained. A Dudgeon's sphygmograph (Fig. 100) may also be employed. In this the clockwork carries the strip of blackened paper aloirg beneath PRACTICAL EXERCISES 209 Fig. 101. — Effect of Exercise on the Pulse (Marey). Upper tracing, normal ; lower, after running. the needle which records the movements of the artery. Or a small glass funnel or thistle-tube connected with a recording tambour may be pressed over the carotid artery. The lever of the tambour writes on a drum, on which at the same time half or quarter seconds are marked by an electro-magnetic signal. 19. Venous Pulse Tracing from the Jugular Vein. — Arrange a recording tambour to write on a drum. Con- nect the tambour with the stem of a small glass thistle-tube or funnel (or with a small metal cup) by a piece of narrow rubber tubing, and apply the cup-shaped end of the thistle-tube over the right jugular bulb of a fellow-student. This lies about i inch external to the right sterno-clavicular articulation, and a little above it. The receiver may have to be moved about a little until the best pulsation is obtained. The ' patient ' should be lying down, the shoulders slightly raised, the head on a pillow and turned slightly to the right, in order to relax the right sterno-mastoid muscle (Mackenzie). 20. Polygraph Tracings.— Arrange the polygraph over the radial artery, as with an ordinary sphygmograph, so that the lever will record the radial pulse when the strip of paper is set moving. If the instru- ment has only one tambour, con- nect the tambour to a receiver or thistle-tube over the jugular bulb. The writing-point of the tambour is arranged so as to be immediately below the writing- point connected with the radial. If the polygraph is provided with clockwork to record time, set off the time -marker writing fifths of a second. When it is seen that the writing-points are marking properly, start the clockwork which moves the strip of smoked paper. Repeat the observation with the tam- bour connected with the apex- beat. Letter the curves as far Fig. 102.— Effect of Amyl Nitrite on the as possible as in Figs. 65 and Pulse (Marey). Upper tracing, normal; 66 (p. 149) without at present lower, after inhalation of amyl nitrite. attempting their exact analysis. If the polygraph has two tam- bours, simultaneous tracing of the radial pulse, the jugular pulse, and the cardiac impulse, or of the carotid pulse, the jugular pulse, and the apex- beat, may be taken, and other combinations as well. If no polygraph is available, a drum may be employed, the tracings being all taken with thistle-tubes connected with recording tambours. The levers of the tambours must be arranged to write on the drurh in the same vertical straight line, or, without making the adjustment quite exact, vertical lines of reference may be drawn through each curve, with the drum at rest, indicating the relative positions of the writing-points. 2io THE CIRCULATION OF THE BLOOD AND LYMPH 21. Plethysmographic Tracings. — Connect the vessel D (Fig. 36, p. 128), directly with a recording tambour by the tube F, omitting for simplicity the recording arrangement in the figure. Place the arm in the plethysmograph, and adjust the indiarubber band to make a watertight connection. Support D so that the arm rests easily within it, and fill it with water at body temperature. No water must get into the tambour, and it is well to insert a piece of glass tubing in the connection between it and the plethysmograph, so that it may be seen when the water is rising too high. A T -piece with a short piece of rubber tubing on the stem should be inserted in the course of the tube leading to the tambour. All adjustments are made with the T-piece open, and when a tracing is to be taken the short rubber tube is closed by a clip. Arrange a time-marker to write half or quarter seconds (Fig. 88, p. 195). Adjust the writing-point to write on a drum, and close the upper tubulure C with a cork. The quantity of blood in the arm is increased with every systole of the left ventricle, diminished in diastole. The lever will therefore rise when the ventricle contracts, and sink when it relaxes. (1) Take tracings with the arm (a) horizontal, (b) hanging down. (2) With the arm horizontal, take tracings to show the effect (a) of closing and opening the fist Inside the plethysmograph;* (b) of apply- ing a tight bandage round the arm a little way above the indiarubber band ; (c) of inhaling 2 drops of amyl nitrite. Instead of the arm plethysmograph, a small plethysmograph to hold a finger may be employed. It consists of a glass tube drawn out at one end. The wide end is provided with a rubber collar. The narrow end is connected by a small rubber tube with a very small and sensitive recording tambour, a T-piece being inserted on the connection as before. With the T-piece closed fill the tube with water. Then, holding up the wide end of the tube, the tip of the finger is put in so as just to close the tube. The T-piece is then raised and opened, and the finger pushed in as far as it will go. The collar must fit the finger so as to form a watertight joint. Now get the proper pressure in the tambour by blowing into the T-piece, and close the clamp. A time-tracing can be taken as before. 22. Pulse-Rate. — (i) Count the radial pulse for a minute in the sitting, supine, and standing positions. Use a stop-watch, setting it off on a pulse-beat and counting the next beat as one. Make three observations in each position. (2) Count the pulse in a person sitting at rest, and then again in the sitting position immediately after active muscular exertion. Note how long it takes before the pulse-rate comes back to normal. (3) Count the pulse in a person sitting at rest. Repeat the observa- tion while water is being slowly sipped, and note any change. (4) With one hand over the thorax of a rabbit, count its pulse. Then notice the effect (a) of suddenly closing its nostrils, (b) of bringing a small piece of cotton-wool sprinkled with ammonia or chloroform in front of the nose (reflex inhibition of the heart}. 23. Blood-Pressure Tracing. — (a) Put a dog under morphine (p. 63). Set up an induction machine arranged for an interrupted current (Fig. 93, p. 200). Fill the U-shaped manometer tube (if this has not already been done) with clean mercury to the height of 10 to 12 cm. in each limb. If the float tends to stick, half an inch of oil may be put above the mercury in the distal (straight) limb before putting in the float. But where the mercury is clean and dry, and the * Closing the fist causes a fall in the curve — i.e., a diminution in the volume of the arm. On opening the hand, the curve regains its level. PRACTICAL EXERCISES 211 size of the float properly adjusted to that of the tube, this is not neces- sary, and is to be avoided. Then, tilting the tube carefully, fill the proximal limb (i.e., the limb which is to be connected with the blood- vessel) with a saturated solution of sodium carbonate or a half -saturated solution of magnesium sulphate, or, what is better for most purposes, a 2 per cent, solution of sodium citrate. This is easily done by means of a pipette furnished with a long point. Now attach a strong rubber tube to the proximal end of the manometer, and fill it also with the solution. All air must be got out of the manometer and its connecting- tube. Raise the end of the rubber tube and blow into it, so as to cause a difference of about 10 cm. in the height of the mercury in the two limbs of the manometer, and, without releasing the pressure, clamp the tube with a pinchcock or screw clamp (Fig. 4 - , p. no). Now smoke a drum, and arrange the writing-point of the manometer- float so that it will write on it. Suspend a small weight by a piece of silk thread from a support attached to the stand of the drum, so that it hangs down outside of the writing-point of the manometer-float and always keeps it in contact with the smoked surface without undue friction. Or a piece of glass rod drawn out to a fine thread in the blowpipe flame answers very well. Below the writing-point of the float, and in the same vertical line with it, adjust the writing-point of a time-marker beating seconds (Fig. 88, p. 195). Next fasten the animal on a holder, back down. Give ether and insert a tracheal cannula (p. 202). (The tracheal cannula is not abso- lutely required for the experiment, but it is convenient, as the animal is more under control, and artificial respiration can be begun at any moment, should this be necessary.) Insert a glass cannula, armed with a short piece of rubber tubing, into the central (cardiac) end of the carotid artery (p. 63). Leaving the bulldog forceps on the artery, fill the cannula and tube with the sodium citrate or one of the other solutions. Slip the rubber tube over a short glass connecting-tube. Fill this also with the solution, and connect it with the manometer-tube, seeing that both are quite full of liquid, so that no air may be enclosed. Where a permanent working place is provided for blood-pressure experiments it is convenient to connect the cannula and manometer with a pressure -bottle containing the sodium citrate solution, and to use a three-way cannula for the bloodvessels (Fig. 103). The cannula has a bulbous enlargement, which hinders clotting. The end of the cannula is connected with the tube from the pressure -bottle, which is closed by a clip, and the side-tube is connected with one limb, E, of the manometer shown in Fig. 104. E is itself provided with a side- tube, F, armed with a short piece of rubber tubing. The cannula does not require to be filled with liquid before being inserted into the artery. By opening F and releasing the clip on the tube from the pressure- bottle the cannula and the tube connecting it with the manometer can be filled, and any blood-clots can be easily washed out in the course of an experiment. Before the bulldog forceps is taken off the artery to obtain a blood-pressure tracing, F must be closed, and the clip on the tube from the pressure-bottle opened. The bottle is attached to a strong cord passing over a pulley, by which it is raised to a height sufficient to balance approximately the pressure in the artery. The tube to the pressure-bottle is then clipped. If no manometer with side-tube is available, a T-piece can be inserted in the connection between the cannula and the manometer, and the cannula can be washed out through this. Now take the bulldog forceps off the artery, and allow the drum to revolve at slow speed. The writing-point of the manometer-float will 212 THE CIRCULATION OF THE BLOOD AXD LYMPH trace a curve showing an elevation for each heart-beat, and longer waves due to the movements of respiration. (6) Isolate the vago-sympathetic nerve in the neck. Ligature doubly, and cut between the ligatures. Stimulate the peri- pheral (lower) end ; the heart will be slowed or stopped, and the blood-pressure will fall. -. Stimulate the central (upper) end; there B may be inhibition of the heart or accelera- tion, and the pressure may fall or rise (p. 170). (c) Expose and divide the other vago- sympathetic while a tracing is being taken. 4gain stimulate the central end of the nerve and observe whether there is any effect. (d) Expose the sciatic nerve in one leg, as follows: The leg having been loosened from the holder, the foot is seized by one hand and lifted straight up, so as to put Fig. 103. — Three-way Cannula. the skin of the thigh on the stretch. An incision is now made in the middle line on the posterior aspect of the thigh, through the skin and subcutaneous tissue. The muscles are separated in the line of the incision with the fingers, and the sciatic nerve comes into view lying deeply be- tween them. Place a double ligature on it, and divide between the ligatures. Stimu- late the upper (central end) ; the blood- pressure probably rises, and the heart may be accelerated. Stimulate the peripheral end of the nerve ; there is little change in the blood-pressure and" none in the rate of the heart. (e) Note, incidentally, that stimulation of the central end of the sciatic or the upper (cephalic) end of the vago-sympathetic may cause increase in the rate and depth of the respiratory movements. Dilatation of the pupil is also caused by stimulation of the upper end of Fig. 104. — Manometer with Side-tube (Guthrie). A, float ; B, collar through which the wire C of the float moves; D. vertical wire fixed to mano- meter-holder, which keeps the writing-point on the drum; E, limb of manometer con- nected with cannula, with its side-piece, F.' ZI3 the vago-sympathetic through the sympathetic (pupillo-dilator) fibres that supply the iris. (/) Again stimulate the peripheral end of one vagus, or of both at the same time, while a tracing is being taken, and see how long it is possible to keep the heart from beating. Sometimes, but rarely in the dog, inhibition can be kept up so long that the animal dies. (g) Close the tracheal cannula so that air can no longer enter the lungs. In a very short time the blood-pressure curve begins to rise (rise of asphyxia). After some minutes the pressure falls, and finally, when the circulation has stopped completely and the pressure has become equalized throughout the whole vascular system, a residual pressure of only a few mm. (usually about 10 mm. Hg) is indicated. In order to^get the true zero pressure, disconnect the arterial cannula S { i m ul oi i on of centrol end .stopped Peripheral end .fet A rn u 1 air e cl c e n t r c* 1 ti inulated- Fig. 105. — Blood- Pressure Tracing from a Dog: Stimulation of Central and Peripheral Ends of Vagus. The other vagus was intact. Stimulation of the peripheral end caused stoppage of the heart and a marked fall of pressure. Stimulation of the central end produced a great rise of pressure, with, perhaps, a slight acceleration of the heart. from the manometer, and allow the writing-point to trace a horizontal straight line (line of zero pressure) on the drum (Figs. 84 and 85). 24. Estimation of the Arterial Blood-Pressure in Man. — Use the Riva- Rocci apparatus, as described on p. 113. Begin with the subject in the sitting position. The observer's left hand may be used for palpating the pulse, and the right for working the bulb. Employ the ausculta- tory method as well as palpation, and determine the systolic and dias- tolic pressures. Repeat the observations with the person standing up and lying down. Investigate the effect of muscular exercise on the blood-pressure. 25. The Influence of the Position of the Body on the Blood-Pressure. — Inject into the rectum of a dog 3 to 4. gnu. of chloral hydrate dis- solved in a little water. See chat it does not run out again immediately 214 THE CIRCULATION OF THE BLOOD AND LYMPH after injection. In ten minutes anaesthetize the animal fully with a mixture of equal parts of alcohol, chloroform, and ether (one of the so-called A.C.E. mixtures), or with chloroform, and tie it very securely, back downward, on a board, which can be rotated around ajiorizontal axis, corresponding in position to the point at which the cannula is to be inserted.* Set up a drum and manometer as in 23 (p. 210), but with a rubber connecting-tube of such length as will allow free rotation of the board. Put a cannula in the trachea. Insert a cannula into the central end of the carotid artery at a point immediately above the axis of rotation of the board, and connect it with the manometer. (a) Take a blood-pressure tracing with the board horizontal. (6) Whilst the tracing is being taken, rotate the board so that the position of the animal becomes vertical, with the feet down. Mark on the tracing the moment when the change of position takes place. The pressure falls. Replace the dog in the horizontal position. The manometer regains its former level. Now rotate the board, till the animal is again vertical, but with feet up and head down, and observe the effect on the blood-pressure. The respiratory variations in the pressure are usually greater with feet down than with head down. Notice in both cases whether there is any change in the rate of the heart. (c) Take the board off the stands, lay it on a table, expose the femoral artery, and insert a cannula into it. Shift the axis so that it now lies below this cannula. Replace the board on the stands, and repeat (a) and (6). The fall of pressure will now take place in the head-down position. f In the feet-down position (with the cannula in the femoral artery) a rise of pressure in general takes place. But sometimes this is very small, and lasts only a few seconds, being succeeded by a fall, during which the heart-beats on the tracing are much weaker than before, since enough blood is not reaching the heart to enable it to maintain the pressure. In the feet-down position see whether the corneal reflex can be got. If not, as is likely, turn the animal into the head-down position. The reflex may now soon be obtained, and it may again disappear on putting the animal in the feet-down position. If the chloroform anaesthesia is light the reflex may not be abolished in the feet-down position, although strong respiratory movements may occur, owing to anaemia of the medulla oblongata. 26. Effects of Haemorrhage and Transfusion on the Blood-Pressure. — Anaesthetize a dog with morphine and ether, and insert a cannula into the trachea. Put a cannula into 'the central end of the carotid artery and another into the central end of the femoral artery. Then insert a cannula, which should have a piece of indiarubber tubing 2 to 3 inches in length on its wide end, into the central end of the femoral vein on the opposite side. In doing this more care is necessary than * A simple arrangement for this purpose is a board with a number of staples fastened in pairs into its lower surface, so that an iron rod can be pushed through any pair, and form a horizontal axis at right angles to the length of the board. The dog having been tied down, the rod is pushed through the pair of staples corresponding to the position of the cannula in the artery that is to be connected with the manometer. The projecting ends of the rod rest in two ordinary clamp-holders, fastened at a convenient height on two strong stands, whose bases are clamped to the end of a table. The other end of the board is supported by a piece of wood that rests on the floor, and can be re- moved when the board is to be rotated. f In 1 6 dogs the fall of pressure in the carotid in the feet-down position varied from 12 to 100 mm. of mercury; average fall, 44-4 mm. In 12 out of the 16 animals the rise of pressure in the head-down position varied from 2 to 36 mm. ; in i there was no change; in 3 there was a fall of 5 to 24 mm. PRACTICAL EXERCISES 215 in putting a cannula into an artery. Feel for the femoral artery, cut down over it, and with forceps or a blunt needle separate the femoral vein from it for about an inch. Pass two ligatures under the vein, and tie a loose loop on each. Put a pair of bulldog forceps on the vein between the ligatures and the heart. Now tie the lower (distal) liga- ture, and cut one end short. The piece of vein between it and the bulldog forceps is thus distended with blood, and this facilitates the next step. With fine-pointed scissors make a snip in the wall of the vein. The cannula is now pushed through the slit in the vein, and the upper ligature tied firmly round its neck. By the aid of a pipette, made by drawing a piece of glass tubing out to a long point, the canmila and rubber tube are then completely filled' with o'g per cent, salt solution. Be sure to pass the point of the pipette right down to the point of the cannula, so as to dislodge any bubble ol air that may tend to cling there. Then, holding up the open end of the rubber tube, close it, without allowing any air to enter, by means of a screw clamp or bulldog forceps, or a small piece of glass rod. Connect the cannula in the carotid with a manometer, arranged to write on a drum as in experiment 23 (p. 210). Take the bulldog off the carotid, and measure the difference in the level of the mercury in the two limbs of the man- ometer with a millimetre scale. (1) (a) While a tracing is being taken, draw off about 10 c.c. of blood from the femoral artery, and observe whether there is any effect on the tracing. Mark on the tracing the moment when the removal of the blood begins and ends. (b) Repeat (a), but run off about too c.c.* of blood, and let this be immediately defibrinated. Then draw off portions of 100 c.c.* at short intervals until a distinct fall of blood-pressure has been produced. All the samples of blood should be denbrinated and strained through cheese-cloth. (2) (a) Now, while a tracing is being taken, inject the whole of the denbrinated blood slowly through the cannula in the femoral vein by means of a funnel supported by a stand at such a height that the blood runs in easily. A pinchcock should be put on the tube connecting the funnel and the cannula, and this should be closed before the funnel is quite empty, so as to obviate any risk of air getting into the vein. Of course, the cannula and connecting-tubes must all be freed from air before injection is begun. Again measure the difference in the level of the mercury and compare the pressure with that observed before the first haemorrhage. (b) Inject into the vein, while a tracing is being obtained, about 100 c.c.* of o'9 per cent, salt solution heated to 40° C., and go on injecting portions of 100 c.c. until a distinct rise of pressure has taken place, keeping a record of the total amount injected, and marking the time of each injection on the curve. (c) After an interval of thirty minutes, again measure the height of the mercury in the manometer. Then bleed the dog to death while a tracing is being recorded. 27. The Influence of Proteoses (and Peptones) on the Blood-Pressure. — Set up the apparatus for taking a blood-pressure tracing as in experi- ment 23 (p. 2 TO), but omit the induction-coil. Weigh a dog. Weigh out a quantity of Witte's peptone equivalent to 0*5 grm. for every kilo of body-weight. Dissolve the peptone in about ten time's its weight of o'Q per cent, salt solution. Anaesthetize the dog with morphine and ether or A.C.E. mixture. Insert a cannula into the trachea. Put cannulae into the central end of one carotid and of one femoral vein * 200 c.c. for a large dog. 216 THE CIRCULATION OF THE BLOOD AND LYMPH (p. 214). Connect the carotid with the manometer, and the femoral vein with a burette or large syringe containing the peptone solution. Take care that the connecting-tube and cannula are free from air. Now commence to take a blood-pressure tracing, and while it is going on inject the peptone solution. The pressure falls owing largely to a dilatation of the small arteries through the direct action of the pep- tone on the ir muscular tissue or on the endings of the vaso-motor nerves.* 28. Effect of Suprarenal Extract on the Blood-Pressure. — Make the arrangements for a blood-pressure tracing from a dog as in 23 (p. 210). Put a cannula in the carotid and another in the femoral vein or one of its branches (p. 214). Expose both vagi in the neck, and pass threads loosely under them. Connect the carotid with the manometer and take a tracing. Then, while the tracing is continued, inject slowly into the femoral vein an amount of watery extract corresponding to about 0*2 grm. of suprarenal, or, what is more convenient, a few c.c. of a solution of adrenalin chloride of the strength of i to 50,000 in 0*9 per cent/ sodium' chloride solution, the dose depending, of course, on the size of the animal. The blood-pressure risesf owing to con- Fig. 106. — Effect of Injection of Peptone on the Blood-Pressure in a Dog. (To be read from right to left.) striction of the arterioles by direct excitation of the junction between their vaso-constrictor nerves and their muscular tissue. The heart is slowed, but its beat is strengthened. At once cut both vagi while a tracing is being taken; the blood-pressure rises still more (p. 655). The rise of pressure is sometimes so great that to prevent the mercury from being forced out of the manometer the tube must be clipped. The rise is not long maintained, but a second injection causes a renewed increase of pressure. 29. Action of Epinephrin (Adrenalin) on Artery Rings. — The experi- ment (8) described on p. 66 in connection with the constrictor action of serum may equally well be performed here. * In 12 dogs the blood-pressure always fell, the amount of the fall varying from 81 to 21 mm. of mercury (average, 60 mm.). It sometimes returned to normal in twenty to thirty minutes, but usually required a longer time. In some dogs, after the injection of the whole of this amount of peptone, death occurs before there has been any considerable recovery of the pressure. f The amount of the initial rise of pressure is very variable, since the slow- ing of the heart tends to diminish the pressure, while the constriction of the arterioles tends to increase it. Thus, in one experiment the increase of pres- sure on injection of the extract was only 6 mm. of mercury, while in another it was 56 mm. On section of the vagi in this second experiment, there was an additional rise of 64 mm., and after a second injection a further rise of 70 mm., making an increase of 190 mm. in all above the original pressure. PRACTICAL EXERCISES 2J7 30. Determination of the Circulation-Time. — (a) Begin with ail arti- ficial scheme (Fig. 107). Fill the syringe with a 0-2 per cent, solution of methylene blue. Allow the water to flow from the bottle by loosen- ing the clamp. Inject a definite quantity of the methylene-blue solu- tion, and with a stop-watch observe how long it takes to pass from the point of injection to the end of the glass tube filled with beads Make ten readings of this kind, and take the mean. Then raise the bottle so as to increase the ra' e of flow of the water, and repeat the observations. The ' circulation- time ' will be found to be diminished. ^a This corresponds to an increase of , .F\^ : blood- pressure due to increased act tivity of the heart, without change in the calibre of the bloodvessels. Next, leaving the bottle in its present position, diminish the outflow by tightening Ihe clamp; the circulation- time will be increased. This corres- ponds to an increase of blood-pressure due to diminution in the calibre of the small arteries. (b) Fill the syringe* with methy- lene-blue solution (0-2 per cent, in Fig. 107. — Artificial Scheme to illustrate a Method of measuring the Circulation- Time. B, bottle containing water, the rate of outflow of which is regulated by screw-clamp a; S, syringe filled with methylene-blue solution, connected with T-piece A; M, beaker containing methylene-blue solution; b, c, screw-clamps; C, T-piece, inserted in the course of the flexible tube E, and connected with the glass tube T, which is filled with beads; F, outflow tube. The clamp c having been closed and b opened, the syringe is filled with the methylene-blue solution; b is then closed, c opened, and a definite quantity of the solution injected into the system. The time from the beginning of injection till the appearance of the blue at G is measured with the stop-watch. 0-9 per cent, salt solution), as in (a). Keep the solution warmed to 40° C. by immersing the small beaker containing it in a water-bath, or heating it over a Bunsen with a small flame. Weigh a rabbit or cat. In the case of the rabbit, inject 1 grm. chloral hydrate into the rectum, and later on give ether if necessary. If a cat, give ether alone, or urethane (1-5 grm. per kilo by stomach tube i to 2 hours before). Fasten it on a holder, back downwards (Fig. 61, p. 136). Cover it with a towel to keep it warm. Clip off the hair on the front of the neck, and make an incision i J inches long in the middle line, beginning a little * A burette, sloped so as to make a small angle with the horizontal, may be substituted for the syringe. The burette is supported on a stand at such a height (say 10-15 cm. above the level of the caiinula) that the methylene- blue solution runs without great force into the jugular. The danger of pro- ducing an abnormal result by suddenly raising the pressure in the right side of the heart is thus avoided. 2i8 THE CIRCULATION OF THE BLOOD AND LYMPH way below the cricoid cartilage. Reflect the skin and isolate the external jugular vein, which is quite superficial. Carefully separate about £ inch of the vein from the surrounding tissue, and pass two ligatures under it. but do not tie them. Compress the vein with a pair of bulldog forceps between the heart and the ligatures. Now tie the uppermost of the two ligatures (that next the head), but only put a single loose loop on the other. The piece of vein between the upper ligature and the bulldog is now distended with blood. With fine- pointed scissors make a small slit in the vein, taking great care not to divide it completely, insert the cannula, and tie the loose ligature firmly over its neck. Fill the cannula and the small piece of rubber tubing attached to it with 0*9 per cent, salt solution by means of a pipette with a long point. Expose the carotid on the other side, isolate it lor f inch, clear it carefully from its sheath, slip under it a strip of thin sheet indiarubber, and between this and the artery a little piece of white glazed paper. Connect the cannula in the jugular with the T-piece attached to the syringe. Care must be taken that no air remains in the cannula or its connecting-tube, as a rabbit not unfrequently dies instantaneously when a bubble of air is injected into the right heart, although a considerable quantity of air can generally be injected into the jugular of a dog without killing it. Now take off the bulldog from the vein, and make a series of observa- tions on the pulmonary circulation-time. The animal must be so placed that a good light falls on the carotid. If necessary, the light of a gas-flame may be concentrated on it by a lens. The student holds the stop-watch in one hand, and injects a measured quantity of the methylene-blue solution with the other. Uniformity in the quantity injected is secured by fastening on the piston of the syringe a screw- clamp, which stops the piston at the desired point. The observation consists in setting off the watch at the moment when injection begins and stopping it when the blue appears in the carotid. After each injection the screw-clamp or pinchcock on the tube connected with the cannula must be tightened, the other opened, and the syringe refilled. Great care must be taken never to open the two clamps at the same time, as in that case blood may regurgitate through the jugular and fill the syringe, or methylene blue may be sucked into the circulation. As many observations as possible should be taken, and the mean determined. The circulation -time observed is approximately that of the lesser circulation, the time taken by the blood to pass from the left ventricle to the carotid being negligible for the purposes of the student. The specific gravity of the blood may also be tested at the beginning and end of the experiment by Hammerschlag's method (p. 62). If a large number of injections have been made in quick succession, the specific gravity will be less than normal; but if a considerable interval has been allowed to elapse after the last injection, little or no differ- ence may be found, as the surplus liquid readily passes out of the bloodvessels. Necropsy. — Observe particularly the state of the lungs, whether the bladder is distended or not, and whether any of the serous cavities or the intestines contain much liquid; so as to determine, if possible, by what channel the water injected into the blood may have been elimin- ated. Study the distribution of the methylene blue in such organs as the kidneys and the muscles immediately after death, and notice that the blue colour becomes more pronounced after exposure for a time to the air. Make a longitudinal section through a kidney, and observe that the pigment is found especially in the cortex and around the PRACTICAL EXERCISES 219 pelvis at the apices of the pyramids, or it may be only in the cortex. The urine is greenish. If some methylene blud has been injected after the heart ceased to beat, the bloodvessels, particularly in the mesentery, may be beautifully mapped out by the pigment. This is not the case if the last injection took place before death, since the methylene blue is rapidly reduced by living tissues to a colourless substance, leuco-methylene blue. 31. Measurement of the Blood-Flow in the Hands. — Arrange the calorimeters as in Fig. 108. The thermometers in the calorimeters should be graduated in tenths of a degree, so that by means of the small lenses or ' readers ' which slide on the stems hundredths of a degree can be estimated. Where it is desirable that a number of students should make observations in as short a time as possible, one calorimeter can be allotted to each subject, the other hand being kept in the pocket or covered with a glove if the room is cool, so as to avoid reflex vaso- motor interference. A felt collar is chosen which fits the wrist closely. A horizontal pencil-mark is made at the lower edge of the styloid process of the ulna, and another parallel mark at a distance above this slightly greater than the thickness of the collar. When this second mark is just kept in view above the collar with the hand in the calorimeter, the first (lower) mark will be jiut below the level of the lid. A large bath holding 20 or 30 litres or niore (a clean ' garbage ' or ' offal ' can is suitable) is filled with water at about 32° C. The exact temperature is not important, but it should be about the same in all measurements which are to be compared. An ordinary ther- mometer graduated in degrees is all that is necessary for reading the temperature of the bath. The calorimeters are now filled from the bath. They are conveniently made of such a size that 3 litres of water and the hand can be contained in them without any slopping over when the water is stirred. Time is saved by having a metal flask which just holds the quantity of water that goes into each calorimeter. The orifices of the calorimeters are closed by felt discs. The subject, sitting in a high chair placed between the calorimeters, now immerses his hands in the bath to a point between the two marks. The fingers are kept spread. The bath is occasionally stirred. An ordinary ther- mometer suspended at the back of the chair gives the room tempera- ture. After ten minutes the hands are withdrawn from the bath, the wrists rapidly dried with a towel, the hands at once introduced into the calorimeters, and the felt collars adjusted round the wrists. The sub- ject leans back comfortably in the chair, allowing the arms to hang down without effort. The fingers are kept slightly spread. The ob- server sits on a low seat behind the subject, and reads the thermometers from time to time, always after stirring the water well with goose- feathers passing through the stirring-holes in the lid. The readings can be made at intervals of a minute, two minutes, or any interval which is convenient. At the end the hands are quickly withdrawn, the felt discs put over the orifices, and the water vigorously. stirred for ten or fifteen seconds before the thermometers are read. In this way any errors due to imperfect stirring or to accidental contact of the hands with the thermometers are eliminated. The volume of each hand is now measured by immersing it exactly to the lower mark in water contained in a glass douche-can connected by a short rubber tube with a pipette furnished with a side-tube at its lower end. The lowest graduation on the burette (50 on a 50 c.c. burette) is brought level with the water before the hand is immersed, While the hand is being held steadily and vertically in the water by an assistant, the level of the water in the burette is read off. All that 220 THE CIRCULATION OF THE BLOOD AND LYMPH is necessary to get the volume of the hand is to pour water into the can from a graduated measure after withdrawal of the hand until the same level is reached. Or the value of a division of the burette can be determined once for all. The burette is simply used as a transparent scale. When the two hands are successively measured, the small amount of water removed by the first is automatically restored by dipping the second into a separate vessel of water, and putting it wet into the douche-can. The rectal t mperature should now be obtained. The temperature of the arterial blood entering the hand is taken as 0-5° C. below that of the rectum. If only the mouth temperature can be got, the thermometer should be put in a second time without shaking Fig. 108. — Calorimetric Method of measuring Blood-Flow in Hands. down to see if it rises any more. The mouth temperature is taken as equal to the arterial blood temperature. After thorough stirring, the calorimeter temperatures can now be read again. The two being noted, the amount of cooling of the calor- imeters can be determined. This has to be added to the actually observed rise of the thermometers during immersion of the hands. Suppose an experiment yielded the following data: Rise of ther- mometer in a calorimeter in twenty minutes during immersion of a hand in it, ro° C. ; temperature of calorimeter at beginning of the twenty minutes, 3i'o° C. ; at end of twenty minutes, 32-0° C. ; cooling of calor- imeter in twenty minutes, o'l8 C. ; water in calorimeter, 3,000 c.c.; PRACTICAL EXERCISES 221 volume of hand, 450 c.c.; rectal temperature, 37'o° C.; water equivalent of calorimeter, 100 c.c. The water equivalent of the hand is 450 x o-8*= 360 c.c. The water equivalent of the calorimeter is - 100 c.c. Water - - 3,000 c.c. Total - - 3,460 c.c. 3,46ox n = 3,8o6 small calories given off by the hand in twenty minutes. Temperature of arterial blood (36' 5°) minus temperature of venous blood (31*5*, the mean temperature of the calorimeter) = 5*0. 3,806 lot Flow per minute through hand=- x — ' =42*3 grm. 20 x 5 9 Flow per 100 c.c. of hand per minute = 9' 4 grm. The readings of the calorimeter thermometers for the first one or two minutes may not be usable, owing to disturbance caused by the intro- duction of the hands. As soon as they begin to rise steadily and uniformly, the readings can be utilized for the calculation of the flow. 32. Vasomotor Reflexes. — Begin as in 31,, Then, after the hands have been in the calorimeters for a sufficient period (say ten minutes) to allow satisfactory readings for the determination of the blood-flow to be obtained, rapidly transfer one hand to cold water (at about 8° C.), while the other remains in the calorimeter. Continue reading the calorimeter thermometer. Its rise will be checked by reflex vaso- constriction. If the hand is kept for a few minutes in the calorimeter, the reflex vaso-constriction of the hand in the calorimeter will probably disappear, and the thermometer will rise faster. When a sufficient number of readings have been obtained for calculating the alteration in the flow, which will usually be the case in eight or ten minutes, transfer the hand from the cold water to warm water (at about 43° C.), and continue reading the calorimeter thermometer. There is usually a reflex vaso-constriction followed by vaso-dilatation. * This factor is the product of the specific gravity and the specific heat of the hand. The volume multiplied by the specific gravity gives the mass of the hand, which multiplied by the specific heat, gives the water equivalent of the hand. t The reciprocal of the specific heat of blood (see formula, p. 122). CHAPTER IV RESPIRATION RESPIRATION in its widest sense is the sum total of the processes by which the ultimate elements of the body gain the oxygen they require, and get rid of the carbon dioxide they produce. SECTION I. — PRELIMINARY ANATOMICAL DATA. Comparative. — In a unicellular organism no special mechanism of respiration is needed; the oxygen diffuses in, and the carbon dioxide diffuses out, through the general surface. The simple wants of such multicellular animals as the coelente rates, the group to which the sea- anemone belongs, are also supplied by diffusion through the ectoderm from and into the surrounding water, and through the endoderm from and into the contents of the body-cavity and its ramifications. But in animals of more complex structure special arrangements become necessary, and respiration is divided into two stages: (i) Ex- ternal respiration, an interchange between the air or water and a cir- culating medium or blood as it passes through richly vascular skin, gills, trachea?, or lungs; and (2) internal respiration, an interchange between the blood, or lymph, and the cells. In the lower kinds of worms respiration goes on solely through the skin, under which plexuses of bloodvessels often exist, but in some higher worms there are special vascular appendages that play the part of gills. The Crustacea also possess gills, while in the other arthropoda respiration is carried on either by the general surface of the body (in some low forms), or more commonly by means of trachea, or branched tubes .surrounded by blood spaces and communicating externally with the air and internally by their finest twigs with the individual cells. Most of the mollusca breathe by gills, but a few only by the skin. Among vertebrates the fishes and larval amphibians breathe by gills, but most adult amphibians have lungs. The skin, too, in such animals as the frog has a very important respiratory function, more of the gaseous exchange taking place through it in some conditions than through the lungs. One small group of fishes, the dipnoi, has the peculiarity of possessing both gills and a kind of lungs, the swim-bladder being surrounded with a plexus of bloodvessels and taking on a respiratory function. In all the higher vertebrates the respiration is carried on by lungs; the trifling amount of gaseous interchange which can possibly take place through the skin is not worth taking into account. The lungs are to be regarded as developed from outgrowths of the alimentary canal, beginning near the mouth. 22* PRELIMINARY ANATOMICAL DATA 223 The object of all special respiratory arrangements being, in the first instance, to facilitate the gaseous exchange between the surrounding medium (air or water) and the blood, a prime necessity of a respiratory organ, be it skin, gill, trachea, or lung, is a free supply of blood, in vessels so fine and thin that diffusion readily takes place into them and out of them. But a free supply of blood would be of no avail if the medium to which the blood gave up its carbon dioxide and from which it drew its oxygen was not being constantly and sufficiently renewed. Sometimes the natural currents of the water or the air are of them- selves sufficient to secure this renewal; in other cases, artificial currents are set up by cilia, or special bailing organs, like the scaphognathites of the lobster. In all the higher animals, active movements by which air or water is brought into contact with the respiratory surfaces, are necessary; and it is possible that such movements take place even in the tracheae of insects and other air-breathing arthropoda. Fishes, by rhythmical swallowing movements, take in water through the mouth and pass it over the gills and out by the gill-slits, while the frog distends its lungs by swallowing air. Physiological Anatomy of the Respiratory Apparatus.— In man the respiratory apparatus consists of a tube (the trachea) widened at its upper part into the larynx, which contains the special mechanism of voice, and communicates through the nose or mouth with the external air. Below, the trachea divides dendritically into innumerable branches, the ultimate divisions of which are called bronchioles. Each bronchiole ends in several openings or vestibula, each of which in turn leads into a dilatation called an atrium. From each atrium are given off two, or more, often funnel-shaped diverticula, the infundibula, the walls of which are everywhere pitted with recesses or alcoves, called alveoli. The atria are also lined with alveoli. The infundibula (with the atria) constitute the essential distensible elements of the lung, by the alternate stretching and relaxation of which the respiratory changes in the volume of the organ are mainly brought about. The trachea and bronchi are strengthened by hyaline cartilage which renders them rela- tively rigid. But the fact that the cartilage does not form complete rings permits small changes of calibre to take place. In the bronchioles, no cartilage is present, but the circularly-arranged muscular fibres still persist, and also form a thin layer in the infundi- bula. In the air-cells, or alveoli, however, there are no muscular fibres. Their walls consist essentially of a network of elastic fibres, continuous with a similar layer in the infundibula and bronchioles, and covered on the side next the lumen by a single layer of large, clear epithelial scales, with here and there a few smaller and more granular polyhedral cells. From the larynx to the bronchioles the mucous membrane is ciliated on its free surface, the cilia lashing upwards so as to move the secre- tion towards the larynx and mouth. In the infundibula the ciliated epithelium begins to disappear, and is absent from the alveoli. Part of the nasal cavity and the upper part of the pharynx are also lined with ciliated epithelium. Mucous glands are present in abundance in the upper porticn; of the respiratory passages, but disappear in the smaller bronchi. Blood-Supply of the Lungs. — The quantity of blood traversing the lungs bears no proportion to the amount required for their actual nourishment. Small, however, as this latter quantity is, it cannot apparently be derived from the vitiated blood of the right ventricle, but is obtained directly from the aortic system by the bronchial arteries. These are distributed with the bronchi, which they supply as well as the connective tissue of the interlobuiar septa running through the 224 RESPIRA TION substance of the lung, the pleura lining it and the walls of the large bloodvessels. Most of the blood from the bronchial arteries is returned by the bronchial veins into the systemic venous system, but some of it finds its way by anastomoses into the pulmonary veins. The branches of the pulmonary artery are also distributed with the bronchi, and break up into a dense capillary network around the alveoli. From the capillaries veins arise which, gradually uniting, form the large pulmonary veins that pour their blood into the left auricle. The same quantity of blood must, on the whole, pass per unit of time through the lesser as through the greater circulation, otherwise equilibrium could not exist, and blood would accumulate either in the lungs or in the systemic vessels. But it does not follow that at each heart-beat the output of the two ventricles is exactly equal. If, indeed, the capacity of the lesser circulation were constant, the quantity driven out at one systole by the right ventricle would be the same as that ejected at the next by the left ventricle. But it is known that the capacity of the pulmonary vessels is altered by the movements of respiration and probably in other ways, so that it is only on the average of a number of beats that the output of the two ventricles can be supposed equal. The time required by a given small portion of blood — e.g., by a single corpuscle — to complete the round of the lesser circulation, is, as we have seen (p. 137), much less than the average time needed to complete the systemic circulation. In man the ratio is probably about i : 5. Since all the blood in a vascular tract must pass out of it in a period equal to the circulation time, the average quantity of blood in the lungs and right heart of a man would thus be about one-fifth of that in the systemic vessels. That is to say, not less than 700 grm. out of the 4^ kilos* of blood in a yo-kilo man would be contained in the lesser cir- culation, and about 3^ kilos in the greater. This corresponds sufficiently well with calculations from other data. For example, the average weight of the lungs in three persons exe- cuted by beheading,' was 457 grm. (Gluge). The average weight of the lungs in a great number of persons who had died a natural death was 1,024 grm. (Juncker). The weight of the pulmonary tissue alone in the first set of cases must be less than 457 grm., for the lungs of a person who has bled to death are never bloodless. In a dog killed by bleeding from the carotid, one-quarter of the weight of the lungs con- sisted of blood. Assuming the same proportion for the decapitated individuals, we get 343 grm. as the net weight of the blood-free lungs. Deducting this from 1,024 gmi., we arrive at 681 grm. as the average quantity of blood in the lungs. Adding to this the quantity in the right side of the heart (p. 140), we get, in round numbers, 750 grm. as the amount in the lesser circulation. It is true that in the living body the conditions are not the same as after death ; but it is probable that in a large number of cases taken at random the differences would be approximately equalized. It has been further calculated that the total area of the alveolar surface of the lungs of a man is about 100 square metres (sixty times greater than the area of the skin), of which, perhaps, 75 square metres are occupied by capillaries. The average thickness of this immense sheet of blood has been reckoned to be equal to the diameter of a red blood-corpuscle, or, say, 8 u. This would give 600 c.c. (630 grm.) as the quantity of blood in tne lunge, which is probably somewhat too low an estimate. * See footnote on p, 139 MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 225 If we take the pulmonary circulation-time as 13 seconds (p. 137). o* /"r x 60 x (3o and the quantity of blood in the lungs as 700 grm., then - = 194 kilos of blood will pass through the lungs in an hour, or 4,656 kilos (say, 4,400 litres) in twenty-four hours. This would fill a cubical tank in which the man could almost stand upright with the lid closed. SECTION II. — MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION. The lungs are enclosed in an air-tight box, the thorax; or it may be said with equal truth that they form part of the wall of the thoracic cavity, arid the part which has by far the greatest capacity of adjustment. The alveolar surface of the lungs is in contact with the air. The pleura, which covers their internal surface, is reflected over the chest-walls and diaphragm, so as to form two lateral sacs, the pleural cavities. In health these are almost obliterated, and the visceral and parietal pleurae, separated and lubricated by a few drops of lymph, glide on each other with every movement of respiration. But in disease the pleural cavities may be filled and their walls widely separated by exudation, as in pleurisy, or by blood, as in rupture of an aneurism, or by air in -the condition known as pneumo-thorax. Between the two pleural sacs lies a mesial space, tne mediastinum, commonly divided into an anterior medias- tinum in front of the heart, and a posterior mediastinum behind it. The pleural and pericardial sacs and the mediastinum constitute together the thoracic cavity. The external surface of the chest- wall and the alveolar surface of the lungs are subjected to the pressure of the atmosphere, to which the pressure in the thoracic cavity (intra-thoracic pressure) would be exactly equal if its bound- aries were perfectly yielding. But in reality the intra-thoracic pressure is always normally something less than this. For even the lungs, the least rigid part of the boundary, oppose a certain resistance to distension, and so hold off, as it were, from the thoracic cavity a portion of the alveolar pressure; and in any given position of the chest the intra-thoracic pressure is equal to the atmospheric pressure minus this elastic tension of the lungs. The object of the respiratory movements is the renewal of the air in contact with the alveolar membrane — in other words, the ventila- tion of the lungs. Two main methods are followed by sanitary engineers in the ventilation of buildings: they force air in, or they draw it in. In both cases the movement of the air depends on the establishment of a slope of pressure from the inlet to the interior. In the first method, this is done by increasing the pressure at the inlet; in the second, by diminishing the pressure at the outlet. In certain animals Nature, in solving its problem of ventilation, has made use of the first principle. Thus, the frog forces air into its 15 226 RESPIRA TION lungs by a swallowing movement. In artificial respiration, as practised in physiological experiments, the same method is usually employed: air is driven into the lungs under pressure. But in the vast majority of air-breathing animals, including man, the opposite principle has been adopted ; and the ' indraught ' of air from nose and pharynx to alveoli is not set up by increasing the pressure in the former, but by diminishing it in the latter. This ' indraught,' or inspiration, is brought about by certain movements of the chest- wall, which increase the capacity of the thoracic cage and lower the pressure in the thoracic cavity. The expansion of the highly- distensible lungs keeps pace with the diminution of pressure in the pleural sacs, and they follow at every point the retreating chest -wall and diaphragm, although they do not expand equally in all directions. The dorsal surface in con- tact with the vertebral column, the mediastinal surface in contact with the pericardium and the contents of the mediastinum, and the surface of the apex, move but little. The surfaces in contact with the diaphragm, ribs, and sternum have the greatest range of movement. Intermediate portions of the parenchyma of the lungs expand in a degree determined by their distance from the relatively stationary and mobile surfaces. The pres- sure of the air in the alveoli during the rapid expansion of the lungs necessarily sinks below that of the atmosphere, and air rushes in through the trachea and bronchi till the difference is equalized. Then commences the movement of ex- piration. The expanded chest falls back to its original limits; the pressure in the thoracic cavity increases; the distended lungs, in virtue of their elasticity, shrink to their former volume; the pressure of the air in the alveoli rises above that of the atmo- sphere, and with this reversal of the slope of pressure air streams out of the bronchi and trachea. In inspiration the chest dilates in all its diameters. Its vertical diameter is increased by the contraction of the diaphragm, which, composed of a central tendon, a peripheral ring of muscular tissue, and the two muscular crura, bulges up into the thorax in the form of two flattened domes, one on each side, and thus closes its lowe* Fig. 109. — Scheme to illus- trate the Movements of the Lungs in the Chest. T is a bottle from which the bottom has been removed; D, a flexible and elastic membrane tied on the bottle, and capable of being pulled out by the string S so as to increase the ca- pacity of the bottle. L is a thin elastic bag repre- senting the lungs. It com- municateswith the external air by a glass tube fitted airtight through a cork in the neck of the bottle. When D is drawn down, the pressure of the external air causes L to expand. When the string is let go, L con- tracts again, in virtue of its elasticity. MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 227 aperture. When the diaphragm contracts, even in ordinary quiet breathing, the central tendon descends distinctly (about half an inch) after the manner of a piston. The acute angle which the muscular ring makes during relaxation with the thoracic wall opens out around its whole circumference, so as to form a groove of trian- gular section. But the most peripheral portion of the ring is always kept in close apposition to the chest-wall by the negative intra- thoracie pressure. The lungs follow the descending diaphragm, their lower borders keeping accurately in contact with it. The descent of the diaphragm is not directly downwards, but downwards and forwards. For it is compounded of two movements, the spinal segment of the muscle (the crura) causing a vertical elongation of the thorax, while the sterno-costal part (the muscular ring) pushes the abdominal viscera downwards and forwards (Keith). Since the diaphragm is attached to the lower ribs, there is a tendency during its contraction for these to be drawn inwards and upwards; but this is opposed by the pressure of the abdominal viscera, and by the action of the quadratics lumborum, which fixes the twelfth rib, and of the serratus posticus inferior, which draws the lower four ribs backward. When these and the other inspiratory muscles that act especially upon the ribs are paralyzed by injury to the spinal cord, and respiration is carried on by the diaphragm alone, the line of its attachment to the ribs is distinctly marked during inspiration by a shallow circular groove. The thorax is also enlarged by the action of certain muscles that act upon the ribs. Among the elevators of the ribs, as their name indicates, are usually reckoned, although erroneously, the levatores costarum — twelve in number on each side. They arise from the transverse processes of the last cervical and first eleven dorsal vertebrae, and passing obliquely downwards and outwards, are in- serted between the tubercle and the angle into the first or second rib below their origin. They do not elevate the ribs, but take part in lateral movements of the spinal column. The scalene muscles, which may in a lean person be felt to be tense during inspiration, fix the first and second ribs (scalenus anticus and medius, the first ; scalenus posticus, the second rib), and so afford a fixed line for the intercostal muscles to work from on the lower ribs. The most important elevators of the ribs are the external inter- costals. The intercartilaginous portions of the internal inter- costals (the intercartilaginei muscles, as they are sometimes called) also contract simultaneously with the diaphragm, and may there- fore be included in the list of inspiratory muscles; but instead of elevating the ribs they depress the costal cartilages, and 'thus help to widen the angles between them and the ribs. In addition to increasing the capacity of the chest, the contraction of the external intercostals and the intercartilaginous muscles aids in inspiration 228 RESPIRATION by augmenting the rigidity of the intercostal spaces, and so pre- venting them from being drawn in as easily as would otherwise be the case when the thorax is expanded by the action of the dia- phragm and the other inspiratory muscles. Leaving out of account the floating ribs, which functionally form a part of the abdominal wall, the ribs in relation to their respiratory functions may be divided into the following groups: (i) The first rib, which, moving itself very little, provides a fixed line towards which the next set of ribs may be raised. (2) An upper costal series consisting of the ribs from the second to the fifth. These are raised in inspiration towards the fixed first rib by the contraction of the intercostal muscles. The movement of these ribs is, mainly at any rate, a rotation around a transverse axis, the axes on which they move corresponding to their necks. The manner in which they are articulated to the vertebrae prevents any sensible rotation around an antero-posterior axis or ' bucket- handle ' movement. Since these ribs slant downwards and forwards to their sternal attachments, the sternum is raised when they are elevated; or, rather, since the manubrium is practically immovable in ordinary breathing, the body of that bone is bent on the manu- brium at the manubrio-sternal joint. This causes an increase in the antero-posterior diameter of the thorax. Further, since the arches formed by the ribs widen in regular progression from above downwards in the upper portion of the thoracic cage, so that the second rib is a segment of a larger circle than the first, and the third than the second, it is clear that a general elevation of the chest will tend to increase the transverse diameter at any given level. Such an increase is also favoured by the opening out of the angles between the bony ribs and the costal cartilages under the influence of the couple (or pair of oppositely directed forces) that acts on them — viz., the upward pull of the external intercostals exerted on the ribs, and the downward pull of the intercartilaginei and the resist- ance of the sternum to further displacement exerted on the carti- lages. The whole arrangement is perfectly adapted to permit the expansion of the roughly conical upper lobes of the lungs. (3) The lower costal series, consisting of the ribs from the sixth to the tenth. These ribs, with their muscles, form a mechanism which normally acts along with the diaphragm (Keith). They are so arranged that in inspiration the lateral and anterior part of each moves outwards to a greater extent than the one above it. There is not only a rotation around a transverse axis, by which the lower end of the sternum, connected to these ribs by the combined cartilages of the sixth to the ninth, is elevated, but also a rotation around an antero-posterior axis. The movement of the lower ribs results, therefore, in increasing both the back-to-front diameter and the transverse diameter of the lower portion of the thorax. The MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 229 widening of the thorax from side to side may also be in a slight degree ascribed to a twisting movement of the ribs, which tends to evert their lower borders. With the diaphragm, these lower ribs arranged in a vertical series of not very different curvature con- stitute a mechanism for the inspiratory expansion of the roughly cylindrical lower lobes of the lungs. Expiration in perfectly tranquil breathing is brought about with less aid from active muscular contraction. The sense of effort disappears as soon as the chest ceases to expand. The diaphragm and the elevators of the ribs relax. The structures that have been stretched or twisted recoil into their original positions; the struc- tures that have been raised against the force of gravity fall back by their weight, and in the measure in which the pressure increases in the thoracic cavity the elasticity of the lungs causes them to shrink. The pressure in the alveoli, which at the end of inspiration was just equal to that of the atmosphere, is thus increased, and the air expelled. It is probable that, even in man and in quiet respira- tion, the interosseous portions of the internal intercostals help by their contraction in depressing the ribs, and that a slight contrac- tion of the abdominal muscles hastens the return of the diaphragm to its position of rest. In reptiles and birds, expiration is normally effected by an active muscular contraction. This is also true in some mammals — the rabbit, for instance, in which the external oblique muscles of the abdominal wall take an important share in the expiratory act. Types of Respiration. — Differences exist also, not only between different groups of animals, but even between women and men, in the relative importance in inspiration of the diaphragm and the muscles that raise the lower ribs on the one hand, and the muscles that elevate the upper ribs on the other. When the movements of the diaphragm predominate, the respiration is said to be of the abdominal or diaphragmatic type ; when the movements of the upper ribs and sternum are most conspicuous, of the costal or thoracic type. In abdominal respiration, the inspiratory movement commences at the diaphragm, and then involves the lower ribs and the tip of the sternum. In costal respiration, the upper ribs initiate the move- ment, and are followed by the abdomen. In the rabbit, during quiet breathing, the respiration is purely diaphragmatic, the ribs remain motionless; and herbivorous animals in general conform more or less closely to this type. In the carnivora, on the contrary, the costal type prevails. Man allies himself as regards his respira- tion with the rabbit and the sheep ; he uses his diaphragm more than his upper ribs. Civilized woman falls into the class of the wolf and the tiger; she uses her upper ribs more than her diaphragm. The cause of the difference between men and women has been much discussed. It is not a primitive sexual difference, for it is far from 23o RESPIRATION being universal; in the uncivilized and semi-civilized races that have been investigated, the women breathe like the men. It is therefore probable that the predominance of the costal type among women of European race is a peculiarity developed by a mode of dressing which hampers the movements of the diaphragm \vhilc permitting the elevation of the ribs. This conclusion is strengthened by the fact that in children no difference exists ; both boys and girls show the abdominal type of respiration. All this refers to ordinary breathing. In forced respiration, when the need for air becomes urgent, costal breathing always becomes prominent alike in men, in women, and in animals, for by elevation of the ribs the capacity of the chest can be increased to a greater degree than by any contraction of the diaphragm. In forced inspiration, indeed, all the muscles that can elevate the ribs may be thrown into contraction, as well as other muscles which give these fixed points to act from. During a paroxysm of asthma, for example, the patient may grasp the back of a chair with his hands, so as to fix the arms and shoulders and allow the pectorals and serratus magnus to raise the ribs. Similarly in forced expiration all the muscles are used which can depress the ribs, or increase the intra-abdominal pressure and push up the diaphragm. Artificial Respiration. — An efficient pulmonary ventilation can be obtained by various methods when the natural breathing is in abey- ance. In animals the method most commonly employed for ex- perimental purposes is the rhythmical inflation of the lungs by a pump or bellows, or by a stream of compressed air which is regularly interrupted, the chest being allowed to collapse after each inflation. When the animal is to be kept alive after the experiment the inflation is produced through a tube introduced through the glottis. If the animal is not to be kept alive, the apparatus is generally connected with a cannula in the trachea. When it is desired to avoid move- ments of the lungs, respiration may be maintained by a stream of oxygen through a catheter passed down the trachea (method of insufflation). In man the exchange of air between the atmosphere and the lungs may be most readily accomplished by strong rhythmi- cal compression of the lower part of the chest. This forces out some of the air from the lungs; on relaxing the pressure the chest expands again and air is drawn in. Schafer has shown that this is the most efficient method of respiration in resuscitation of the ap- parently drowned. ' The patient is placed face downwards on the ground, with a folded coat under the lower part of the chest. The operator puts himself athwart or at the side of the patient, facing his head and kneeling upon one or both knees (Fig. no), and places his hands on each side over the lower part of the back (lowest ribs) . He then slowly throws the weight of his body forward to bear upon his own arms, and thus presses upon the thorax and forces air out of the lungs. He then gradually relaxes the pressure by bringing his MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 231 own body up again to a more erect position, but without moving the hands.' Air is thus drawn into the lungs. The process is repeated twelve to fifteen times a minute. Certain accessory phenomena (movements and sounds) are asso- ciated with the proper movements of respiration. The larynx rises in expiration, and sinks in inspiration. The glottis (and particu- larly its posterior portion, the glottis respiratoria) is widened during deep inspiration and narrowed during deep expiration. The same is the case with the nostrils, and, indeed, iri some persons the alae nasi move even in ordinary breathing. It has long been known that in deep respiration changes in the calibre of the bronchi syn- chronous with the respiratory movements may occur. In young persons it may be directly observed with the bronchoscope, an instrument used by laryngologists for exploring the larger bronchi, that these dilate in inspiration and constrict in expiration (In- galls). In part at least these movements are passively pro- duced by the changes of intra- thoracic pres- sure, but it has not been defi- nitely deter- mined whether they are not in part caused by alternate contraction and relaxation of the circular bronchial muscles. To these muscles has sometimes been attributed the function of regulating the flow of air into and out of the infundib- ula, as the muscle of the arterioles regulates the distribution of the blood in the organs. As regards the respiratory sounds, all that is necessary to be said here is that when we listen over the greater portion of the lungs with the ear, or, much better, with a stethoscope, a soft breezy murmur, that has been compared to the rustling of the wind through distant trees, is heard. This has been called the vesicular murmur. It is only heard in health during inspiration and the very beginning of expira- tion, and is louder in children than in adults. Around the larger bronchi and the trachea a blowing sound is heard, which certainly originates at the glottis, and is strengthened by the resonance of the air-tubes. In health this is not recognized over the greater portion of the lung. But in certain diseases in which the alveoli are devoid of air, whether from compression or because they are filled up with exudation, and in other conditions, this bronchial or tubular breathing Fig. no. — Artificial Respiration in Cases of Drowning (after Schafer). 232 RESPIRATION may be heard over the affected area. The bronchi themselves, how- ever, must still be patent and contain air. The most commonly accepted explanation is that the laryngeal sound is better conducted through the smaller bronchi towards the surface of the lungs when their walls have been rendered more rigid by the solidification of the parenchyma, in spite of the fact that the consolidated tissue as such docs not conduct the sound so well as the air-containing alveoli. It seems probable that, in addition, the columns of air in the bronchi, which arc encased in solid tissue, may actually increase the intensity of the transmitted laryngeal murmur by resonance. It has been much debated whether the vesicular murmur also arises at the glottis, and is modified by transmission through the pulmonary tissue, or whether it arises somewhere in the terminal bronchi, the infundibula or the alveoli. Both views may be supported by certain arguments, and to both some objections may be raised, 'the fact appears to be that there are two elements in the inspiratory murmur — a true vesicular sound, produced about the place where the terminal bronchioles give off the infundibula, and a resonance sound set up in the trachea and bronchi by the glottic murmur. This resonance sound as heard over portions of the lung containing only small bronchi has a different character from that heard over large bronchi, inasmuch as the fundamental note, and to a still greater extent the overtones ]p. 310), are much weakened in those small and easily-distensible tubes. The true vesicular element is heard all over the lungs, but the resonant laryngeal element in large animals, like the horse and ox, dies out as an audible murmur before it reaches the remotest lobules, and can only be distinguished over a portion of the pulmonary area. When the glottic sound is eliminated by causing an animal to breathe through a tracheal fistula, the vesicular murmur is still heard, and in the horse is even somewhat sharper than normal, although in the dog it is softer and weaker. The expiratory murmur does not seem to contain a true vesicular element, but is exclusively due to the resonance of the expiratory glottic sound (Marek). It is ge'nerally admitted, and this is of great importance in practical medicine, that when the normal vesicular sound is heard over any portion of the lung tissue, it may be inferred that this portion is. being properly distended, and that air is freely entering its alveoli. Up to this point we have contented ourselves with a purely qualitative description of the mechanical phenomena of respiration. We have now to consider their ^ quantitative relations, and the methods by which these have been studied. The expansion of the lungs in inspiration may be easily demonstrated in man, and even a rough estimate of its amount obtained, by the clinical method of percussion. For example, the resonant note that is elicited when a finger laid on the chest at a part where it overlies the right lung is smartly struck can be followed down until it is lost in the ' liver dulness.' If the lower limit of the resonant area be marked on the chest -wall first in full inspiration, and then in full expiration, the mark will be lower in the former than in the latter, and the difference will represent the difference in the vertical length of the shrunken and dis- tended lung. A similar enlargement in the transverse direction may be demonstrated in the same way, the inner borders of the lungs coming nearer to the middle line in inspiration, and receding from it in expira- tion. The examination of the chest by the Rontgen rays has also MECHANICAL PHENOMENA OP EXTERNAL RESPIRATION 233 yielded results of importance in the study of normal respiratory con- ditions, and still more important results in pulmonary disease. For most physiological purposes, however, a faithful graphic record of the respiratory movements is indispensable. This may be obtained — (1) By registering the movements of a single point, or the variations in a single circumference, of the bound- ary of the thoracic cavity. In man changes in the circumference of the thorax at any level can be recorded by means of a tambour adjusted to the chest (Figs, in and 134), and in communication with another, which is provided with a writing lever (Figs. 99 and 137). Or an elastic tube, with a spiral spring in its lumen, may be fastened around the thorax or abdomen and connected with a piston-recorder (a small cylin- der in which works a piston carrying a writing-point) (Fitz). (2) By recording the changes of pressure produced in the air-passages by the respiratory movements. This can be done by connecting a cannula in the trachea of an animal with a recording tambour in the manner described in the Practical Exercises (p, 3oo). The variations of pressure may be measured by connecting a manometer with the trachea, or in man with the nostril. (3) ; By writing off the changes of pressure which occur in the thoracic cavity during respiration. For this purpose a trocar (Fig. 113) is intro- duced through an intercostal space into one of the pleural sacs, without the admission of air, or into the pericardium, and then connected with a manometer or other recording apparatus. Or a tube, similar in con- struction to a car- diac sound (p. 96), may be pushed down the cesopha- gu?. The varia- tions in the intra- thorrxic pressure are transmitted to the air in the elas- tic bag, and thence to a tambour. (4) In the rabbit the part of the dia- phragm attached to the ensiform cartilage may be isolated from the rest and its contractions recorded by a lever (Head). For scone purposes this is the best method. When the respiratory movements are studied in any of these ways, it is found that there is practically no pause between the end of Fig. in. — Scheme of Tambour for recording Respiratory Movements. C, a metal capsule connected airtight with B, A, two caoutchouc mem- branes, the chamber formed by which can be inflated by means of the tube and stopcock E. The tube D con- nects the space H with a registering tambour provided with a lever. The membrane A is applied to the chest, round which the inextensible strings Fare tied. At every expansion of the chest the pressure in H is increased, and the increase of pressure is trans- mitted to the registering tambour. Fig. ii2.— Respiratory Tracing from Man (Marey). Down stroke, inspiration; up stroke, expiration. 234 RESPIRATION inspiration and the beginning of expiration. Nor, although the chest collapses more gradually than it expands, is there any distinct interval in ordinary breathing between the end of expiration and the beginning of the succeeding inspiration. When, however, the respiration is unusually slow, an actual pause (expiratory pause) may occur at this point. Expiration takes somewhat longer time than inspiration, the ratio varying from 7 : 6 to 3 : 2, according to age, sex, and other circumstances. The frequency of respiration is by no means constant even in health. All kinds of in- fluences affect it. It is difficult even to direct the attention to the respiratory act without bringing about a modification in its rhythm. In the adult 15 to 20 respirations per minute may be taken as about the normal. In young children the frequency may be twice as great (new-born child, 50 to 70; child from i to 5 years old, 20 to 30 per minute). It is greater in a female than in a male of the same age. A rise of temperature increases it; 150 respira- tions per minute have been seen in a dog with a high temperature. Sudden cooling of the skin, exercise, and various emotional states, increase the rate, and sleep diminishes it. The will can alter the frequency and depth of respiration for a time, and even stop it altogether, but in less than a minute, in ordinary individuals, the desire to breathe becomes imperative. Cato's assertion that he could kill himself at any time ' merely by holding his breath ' is only a proof that he was a better philosopher than physi- ologist. After a period of forced respiration the breath can be held for a much longer time. This is due to the ' washing out ' of the carbon dioxide, the normal stimulus to the respiratory centre (p. 281). After six minutes of forced breathing the interval of voluntary inhibition can be extended be- yond four minutes. A professional diver has remained under water in a tank for about four and three-quarter minutes. When oxygen is inhaled instead of air during the last few breaths of the forced respiration, the interval during which the breath can be held may be much increased (up to nine or ten minutes). In animals the rate of respiration can be greatly affected by drugs and by the section and stimulation of certain nerves ; but to this we shall return when we come to consider the nervous mechanism of respiration. L-- Fig. 113. — Simple Pleural Cannula. B, a line of small spurs which, after the can- nula C has been pushed without ad- mission of air through an intercostal space into the pleural cavity, stick in the parietal pleura and securely fasten the cannula. Traction being made on the cannula, a ligature is tied at L around the protruding tissue for greater security. S, side-tube by which the cannula is connected with a manometer or tambour. MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 235 It cannot fail to be observed that to a great extent the rate of respiration is affected by the same circumstances as the frequency of the heart (p. 107), and in the same direction. And, indeed, in health, these two physiological quantities, amid all their absolute variations, maintain to each other a fairly constant ratio (i to 4 or f _ -.._._ i to 5 in man). Even in many diseases this proportion remains tolerably stable, although in others it is dis- turbed. The total quantity of air expired, or, what comes to the same thing, the alteration in the capacity of the chest during expiration, can be measured by means of a gas-meter or of a spiro- meter (Fig. 114), which consists of an inverted graduated glass cylinder dip- ping by its open mouth into water and balanced by weights. The vessel is sunk till it is full of water, the air being allowed to escape by a cock. The expired air is now permitted to enter it through a tube, and displaces some of the water. The spirometer is adjusted so that the level of the water inside and outside is the same, and then the volume of air contained in it is read off. This gives the volume of the expired air at atmospheric pressure. Similarly, by breathing air from the spirometer the amount inspired can be measured (p. 303). From 400 to 500 c.c. of air* are taken in and given out at each respiration in quiet breathing. This is called tidal air. It amounts to 35 pounds by weight I Fig. 114. — Diagram of Spirometer. A, vessel filled with water. B, glass cylinder with scale C, swung on pulleys and counter- poised by weights W. D, tube for breathing through. Vital Copacit(j\ ear Tidal _.. Supplemental air Residual air Fig. 115. — Diagram to illustrate the Relative Amount of Complemental, Tidal, Supplemental, and Residual Air. in twenty-four hours, or enough to fill, at atmospheric pressure, a cubical box with a side of 8 feet. With the deepest possible in- spiration room can be made for 2,000 c.c. more ; this is called complemental air. By a forced expiration 1,500 c.c. can be expelled besides the tidal air ; and to this * The average for 81 healthy students, with an average body-weight of 66 kilos, was 460 c.c., or 7 c.c. per kilo. In 4 new-born children the tidal air varied from 20 to 30 c.c., and from y6 to 7' 3 c.c. per kilo, which is not very different from the amount in the adult. The pulmonary ventilation must therefore be far more rapid in the child, since its respiratory frequency is so much greater. 236 RESPIRA TION quantity the name of supplemental or reserve air has been given. After the deepest expiration there always remains 1,000 to 1,200 c.c. of air in the lungs (Durig), and this is called the residual air. After a normal expiration following a normal inspiration the lungs still contain stationary air to the amount of about 2,500 c.c. The term vital or respiratory capacity is applied to the quantity of air which can be expelled by the deepest expiration following the deepest inspiration, and amounts in an adult of average height to 3,500 or 4,000 c.c. The maximum quantity of air which the lungs can contain is evidently equal to vital capacity plus residual air. At one time the vital capacity was thought to be capable of affording valuable information in the diagnosis of chest diseases; but little stress is now laid upon it, as it varies from so many causes. For instance, it can be increased by practice with the spirometer. It is greater in mountaineers than in the inhabitants of lowland plains. The Dead Space. — It is clear from the figures we have given that in ordinary breathing only a small proportion of the air in the lungs comes in direct at each inspiration from the atmosphere, and only a small proportion escapes into the atmosphere at each expiration. The greater part of the air in the lungs is simply moved a little farther from the upper respiratory passages, or a little nearer them; and fresh oxygen reaches the alveoli, as carbon dioxide leaves them, mainly by diffusion, aided by convection currents due to inequali- ties of temperature, and to the churning which the alternate expan- sion and shrinking of the lungs, and the pulsations of their arteries must produce. But that some of the tidal air strikes right down to the alveoli is evident enough. For the respiratory ' dead space ' — that is, the capacity of the upper air-passages and the bronchial tree down to the infundibula — is only 140 c.c., or one-third of the amount of the tidal air (Zuntz, Loewy). There is no direct way of determining whether any respiratory exchange goes on through the walls of the upper air-passages. But by indirect methods it has been estimated that about 30 per cent, of the volume of the tidal air is pure air (Haldane and Priestley). This, of course, corresponds to the ' effective ' dead space. Taking the average tidal air at 460 c.c. (p. 235), it is clear that the effective corresponds very closely with the anatomical dead space— that is to say, the respiratory function of the air-passages above the point where the infundibula are given off is negligible. Although such calculations can only b2 approximately correct, the agreement is of interest. Some observers have stated that great variations occur in the size of the dead space with changes in the depth of respiration, its volume being increased four or even eight times by very deep breathing. This, however, is incorrect, although with maximum expansion of the lungs the increase may amount to 100 c.c. (Krogh and Lindhard, Pearce). The Amount and Variations of the Intrathoracic Pressure. — In the deepest expiration the lungs are never completely collapsed ; their MECHANICAL PHENOMENA OF EXTERNAL RESPIRATION 237 elastic fibres are still stretched; and the tension of these acts in the opposite direction to the external atmospheric pressure, and dimin ishes by its amount the pressure inside the thoracic cavity. In th^ dead body Bonders measured the value of this tension, and there- fore of the negative pressure of the thorax, by tying a manometer into the trachea, and then causing the lungs to collapse by opening the chest. It varied from 7-5 mm. of mercury in the expiratory position to 9 mm. in the inspiratory. So far as can be judged from observations made on persons suffering from various diseases of the respiratory organs, the alterations during ordinary breathing do not Fig. 116. — Variations of Intraihoracic Pressure. Upper curve>*carotid blood-pres- sure (dog) lower curve, intrapleural pressure. At 42 the trachea was closed ; the blood-pressure curve shows the rise of asphyxia, and the intrapleural curve, greatly exaggerated pressure variations due to the strong and slow but abortive respirations. amount to more than 3 or 4 mm. of mercury. But when an attempt is made in the dead body to imitate a deep inspiration by making traction on the chest walls so as to expand the lungs, the intra- thoracic pressure may fall to —30 mm. of mercury; and in a living rabbit, during a deep natural inspiration, a pressurs of —20 mm. has been seen. The reason why the lungs collapse when the chest is opened is that the pressure is now equal on the pleural and alveolar surfaces, being in both cases that of the atmosphere. There is therefore nothing to oppose the elasticity of the lungs, which tends to con- tract them. So long as the chest is unopened, the pressure on the pleural surface of the lunge is less than that on the alveolar surface 238 RESPIRATION and the elastic tension can only cause them to shrink until it just balances this difference. In intra-uterine life, and in stillborn children who have never breathed, the lungs are completely collapsed (atelectatic), and there is no negative intrathoracic pressure. They are kept in this con- dition by adhesion of the walls of the bronchioles and alveoli. If the lungs have been once inflated, this adhesion ceases to act, and they never completely collapse again. Amount and Variations of the Respiratory or Intrapulmonary Pressure. — As we have already remarked, the pressure in the alveoli and air-passages is less than that of the atmosphere while the in- spiratory movement is going on, greater than that of the atmosphere during the expiratory movement, and equal to that of the atmo- sphere when the chest-walls are at rest. When the external air- passages are closed — e.g., by connecting a manometer with the mouth and pinching the nostrils — the greatest possible variations of pressure are produced. In the deepest inspiration under these conditions a negative pressure of about 75 mm. of mercury (i.e., a pressure less than that of the atmosphere by this amount) has been found, and in deep expiration a somewhat greater positive pressure* (Practical Exercises, p. 304). But with ordinary breathing, the variations of pressure as measured by this method do not exceed 5 to 10 mm. of mercury above or below the pressure of the atmosphere. When the external openings are not obstructed, as, for example, when the lateral pressure is taken in the trachea of an animal by means of a cannula with a side-tube connected with a manometer, still smaller, and doubtless truer, values have been found (2-3 mm. of mercury as the positive expiratory pressure and i mm. as the negative inspiratory pressure in dogs). But since the respiratory passages are abruptly narrowed at the glottis, the variations of pressure must be greater below than above it, and in general they must increase with the distance from that orifice, being greater, for instance, in the alveoli than in the bronchi. The mechanical phenomena of respiration having been described, it might seem logical to consider next the nervous mechanism by which the respiratory movements are controlled ; but the regulation of these movements through the nervous system is in so important a degree a chemical regulation that it cannot be properly understood without some knowledge of the chemical changes in the blood associated with external and internal respiration. We therefore pass to the consideration of — * The maximum negative pressure in deepest inspiration averaged for 49 students -73 mm. (highest observation -137 mm.) of mercury; the maxi- » mum positive pressure in deepest expiration, -f- So mm. (highest observation ' + 140 mm.). THE CHEMISTRY OF EXTERNAL RESPIRATION 239 SECTION III. — THE CHEMISTRY OF EXTERNAL RESPIRATION. Our knowledge of this subject has been entirely acquired in the last 200 years, and chiefly in the last century. Boyle showed by means of the air-pump that animals die in a vacuum, and Bernouilli that fish cannot live in water from which the air has been driven out by boiling. Mayow, of Oxford, seems to a considerable extent to have antici- pated Black, who in 1757 demonstrated the presence of carbonic acid (carbon dioxide) in expired air by the turbidity which it causes in lime-water. A fundamental step was the discovery of oxygen by Priestley in 1771, and his proof that the venous blood could be made crimson, like arterial, by being shaken up with oxygen. Lavoisier discovered the composition of carbonic acid, and applied his discovery to the explanation of the respiratory processes in animals, the heat of which he showed to be generated, like that of a candle, by the union of carbon and oxygen. He made many further important experiments on respiration, publishing some of hi 3 results in 1789, when the French Revolution, in which he was to be one of the most distinguished victims, was breaking out. He made the mistake, however, of supposing that the oxidation of the carbon takes place in the blood as it passes through the lesser circulation. That some carbon dioxide is formed in the lungs there is no reason to doubt, and the quantity may even be considerable. But that they are not the chief seat of oxidation was sufficiently proved as soon as it was known that the blood which comes to them from the right heart is rich in carbon dioxide, while the blood which leaves them through the pulmonary veins is comparatively poor. There are two main lines on which research has gone in trying to solve the chemical problems of respiration: (i) The analysis and comparison of the inspired and expired air, or, in general, the in- vestigation of the gaseous exchange between the blood and tho air in the lungs. (2) The analysis and comparison of the gases of arterial and venous blood, of the other liquids, and of the solid tissues of the body, with a view to the determination of the gaseous exchange between the tissues and the blood. We shall take these up as far as possible in their order. The methods which have been used for comparing the composi- tion of inspired and expired air and estimating the respiratory ex- change are verv various. (i) Breathing into one spirometer and out of another, the inspired and expired air being directed by valves. The contents 'of the spiro- meters are analyzed at the end of the experiment (Speck). In the arrangement of Zuntz and Geppert, instead of the whole of the expired air, a sample is collected for analysis during the entire duration of the experiment, while the total volume expired is measured by a gas-meter. This is a very convenient method for observations on man, especially 24o RESPIRATION in disease, but each experiment can only be carried on at most for fifteen to twenty minutes. (2) A small apparatus, much on the same principle, was used for rabbits by Pfliiger and his pupils. A cannula in the trachea was con- nected with a balanced and self-adjusting spiro meter containing oxygen, and the inspired and expired air separated by potassium hydroxide valves, which absorbed the carbon dioxide. The amount of oxygen used could be read off on the spirometer, and the amount of carbon dioxide produced estimated in the liquid of the valves. (3) Elaborate arrangements, such as Pettenkofer's great respiration apparatus, and the still larger and more efficient modifications of it constructed since his time, in which a man, or even several men, can remain for an indefinite period, working, eating, and sleeping: Air is drawn out of the chamber by an engine, its volume being measured by a gas-meter. But as it would be far too troublesome to analyze the whole of the air, a sample stream of it is constantly drawn off, which also passes through a gas-meter, through drying-tubes containing sulphuric acid, and through tubes filled with baryta water. The baryta solution is titrated to determine the quantity of carbon dioxide; the increase in weight of the drying tubes gives the quantity of aqueous vapour. A similar sample stream of the air before it passes into the chamber is treated exactly in the same way, and from the data thus got the quantity of carbon dioxide and aqueous vapour given off can readily be ascertained. The oxygen can be calculated, as the difference be- tween the final body-weight and the original body-weight plus the weight of the carbon dioxide and water eliminated, but may also be directly estimated by special methods. (4) Haldane and Pembrey have elaborated a gravimetric method, which is very suitable for small animals. It depends upon the absorp- tion of carbon dioxide by soda lime. (See Practical Ex'ercises, p. 30%) In Atwater's so-called respiration calorimeter, which will be referred to again under ' Animal Heat,' and by which, not only the gaseous metab- olism, but the heat production can be measured in man, the carbon dioxide is estimated in the same way. Inspired and Expired Air. — The expired air is at or near the body temperature, and is saturated with watery vapour. In ordinary breathing it contains about 4 per cent, of carbon dioxide, while the inspired air only contains a trace. The expired air contains 16 or 17 per cent, of oxygen, the inspired air about 21 per cent. There are in addition in expired air small quantities of hydrogen and marsh- gas derived from the alimentary canal, either directly from eructa- tion or after absorption into the blood. Sometimes a trace of ammonia can be detected in the air of expiration, but this is due to decomposition of proteins taking place in the mouth, especially in carious teeth, or in the air-passages and lungs in disease of these organs. It has indeed been shown that the lungs are practically impermeable for ammonia. Expired air is entirely free from float- ing matter (dust), which is always present in the inspired air. The volume of the expired air, owing to its higher temperature and ex- cess of watery vapour, is somewhat greater than that of the inspired air, but if it be measured at the temperature and degree of satura- tion of the latter, the volume is somewhat less. Since the oxygen of a given quantity of carbon dioxide would have exactly the same THE CHEMISTRY OF EXTERNAL RESPIRATION 24i volume as the carbon dioxide itself at a given temperature and pressure, it is clear that the deficiency is due to the fact that all the oxygen which is taken up in the lungs is not given off as carbon dioxide. Some of it, going to oxidize hydrogen, reappears as water. Alveolar Air. — The percentage of carbon dioxide in the air of the alveoli is of course greater and that of oxygen less than in the ordinary expired air. For the relatively pure air of the dead-space makes up a substantial fraction of the tidal air. The mean of the oxygen or carbon dioxide percentages in samples taken from the last portions of the air of two deep expirations, one following an ordinary inspiration and the other following an ordinary expiration, corresponds fairly well with the mean percentage in the alveoli during ordinary breathing. This method becomes untrustworthy during muscular work. The average percentage of oxygen may be taken as 14*5, corresponding to a partial pressure (p. 248) of 109 mm. of mercury. The percentage of carbon dioxide in the alveolar air with the body at rest is remarkably constant in one and the same individual at constant atmospheric pressure. It varies in different men from 4-6 to 6*2 (mean 5-5) per cent, of the dry alveolar air. In women and in children of both sexes it is less than in men. From this we conclude that in men the partial pressure of carbon dioxide in the alveoli may be at least one-eighteenth of an atmosphere, or 42 mm. of mercury (Fitzgerald and Haldane). Respiratory Quotient. — The quotient of the volume of oxygen given out as carbon dioxide by the volume of oxygen taken in is the respiratory quotient. It shows what proportion of the oxygen is used to oxidize carbon. It may approach unity on a carbo-hydrate diet which contains enough .ox^gyen to oxidize all its own hydrogen to water. With a diet rich in fat it is least of all; with a diet of lean meat it is intermediate in amount. For ordinary fat contains no more than one-sixth, and proteins not one-half, of the oxygen needed to oxidize their hydrogen (p. 620). In man on a mixed diet the respiratory quotient may be taken as 0-8 or 0-9. So long as the type of respiration is not .changed, the respiratory quotient may remain constant for a wide range of metabolism. In hibernating animals, however, the respiratory quotient may become very small during winter sleep (as low as> 0-25), both the output of carbon dioxide and the consumption of oxygen falling enormously, but the former in general more than the latter. This has been explained on the assumption that oxygen is stored away in winter sleep in the form of incompletely oxidized substances. On the other hand, in dyspnoea accompanying muscular exertion the respiratory quotient has been found as high as 1-2. It must be remembered that even a voluntary increase in the respiratory movements causes an imme- diate temporary increase in the respiratory quotient, owing to the ' washing out ' of carbon dioxide from the blood and tissues. This change has no metabolic significance. Indeed, the determination 16 of the respiratory quotient for short periods has only a limited value, and such observations must be interpreted with great care. In starvation the respiratory quotient diminishes, the production of carbon dioxide falling off at a greater rate than the consumption of oxygen, for the starving organism lives on its own fat and pro- teins, and has only a trifling carbo-hydrate stock to draw upon. In a diabetic patient, fed on a diet of fat and protein alone, the respiratory quotient was only 0-6 to 07, just as in a starving man. Total Respiratory Exchange. — The amount of oxygen absorbed in a man at rest has been determined under certain conditions as about 0-29 gramme per hour, and the discharge of carbon dioxide as about °'33 gramme per hour per kilogramme of body- weight. In an average man weighing 70 kilos the mean production of carbon dioxide is about 800 grammes (400 litres) in twenty-four hours, and the mean consumption of oxygen about 700 grammes (490 litres). But there are very great variations depending upon the state of the body as regards rest or muscular activity and on other circum- stances. In hard work the production of carbon dioxide was found to rise to nearly 1,300 grammes, and in rest to sink to less than 700 grammes, the consumption of oxygen in the same circumstances increasing to nearly 1,100 grammes and diminishing to 600 grammes. In rest, in moderate exertion, and in hard work, the production of carbon dioxide was found to be nearly proportionate to the numbers 2, 3, and 6 respectively. When unaccustomed work is performed, the increase in the carbon dioxide output (and oxygen intake) may be much greater. With training it diminishes. In a case of diabetes the consumption of oxygen was 50 per cent, greater than in a healthy man, corresponding to the higher heat-equivalent of the food of the diabetic patient. Ventilation. — Taking 400 litres per twenty-four hours, or 17 litres per hour, as the mean production of carbon dioxide by an average male adult at rest or doing only light work, we can calculate the quantity of fresh air which must be supplied to a room in order to keep it properly ventilated. It has been found that when the carbon dioxide given off in respiration amounts to no more than 2 parts in 10,000 in the air of an ordinary room, the air remains sweet. When the carbon dioxide given off reaches 4 parts in 10,000, the room feels distinctly, and at 6 in 10,000 disagree- ably, close, while at 9 parts in 10,000 it is oppressive and almost in- tolerable. This is not due to the carbon dioxide as such, for pure carbon dioxide added alone in similar proportions to the air of a room has not the same bad effect, and the amount of this gas is only taken as an index of the extent to which the air has been vitiated by some other products or processes connected with the occupation of the room. Very often the mere rise of temperature in a crowded and ill- ventilated space is sufficient to induce disagreeable symptoms, especially as it is inevitably associated with an ijncrease in the humidity of the air, which reduces the capacity of the body to cool itself by increasing the secretion of sweat. Thus it has been found that persons in a respiratory chamber feel quite comfortable with only moderate ventila- tion when the carbon dioxide has risen to i per cent., if care is taken THE CHEMISTRY OF EXTERNAL RESPIRATION 243 that the temperature and the proportion of watery vapour do not rise too high. In addition, however, it has been supposed by some that a volatile poison exhaled from the lungs is peculiarly responsible for the evil effects. Certain observers, indeed, alleged that the condensed vapour of the breath, when injected into rabbits, produced fatal symp- toms. But this has been shown to be erroneous; and the most careful experiments have failed to detect in the air expired by healthy persons any trace of such a poison. It has therefore been suggested that the odour and some of the other ill-effects of a close room are due to sub- stances given off in the sweat and the sebum, and allowed by persons of uncleanly habits to accumulate on the skin, and also to the products of slow putrefactive processes constantly going on, under favourable conditions, on the walls, floor, or furniture, but only becoming per- ceptible to the sense of smell when ventilation is insufficient. In a small, newly-painted chamber, presumably free from such impurities, it was not until the carbon dioxide reached 3 to 4 per cent., an immensely greater proportion than occurs even in very badly ventilated rooms, that marked discomfort, with dyspnoea, began to be felt. No close odour could be detected. Nevertheless, experience has shown that it is a good working rule for ventilation to take the limit of permissible respiratory impurity at 2 parts of carbon dioxide per 10,000; and the 17 litres of carbon dioxide given off in the hour will require 85,000 litres (or 3,000 cubic feet) of air to dilute it to this extent. This is the average quantity required for the male adult per hour. For men engaged in active labour, as in factories or mines, twice this amount may not be too much. For women and children less is required than for men. If a room smells close, it needs ventilation, whatever be the proportion of carbon dioxide in the air. It must be remembered that in permanently occupied rooms mere increase in the size will not compensate for incomplete renewal of the air, although it may be easier to ventilate a large room than a small one without causing draughts and other inconveniences. But as few apartments are occupied during the whole twenty-four hours, a large room which can be thoroughly ventilated in the absence of its inmates has a distinct advantage over a small one in its great initial stock of fresh air. The cubic space per head in an ordinary dwelling- house should be not less than 28 cubic metres or 1,000 cubic feet. The quantity of carbon dioxide given off (and of oxygen consumed) is not only affected by muscular work, but also by everything which influences the general metabolism. In males it is greater on the average than in females (in the latter there is a temporary increase during pregnancy), but for the same body- weight and under similar external conditions there is no difference between the sexes. The gaseous exchange is greater in proportion to the body-weight in the child than in the adult. This depends largely on the fact that, other things being equal, the metabolism is relatively to the body- weight more active in a small than in a large organism, since the surface (and therefore the heat loss) is relatively greater in the former. But it has been shown that even in proportion to the surface the metabolism is greater in youth than in adult life, and greater in the vigorous adult than in the old man. So that the age of the organism has an influence apart from the extent of surface. The taking of food increases the gaseous exchange, partly from the 244 RESPIRATION increased mechanical and chemical work performed by the ali- mentary canal and the digestive glands. But that this is not the sole cause of the increase is shown by the fact that it varies with different kinds of food to a greater extent than can be explained by differences in the ease with which they are digested. For in- stance, maize produces a much greater increase than oats when given in equal amount, and a protein diet a greater increase than a diet of carbo-hydrate or fat. Sleep diminishes the production of carbon dioxide partly because the muscles are at rest, but also to some extent because the external stimuli that in waking life excite the nerves of special sense are absent or ineffective. Even a bright light is said to cause an increase in the atmount of carbon dioxide produced and of oxygen consumed ; but probably only by increasing muscular movements, including the movements of respiration. The external temperature also has an influence. In poikilothermal animals (such as the frog), the temperature of which varies with that of the surrounding medium, the production of carbon dioxide, on the whole, diminishes as the external temperature falls, and increases as it rises. In homoiothermal animals, that is, animals with constant blood temperature, external cold increases the pro- duction of carbon dioxide and the consumption of oxygen. But if the connection of the nervous system with the striated muscles has been cut out by curara, the warm-blooded animal behaves like the cold-blooded (Pfliiger and his pupils in guinea-pig and rabbit). These interesting facts will be returned to under ' Animal Heat.' Cold-blooded animals produce far less carbon dioxide, and con- sume far less oxygen, per kilo of body-weight than warm-blooded. The following table shows the relation between the body-weight and the excretion of carbon dioxide in man : Age. Weight in Kilos. COj excreted per Kilo per Hour. r.58 84-6 0-41 gramme 44 76-5 0-48 Male 35 6o5 82 0-51 0-49 16 57'7 o-59 I 9'6 22 0-92 C66 66-9 o-39 Female j 3° 53-9 55'7 o-54 o-45 Uo 23 0-83 The next table illustrates the difference in the intensity of metab- olism in different kinds of animals, a difference, however, largely dependent upon relative size: THE CHEMISTRY OF EXTERNAL RESPIRATION 245 Oxygen absorbed per Kilo per Hour. Carbon Dioxide given off per Kilo per Hour. Respiratory Quotient Animal. CO2 Oo (in COa) In Grms. In C.C. In Grms. InCC. 02 °r 02 Greenfinch - I3-000 9091 13-590 6909 0-76 Hen - I-058 740 I-327 675 O-gi Dog - I-303 911 I-325 674 o-74 Rabbit 0-987 690 1-244 632 0-91 Sheep - 0-490 343 0-671 341 0-99 Boar - 0-3QI 273 0-443 225 0-82 Frog - O'IO5 73'4 0-113 577 0-78 Crayfish 0-054 38 0-064 32-7 0-86 Forced respiration, although it will temporarily increase the quantity of carbon dioxide given off by the lungs, and thus raise for a short time the respiratory quotient, does not sensibly affect the production; it is only the store of already formed carbon dioxide in the body which is drawn upon. The amount of oxygen taken up is little altered by changes in the movements of respiration. Within wide limits the oxygen consumption of the organism is in- dependent of the supply of oxygen offered to it. How it is that the depth of the respiration may affect the rate at which carbon dioxide is eliminated, we can only understand when we have examined the process by which the gaseous interchange between the blood and the air of the alveoli is accomplished; and before doing this it is necessary to consider the condition of the oxygen and carbon dioxide in the blood. SECTION IV. — THE GASES OF THE BLOOD. Physical Introduction. — Matter may be assumed to be made up of molecules beyond which it cannot be divided without altering its essen- tial character. A molecule may consist of two or more particles of matter (atoms) bound to each other by chemical links. The kinetic theory of matter supposes the molecules of a substance to be in constant motion, frequently colliding with each other, and thus having the direc- tion of their motion changed. In a gas the mean free path, that is, the average distance which a molecule travels without striking another, is comparatively long, and far more time is passed by any molecule without an encounter than is taken up with collisions. Although the average velocity of the mole- cules is very great, these collisions will produce all sorts of differences in the actual velocity of different molecules at any given time. Some will be moving at a greater, some at a slower rate, than the average; while some may be for a moment at rest. If the gas is in a closed vessel, the molecules will be constantly striking its sides and rebounding from them. If a very small opening is made in the vessel, some mole- cules will occasionally hit on the opening and escape altogether. If the opening is made larger, or the experiment continued for a longer time 24« RESPIRA TION with the small opening, all the molecules will in course of time have passed out of the vessel into the air, while molecules of the oxygen, nitrogen, and argon of the air will have passed in. In a gas, then, not enclosed by impenetrable boundaries, there is no restriction on the path which a molecule may take, no tendency for it to keep within any limits. When two chemically indifferent gases are placed in contact with each other, diffusion will go on till they are uniformly mixed. The diffusion of gases may be illustrated thus. Suppose we have a perfectly level and in every way uniform field divided into two equal parts by a visible but intangible line, the well-known whitewash line, for instance. On one side of the line place 500 blind men in green, and on the other 500 blind men in red. At a given signal let them begin to move about in the field. Some of the men in green will pass over the line to the ' red ' side; some of the men in red will wander to the ' green ' side. Some of the men may pass over the line and again come back to the side they started from. But, upon the whole, after a given interval has elapsed, as many green coats will be seen on the red side as red coats on the green. And if the interval is long enough there will be at length about 250 men in red and 250 in green on each side of the boundary- line. When this state of equilibrium has once been reached, it will henceforth be maintained, for, upon the whole, as many red uniforms will pass across the line in one direction, as will recross it in the other. In a liquid it is very different; the molecule has no free path. In the depth of the liquid no molecule ever gets out of the reach of other molecules, although after an encounter there is no tendency to return on the old path rather than to choose any other; so that any molecule may wander through the whole liquid. Although the average velocity of the molecules is much less in the liquid state than it would be for the same substance in the state of gas or vapour (gas in presence of its liquid), some of them may have velocities much above the average. If any of these happen to be moving near the surface and towards it, they may overcome the attraction of the neighbouring molecules and escape as vapour. But if in their further wanderings they strike the liquid again, they may again become bound down as liquid molecules. And so a constant interchange may take place between a liquid and its vapour, or between a liquid and any other gas, until the state of equi- librium is reached, in which on the average as many molecules leave the liquid to become vapour as are restored by the vapour to the liquid, or as many molecules of the dissolved gas escape from solution as enter into it. For the sake of a simple illustration, let us take the case of a shallow vessel of water originally gas-free, standing exposed to the air. It will be found after a time that the water contains the atmospheric gases in certain proportions — in round numbers, about Tfoj of its volume of oxygen and £G of its volume of nitrogen (measured at 760 mm. mercury and o° C.). Now, let a similar vessel of gas-free water be placed in a large airtight box filled with air at atmospheric pressure, and let the oxygen be all absorbed before the water is exposed to the atmosphere of the box. The latter now consists practically only of the nitrogen of the air, and its pressure will be only about four-fifths that of the external atmo- sphere. Nevertheless, the quantity of nitrogen absorbed by the water will be exactly the same as was absorbed from the air. If the box was completely exhausted, and then a quantity of oxygen, equal to that in it at first, introduced before the water was exposed to it, the pressure would be found to be only about one-fifth that of the external atmo- sphere; but the quantity of oxygen taken up by the water would be exactly equal to that taken up in the first experiment. THE GASES OF THE PJ.OOD *\\ Two well-loiown physical laws are illustrated by our supposed ex- periments: (i) In a mixture of gases which do not act chemically on each other the pressure exerted, by each gas (called the partial pressure of the gas) is the same as it ivould exert if the others were absent. (2) The quan- tity (mass) of a gas absorbed by a liquid which does not act chemically upon it is proportional to the partial pressure of the gas. It also depends upon the nature of the gas and of the liquid, and on the temperature, increase of temperature in general diminishing the quantity of gas absorbed. It is to be noted that when the volume of the absorbed gas is measured at a pressure equal to the partial pressure under which ic was absorbed, the same volume of gas is taken up at every pressure. The volume of a gas (reduced to o° C. and 760 mm. pressure) physi- cally absorbed or dissolved in i c.c. of a liquid exposed to the gas at 760 mm. pressure is called the absorption coefficient of the gas in that liquid. The following table from Bohr shows the absorption coefficients of the three gases of physiological interest — oxygen, nitrogen, and carbon dioxide in water, blood-plasma, whole blood and blood-corpuscles at the body temperature (38° C.) : Oxygen. Nitrogen. Carbon Dioxide. Water 0-0237 O-OI22 o'555 Blood-plasma 0'023 O-OI2 0-541 Blood - --- O-O22 O-OII 0-511 Blood-cells - 0-019 o-oio 0-450 Suppose, now, that a vessel of water, saturated with oxygen and nitrogen for the partial pressures under which these gases exist in the air, is placed in a box filled with pure nitrogen at full atmospheric pres- sure. As we have seen, there is a constant interchange going on between a liquid which contains gas in solution and the atmosphere to which it is exposed. Oxygen and nitrogen molecules will therefore continue to leave the water; but if the box is large, few oxygen molecules will find their way back to the water, and ultimately little oxygen will remain in it. In other words, the quantity of oxygen absorbed by the water will become again proportional to the partial pressure of oxygen, which is not now much above zero. On the other hand, molecules of nitrogen will at first enter the water in larger number than they escape from it, for the pressure of the nitrogen is now that of the external atmosphere, of which its partial pressure was formerly only four-fifths. In unit volume of the gas above the water there will be 5 molecules of nitrogen for every 4 molecules in the same volume of atmospheric air. There- fore, on the average 5 nitrogen molecules will in a given time get en- tangled by liquid molecules for every 4 which came within their sphere of attraction before. On the whole, then, the water will lose oxygen and gain nitrogen, while the atmosphere of the airtight box will gain oxygen and lose nitrogen. In the case of water, in which oxygen and nitrogen are absorbed solely in solution, the partial pressures of these gases under which the water was originally saturated could, of course, be easily" calculated from the amount dissolved and the ccefficient of absorption. But supposing that these partial pressures were unknown, it is evident that by exposing it to an atmosphere of known composition, and afterwards determining the changes produced m the composition of that atrnp' 248 RESPIRATION feFQ •G sphere by loss to, or gain from, the gases of the water, we could find out something about the original partial pressures. If, for example, the quantity of oxygen in the atmosphere or the chamber was increased, we could conclude that the partial pressure of oxygen under which the wattr had been saturated was greater than that in the chamber at the beginning of the experiment. And if we found that with a certain partial pressure of oxygen in the atmo- sphere of the chamber there was neither gain nor loss of this gas, we might be sure that the partial pressure (the temperature being sup- posed not to vary), was the same when the water was saturated. We shall see later on how this principle has been applied to deter- mine the partial pressure of oxygen or carbon dioxide which just suffices to prevent blood, or any other of the liquids of the body, from losing or gaining these gases when they are not merely dissolved, but also combined in the form of dissociable compounds. This pressure is evidently equal to that exerted by the gases of the liquid at its surface, which is sometimes called their ' tension ' ; for if it were greater, gas would, upon the whole, pass into the blood ; and if it were less, gas would escape from the blood. Thus, the tension of a gas in solution in a liquid is equal to the partial pressure of that gas in an atmosphere to which the liquid is ex- posed, which is just sufficient to prevent gain or loss of the gas by the liquid (p. 258). The following imaginary experiment may further illustrate the meaning of the term ' ten- sion ' of a gas in a liquid in this connection. Suppose a cylinder filled with a liquid con- taining a gas in solution, and closed above by a piston moving airtight and without friction, in contact with the surface of the liquid (Fig. 117). Let the weight of the piston be balanced by a counterpoise. The pressure at the sur- face of the liquid is evidently that of the atmosphere. Now; let the whole be put into the receiver of an air-pump, and the air gradually exhausted. Let exhaustion proceed until gas begins to escape from the liquid and lies in a thin layer between its surface and the piston, the quantity of gas which has become free being very small in proportion to that still in solution. At this point the piston is acted upon by two forces which balance each other, the pressure of the air hi the receiver acting downwards, and the pressure of the gas escaping from the liquid acting upwards. If the pressure in the receiver is now slightly increased, the gas is again absorbed. The pressure at which this just happens, and against which the piston is still supported by the impacts of gaseous molecules flying out of the liquid while no pressure is as yet Fig. 117. — Imaginary Ex- periment to illustrate ' Tension ' of a Gas in a Liquid. P, frictionless piston; L, liquid in cy- linder; G, gas beginning to escape from liquid. P is exactly counter- poised. In addition to the manner described in the text, the experiment may be supposed to be performed thus: Let the weight, W, be deter- mined which, when the receiver is completely exhausted, suffices just to keep the pir :on in contact with the liquid. The pressure of the gas is then just counter- balanced by W; and if S is the area of the cross- section of the piston, the pressure of the gas per W unit of area is ^-. Or, if o the piston is hollow, and mercury is poured into it so as just to keep it in contact with the liquid, the height of the column of mercury required is also equal to the pressur e or tension of the gas. THE GASES OF THE BLOOD 249 exerted directly between the liquid and the piston, is obviously equal to the pressure or tension of the gas in the liquid. From the above principles it follows that a gas held in solution may be extracted by exposure to an atmosphere in which the partial pressure of the gas is made as small as possible. Thus, oxygen can be obtained from liquids in which it is simply dissolved by putting them in an atmosphere of hydrogen or nitrogen, in which the partial pressure of oxygen is zero, or in the vacuum of an air-pump, in which it is extremely small. Heat also aids the expulsion of dissolved gases. Some gases held in weak chemical union, like the loosely-combined oxygen of oxyhaemoglobin, can be obtained by dissociation of their compounds Fig. 1 1 8.- — Scheme of Gas-Pump. A, the blood bulb; B, the froth chamber; C, the drying tube; D, fixed mercury bulb ; E, movable mercury bulb connected by a flexible tube with D; F, eudiometer; G, a narrow delivery tube; i, 2, 3, 4, taps, 4 being a three-way tap. A is filled with blood by connecting the tap i by means of a tube with a bloodvessel. Taps i and 2 are then closed. The rest of the apparatus from B to D is now exhausted by raising E, with tap 4 turned so as to- place D only in communication with G, till the mercury fills D. Tap 4 is now turned so as to connect C with D, and cut off G from D, and E is lowered. The mercury passes out of D, and air passes into it from B and C. Tap 4 is again turned so as to cut off C from D and connect G and D. E is raised and the mercury passes into D and forces the air out through G, the end of which has not hitherto been placed under F. This alternate raising and lowering of E is continued till a man- ometer connected between C and 4 indicates that the pressure has been sufficiently reduced. The tap 2 is now opened ; the gases of the blood bubble up into the froth chamber, pass through the drying-tube C, which is filled with pumice-stone and sulphuric acid, and enter D. The end of G is placed under the eudiometer F, and by raising E, with tap 4 turned so as to cut off C, the gases are forced out through G and collected in F. The movements required for exhaustion can be repeated several times till no more gas comes off. The escape of gas from the blood is facilitated by immersing the bulb A in water at 40° to 50° C. when the partial pressure is reduced. More stable combinations may require to be broken up by chemical agents — carbonates, for instance, by acids. Extraction of the Blood-Gases. — This is best accomplished by ex- posing blood to a nearly perfect vacuum. The gas-pumps which have been most largely used in blood analysis are constructed on the principle of the Torricellian vacuum. A diagram of a simple form of Pfliiger's gas-pump is given in Fig. 118. The gases obtained are ultimately dried and collected in a eudiometer, which is a graduated glass tube with its mouth dipping into mercury. The carbon dioxide is estimated by introducing a little potassium hydroxide to absorb it. The diminution in the volume of the gas contained in the eudiometer gives the volume of the carbon dioxide. The oxygen may be estimated by putting into the eudiometer more than enough hydrogen to unite with all the oxygen so as to form water, and then, after reading off the volume, exploding the mixture by means of an electric spark passed through two platinum wires fused into the glass. One-third of the diminution of volume represents the quantity of oxygen present. It can also be estimated 250 RESPIRATION by absorption with a solution of pj-rogallic acid and potassium hydrox- ide, or an alkaline solution of sodium hydrosulphite, which is more cleanly. The remainder of the original mixture of blood-gases, after deduction of the carbon dioxide and oxygen, is put down as nitrogen (with, no doubt, a small proportion of argon). For the sake of easy comparison, the observed volume of gas is always stated in terms of its equivalent at a standard pressure and temperature (760 mm., or some- times on the Continent i metre of mercury, and o° C.). It is also possible in various ways to estimate the amount of oxygen in blood without the use of the pump. Thus, since a definite volume of oxygen (i -338 c.c. at o° C. and 760 mm. pressure) combines with it gramme of haemoglobin, we can calculate the total volume of oxygen present if we know how much of the blood-pigment is in the form of oxyhaemoglobin ; and this can be determined by means of the spectro- photometer. Or potassium ferricyanide may be added to the blood. This expels the oxygen from its combination with the haemoglobin, which then unites with an exactly equal amount of oxygen obtained from the ferricyanide to form methaemoglobin (Haldane) (p. 75). In the hands of Barcroft and his pupils this method has been highly developed, so that accurate results can be obtained with small quantities of blood (i c.c., and with the smaller apparatus even o-i c.c.). Bar- croft's differential apparatus consists essentially of two small bottles, as nearly alike as possible, connected by a manometer filled with oil (of cloves). The amount of oxygen liberated in one of the bottles by potassium ferricyanide from a measured amount of blood can be esti- mated from the displacement of the liquid in the manometer. The function of the second bottle is to automatically eliminate effects due to changes in the temperature of the bath in which the apparatus is immersed, etc., since both bottles are affected alike. The Quantity of the Blood-Gases.— In arterial and in venous blood oxygen, carbon dioxide, nitrogen, and argon are constantly found. Both the oxygen and the carbon dioxide vary considerably in amount in the arterial blood, even of individuals of the same animal group, and, of course, much more in the venous blood, as might naturally be expected, since even to the eye it varies greatly accord- ing to the vein it is obtained from, the rapidity of the circulation, and the activity of the tissues which it has just left. In one observation on blood obtained directly from a human artery, 2i«6c.c. of oxygen, 40*3 c.c. of carbon dioxide, and 1-6 c.c. of nitrogen were found in 100 c.c. of blood. The quantity of oxygen taken up outside of the body by specimens of human blood drawn from six normal persons, when shaken up with atmospheric air, varied from 17-6 c.c. to 22-5 c.c. per 100 c.c. of blood, the variations depending mainly on the haemoglobin content. The arterial blood as it actually left the lungs of those persons must have contained somewhat less oxygen (about i c.c. less per looc.c. of blood), since the partial pressure of oxygen in the alveolar air is decidedly below that in atmospheric air. In dogs the amount of carbon dioxide in arterial blood has been found to vary from 35 to 45 c.c. per 100 c.c. of blood, the differences being due to variations in the extent of the pulmonary ventilation and to other factors. THE GASES OF THE BLOOD 251 In a series of observations on the venous blood of dogs the oxygen content ranged from 5-5 to 16-6 c.c. (average 11-9 c.c.), and the carbon dioxide content from 38-8 c.c. to 47-5 (average 44-3 c.c.) per 100 c.c. of blood (Schoffer) . It will be sufficiently accurate to assume that on the average, Volumes of 67 C02. N£ 100 volumes of arterial blood yield 20 40 1-2 ,, ,, mixed venous blood (from right heart) yield - 10-12 45-50 1-2 (reduced to o° C. and 760 mm. of mercury). Average venous blood contains 7 or 8 per cent, by volume less oxygen, and 7 or 8 per cent, more carbon dioxide, than arterial blood. Thus, in the lungs the blood gains about twice as many volumes of oxygen per cent, as the air loses, and the air gains about half as many volumes of carbon dioxide per cent, as the blood loses. It is easy to see that this must be so, for the volume of air inspired in a given time is about twice as great as that of the blood which passes through the pulmonary circulation (pp. 223, 236). Even arterial blood is not quite saturated with oxygen; it can still take up a variable small amount. The percentage saturation with oxygen of the arterial blood of a normal woman from whom blood was being transfused into a patient was directly determined. The blood proved to be 94 per cent, saturated — i.e., it could stili have taken up about one-sixteenth of the quantity contained in it. Nor is venous blood nearly saturated with carbon dioxide; when shaken with the gas it can take up about 150 volumes per cent. The total oxygen capacity of the blood in any individual can be determined, if the volume of the blood is known (p. 56), by estimating the amount of oxygen needed to saturate a sample of the blood. This is most conveniently done by the ferricyanide method with Barcroft's apparatus. Suppose, for example, that i c.c. of blood, after being thoroughly shaken up with air or oxygen, gives off 0-2 c.c. of oxygen when acted upon by ferricyanide, and that the total volume of the blood has been determined to be 5 litres, then the total oxygen capacity of the blood will be 1,000 c.c. This quantity, it is clear, is a measure of the power of the blood to trans- port oxygen. The oxygen capacity of a sample of blood is propor- tional to the amount of hemoglobin in it, so that the estimation of the percentage amount of haemoglobin by a properly standardized haemoglobinometer is really an estimation of the oxgyen capacity of the quantity of blood used for the determination. The total oxygen capacity of the body cannot, of course, be derived from such an observation, any more than the total quantity of haemoglobin, unless the amount of blood in the body is known. When the gases are not removed from blood immediately after 252 RESPIRATION it is drawn, it yields more carbon dioxide and less oxygen than if it is evacuated at once (Pfliiger). From this it is concluded that oxidation goes on in the blood for some time after it is shed. The oxidizable substances are, however, confined to the corpuscles, which suggests that ordinary metabolism simply continues for some time in the formed elements of the shed blood, and that the disappearance of oxygen is not due to the oxidation of substances which have reached the blood from the tissues. It is an interesting fact that the rate of oxygen consumption of nucleated (bird's) erv- throcytes is much greater than that of non-nucleated mammalian corpuscles. The young non-nucleated erythrocytes, which in experi- mental anaemia in mammals (e.g., after haemorrhage) find their way in large numbers into the circulation, have a relatively intense metabolism, and therefore consume a relatively large amount of oxygen. The Distribution and Condition of the Oxygen in the Blood. — The oxygen is nearly all contained in the corpuscles. A little oxygen can be pumped out of serum (0-2 or 0-3 per cent, by volume), but this follows the Henry-Dalton law of pressures — that is, it comes off in proportion to the reduction of the partial pressure of the oxygen in the pump, and is simply in solution. When blood at body temperature is shaken up with air at the ordinary pressure, corresponding to a partial pressure of oxygen of a little over one-fifth of an atmosphere (in round numbers 160 mm. of mercury), the blood-pigment becomes saturated with oxygen or nearly so. When the blood is now pumped out, very little oxygen comes off till the pressure has been reduced to about half an atmo- sphere, corresponding to a pressure of oxygen of about 80 mm. At about 70 mm. partial pressure the dissociation is somewhat greater. At a third to a quarter of an atmosphere (50 to 40 mm.; the amount of oxygen liberated is markedly increased, and the dis sociation becomes more and more rapid as the pressure falls to- wards zero. This behaviour shows that the oxygen is not simply absorbed, but is united, as a dissociable compound, to some con- stituent of the blood. The same thing is, of course, seen when defibrinated blood is saturated at body temperature with oxygen at different pressures. As the partial pressure of the gas is increased from zero the first increments of pressure correspond to a much greater absorption of oxygen than further equal increments. Thus as is seen in Fig. 120, with an oxygen pressure of 10 mm. 100 c.c. of blood took up 6 c.c. of oxygen, or 30 per cent, of the amount required to saturate it. When the pressure of oxygen was 30 mm. over 16 c.c. of oxygen was absorbed, the blood being 80 per cent, saturated. A further increase of the oxygen pressure to 40 mm. increased the quantity of the gas taken up by only 2 c.c. (to 90 per cent, saturation). Ttu next increment of 10 mm. in the oxygen THE GASES OF THE BLOOD 253 385 7 pressure only produced an additional absorption of I c.c., and above this increasing the pressure had very little effect. We may suppose that at the ordinary temperature and pressure some oxygen is continually escaping from the bonds by which it is tied to the haemoglobin; but, on the whole, an equal number of free mole- cules of oxygen, coming within the range of the haemoglobin molecules, are entangled by them, and thus equilibrium is kept up. If now the atmospheric pressure, and therefore the partial pressure of oxygen is reduced, the tendency of the oxygen to break ne»-ce"tage of OxvQC'> >s>. off from the haemoglobin will be unchanged, and as many molecules on the whole will escape as before ; but even after a consider- able reduction of pressure the haemoglobin, such is its avidity for oxygen, will still be able to seize as much oxygen as it loses. The more, however, the partial pressure of the oxygen is diminished — that is to say, the fewer oxygen molecules there are in a given space above the haemoglobin— the smaller will be the chance of the loss being made up by accidental cap- tures. At a certain pressure the escapes will become conspicuously more numer- ous than the captures ; and the gas-pump will give evi- dence of this. The higher the temperature of the haemoglobin is, the greater will be the average velocity of the molecules, and the Fig. 119. — Curves of Dissociation of Oxyhaemo- globin freed from salts by dialysis (after Bar- croft). Along the horizontal axis are plotted the partial pressures (numbers below the curve) of oxygen in air, to which a solution of haemo- globin was exposed. The corresponding per- centages of oxygen are given above the curve. Along the vertical axis is plotted the percentage saturation of the haemoglobin with oxygen. greater the chance of escape of molecules of oxygen. It is easily proved that the substance in the corpuscles which unites with oxygen is the blood-pigment. Although a solution of oxyhaemoglobin crystals behaves towards oxygen somewhat differ- ently from blood containing the same proportion of the native pig- ment, the maximum amount of oxygen taken up is the same for each. Much labour has been spent in determining curves which express the relation between the partial pressure of oxygen to which blood or a haemoglobin solution is exposed, and the proportion to, which the blood pigment becomes saturated with oxygen under each pressure. The differences in the results of the various in- vestigators who have worked out the curves of dissociation for haemoglobin and for blood (Figs. 119, 120, 121) have been largely cleared up by the researches of Barcroft and his co-workers. One 254 RESPIRATION factor which was overlooked in the earlier observations is the influence of salts. The form of curve (a rectangular hyperbola, Fig. 119) which Hiifner put forward by an unjustifiable generalization, as the curve of dis- sociation of oxyhaemoglobin is only found when the haemoglobin solu- tion is thoroughly freed from salts, as by prolonged dialysis. When this condition is fulfilled it can be shown that the curve is always a rectangular hyperbola, but a different one for each temperature at which the observations are made. For haemoglobin solutions not freed from salts or for blood or dilutions Of blood the curve is of a different order altogether, a curve with a double contour (S-shaped). (Figs. 120, 121). If to a solution of haemoglobin freed from salts and yielding a rectangular hyperbola as its dissociation curve, salts be added in the quantities and of the kind known to exist in dog's blood, the curve of no 7 is &a 16 ec 14 cc 1 2 co IOC& see. 6 C.C 4CC. 2 C.C. 03C.C, 0 10 to 30 40 10 M 70 80 00 100 110 120 130 140 ISO Fig. 120. — Curves of Dissociation of Oxygen for Horse's Blood (B) and Dog's Hemo- globin solution (H) at 38° C. (Bohr). The figures along the base-line are the partial pressures of oxygen to which the bLood and haemoglobin solution were exposed. Those along the vertical axis on the left are the percentage saturations with oxygen. The figures along the vertical at the right give the actual number of c.c. of oxygen chemically combined by 100 c.c. of the blood for each pressure of oxygen. The interrupted line P indicates the amount of oxygen dissolved in the plasma of the blood at each partial pressure on the assumption that the plasma is two-thirds of the volume of the blood. Thus, at 150 mm. oxygen pressure the plasma of 100 c.c. of blood took up 0-3 c.c. oxygen. dissociation given by dog's blood is obtained. The addition of the salts appropriate to human blood in the proper amounts causes the hyperbolic curve of the pure haemoglobin solution to change into an S-shaped curve, such as is given by human blood, and so on. The foundation has thus been removed from the theory that different animals possess haemoglobins differing in their capacity to take up oxygen. The salts are supposed to alter the dissociation curve, by changing the degree of aggregation of the haemoglobin molecules- — i.e., by causing them to adhere to each other to a greater or less extent according to the quantity and kind of salts present. Another factor which greatly influences the binding power of haemoglobin for oxygen, and therefore the dissociation curve, is the reaction of the blood (p. 24). When the hydrogen-ion concen- TtiE CASES OF THE BLOOD 255 tration is increased, as by addition of carbon dioxide (Fig. 121) or lactic acid, the effect is to increase the dissociation tension of o. \\-lKemoglobin, or what is a different way of expressing the same thing, to diminish the amount of ox3 (42-5) CO (^ [Wolffberg - — (26) 18-37 C J^ H Nussbaum - — (27) 24-33 f« Loewy and Schrotter ~\ ~ Haldane and Smith J man (.200 + — (37'7) 34-59 (45) It is chiefly the enormous differences in the recorded oxygen tensions of the arterial blood which excite surprise. To some ex- tent, indeed, these also may depend upon differences in the partial pressure of the oxygen in the alveoli, and it has been shown experi- mentally (by the aerotonometer) that with increasing oxygen tension of the inspired air the oxygen tension of the arterial blood increases (Fredericq). Still, the differences which can possibly have existed in the partial pressure of the oxygen in the alveoli in the various series of observations can only to a small extent account for the differences in the results. The main reason for the great range of values lies unquestionably in the different experimental procedures * The numbers in brackets are averages. 262 RESPIRA TION by which the)' were obtained. There is no doubt that in the earlier observations with the aerotonometer (Strassburg) the oxygen of the blood could not have come into equilibrium with the mixture in the gas space, in which the oxygen pressure was at the beginning much lower than that in the blood; the results are therefore too low. The same is true for the oxygen tension of the venous blood, but as this is in any case considerably smaller than that of the arterial blood, the proportional error is not so great. The later experiments (of Herter), given in the second line of the table, yield much higher values, owing to improved technique, but the findings are still to be regarded as minimal and not average results. At the other end of the scale stand the results of Haldane and Smith, who found in man an oxygen tension in the arterial blood of over 200 mm. of mercury equal to more than 26 per cent, of an atmosphere. This exceeds the partial pressure of oxygen in the external air, and is about twice as great as that of the air of the alveoli. In the bird they found an oxygen tensien of between 300 and 400 mm., equal to 45 per cent, of an atmosphere. The chief interest of this discussion of the blood-gas tensions lies in their fundamental importance in the problem of the gaseous exchange in the lungs, on the one hand, and between the blood and tissues on the other. In the presence of such results Haldane and Smith necessarily adopted a secretion theory of gaseous exchange in the lungs. For it was manifestly impossible for oxygen to diffuse from the alveoli into the blood against the slope of pressure. Their findings, however, differed so vastly from those of all observers who had used tonometric methods, and it seemed so difficult to assign an adequate physiological value to the slight increase in the percentage saturation of the haemoglobin with oxygen (see Figs. 119-121), which could be brought about by even a great excess of oxygen tension above that of atmospheric air, that there was a general disposition to distrust the accuracy of their method. At the same time the ordinary tono- metric technique left so much to be desired that there seemed little hope of bringing the matter to a decisive test. The introduction a few years ago by Krogh, of better aerotonometric methods, and the remarkable series of researches carried out by their aid, have changed the whole aspect of the question. He showed that in the arterial blood of rabbits the oxygen tension was in all cases somewhat (usually 2 to 3 per cent, of the atmospheric pressure, or 15 to 23 mm. of mercury) lower than the oxygen tension in air from the bifurcation of the trachea. In animals it is not possible to obtain the actual alveolar air. These results agree very well with those of Fredericq, who found in dogs oxygen tensions of 13 to 14 per cent, of an atmosphere — i.e., something like 100 mm. of mercury, when the animals breathed atmospheric air. Krogh proved that the amount of oxygen lost by self-reduction of the blood (see p. 261) in his aerotonometer was negligible, even in rabbits, although it is known that rabbit's blood uses up more oxygen than that of higher mammals. He also pointed our some sources of error in the Haldane-Smith method. About the same time Haldane, after a careful re-examination of the question, came to the con- clusion that his previous results were much too high; and that THE GASES OF THE BLOOD 263 during rest and when normal air is breathed the oxygen tension in arterial blood never exceeds the tension in the alveolar air. He still believes, and brings forward evidence for his belief, that even the best aerotonometric methods yield with arterial blood oxygen pressures lower than the true oxygen pressure in the blood leaving trie pulmonary capillaries. But he no longer doubts that a sufficient slope of pressure exists between the alveoli and the blood to explain the absorption of oxygen under ordinary conditions. To all intents and purposes, then, we may look upon the contro- versy as closed, and with the happy result, too rare in physiological disputes, that a practically unanimous conclusion has been arrived at. While the oxygen tension in arterial blood may fall only slightly below that in alveolar air, the difference perhaps being even less than that found by Krogh, the slope of pressure is always, under ordinary conditions at least, from the alveoli to the blood. Mechanism of the Gaseous Exchange in the Lungs. — Granting that this is so, it must still be asked whether the diffusion of oxygen from the lungs into the blood can take place rapidly enough to account for the quantities actually absorbed. Calculations made on the basis of such anatomical and physical data as are available (total surface of the lungs, thickness of the membrane which separates the air of the alveoli and the blood in the capillaries, rate of ' invasion,' or entrance of the gases into water; velocity of diffusion of oxygen and carbon dioxide), indicate that even with differ- ences of oxygen tension between the blood and the alveolar air, which would lie within the limits of error of our present methods of measure- ment, enough oxygen could diffuse across the pulmonary membrane to cover the whole normal intake. For example, it has been shown from determinations of the ' invasion- coefficient ' of oxygen that 300 c.c. of oxygen, the amount ordinarily absorbed in a minute, in an average man, can be carried from the alveo- lar air into the wet surface of the pulmonary epithelium with a difference of oxygen pressure of a little over 3 mm. of mercury. If during mus- cular exercise six or seven times as much oxygen were absorbed, the necessary difference of oxygen pressure would still be only about 20 mm. of mercury, less than 3 per cent, of an atmosphere. It is one thing, however, to know that the necessary oxygen can be taken up into the surface of the pulmonary membrane by a purely physical process, but quite another to prove that it can also be trans- ported with sufficient speed across the thickness of the alveolar wall into the blood. A direct test of this question, made by Krogh, has also yielded an affirmative answer. He determined the amount of carbon monoxide actually taken up in a given time in a human subject. It was about 20 c c. per mm. of partial pressure in. the alveolar air per minute when the subject was at rest, and a little above 30 c.c. when breathing was forced, as it would be during heavy work. Now the speed of diffusion of oxygen is somewhat greater than that of carbon monoxide, so that at rest the subject could have absorbed between 200 and 300 c.c. of oxygen per minute with a difference _of oxygen tension of only 10 mm. of mercury between the alveolar air and the arterial blood. Even during hard muscular work the observed differ- ences of oxygen tension seem adequate to the transport of the necessary amount of oxygen. 264 RESPIRATION So far, then, as the absorption of oxygen is concerned, there is every reason to conclude that it is managed by physical processes alone. Haldane, it is true, still contends that under exceptional conditions when the call for oxygen is much increased, as during active mus- cular exercise, or when an adequate intake is hampered by reduced atmospheric pressure, as at high altitudes, the oxygen tension in the arterial blood may materially exceed that in the alveolar air. Those who have worked with other methods do not, however, grant that even under these conditions there is any excess of oxygen pressure in the blood. It may, of course, be formally admitted that even if diffusion can account for the absorption of the whole of the oxygen, this is not of itself a proof that it is by diffusion that the thing is actually done ; it is only a reason for refusing to call in the aid of a more recondite hypothesis, until the necessity for doing so is clearly demonstrated. As regards the carbon dioxide the evidence is fully as clear. The speed of diffusion of carbon dioxide across such a membrane as the alveolar wall being much greater than that of oxygen, still smaller differences of tension would suffice to permit the whole normal out- put of that gas to be eliminated by diffusion. According to the observations of Bohr, nearly 500 c.c. of carbon dioxide per minute could pass out of the blood by diffusion into the alveolar air under a difference of partial pressure of i mm. of mercury. This is with the ordinary breathing of a man at rest. With the in- creased respiration associated with hard muscular work, this quantity would be increased to between 700 and 800 c.c. The amount of carbon dioxide eliminated during rest in an average adult (300 c.c.) can there- fore be easily excreted by diffusion with a tension-difference of i mm. of mercury. Even during very hard work the necessary tension differ- ence need not be more than 3 mm. With a movement of carbon dioxide so free as this, it is obvious that if the excretion of that gas is solely a matter of diffusion its partial pressure in the arterial blood must be nearly identical with its partial pressure in the alveolar air. A glance at the table on p. 261 shows that while the carbon di- oxide tension of venous blood may sometimes, perhaps generally, exceed that of the alveolar air, the difference is quite small. The average for the observations on man with the pulmonary catheter was 45 mm., which compares with an average alveolar tension of 42 mm. Experiments made by more modern methods have placed the matter beyond doubt. Krogh proved, for example, that in rabbits the carbon dioxide tension in the arterial blood was always slightly (on the average 0-4 per cent, of an atmosphere, or 3 mm. of mercury) higher than in air taken from the bifurcation of the trachea. Now this air is known to contain slightly less carbon dioxide than the alveolar air. Haldane 's results on the regulation of the respiration (p. 281) would be unintelligible unless the carbon dioxide pressure in the arterial blood adjusted itself with great rapidity, to changes THE GASES OF THE BLOOD 265 in the tension of carbon dioxide in the alveolar air, so as always to preserve a definite relation. It is quite generally assumed that this relation is one of practical equality, and the classical method of measuring the carbon dioxide tension in the arterial blood in man is to determine it in the alveolar air. Evidence in favour of the view that there is, besides diffusion, an element of selective secretion in the exchange of gases through the pulmonary membrane has been found by some writers in the results of a study of the gases of the swim-bladder in fishes. This study has demonstrated the existence of animal cells which actually secrete gases. This fact, however, even if it removes a presump- tion against, does not establish a presumption in favour of, the secretion theory of external respiration. These gases consist of oxygen, nitrogen, and usually a small quantity of carbon dioxide, but in very different proportions from those in which they exist in the air or the water. Thus, as much as 87 per cent, of oxygen has been found in the bladder of fishes taken at a considerable depth, but a smaller amount in those captured near the surface. When the gas is withdrawn by puncturing the bladder with a trocar, the organ rapidly refills, and the percentage of oxygen increases. Further, this process of gaseous secretion is under the influence of nerves, for gas ceases to accumulate in the organ when the branches of the vagi that supply it are cut. In the tortoise stimulation of the peripheral end of the vagus causes a fall of gaseous exchange in the corresponding lung, with an accompanying rise in the other lung. But this is a consequence of an alteration in the pulmonary circulation through the vasomotor fibres for the lungs which are known to run in the vagus in this animal. In the mammal, no such effect has been demonstrated, and the decisive proof that the lungs are gas-secreting glands which would be afforded by the discovery of secretory nerves is wanting. We have now completed the description of the phenomena of external respiration, with the discussion of its central fact, the exchange of gases between the blood and the air at the surface of the lungs. It remains to trace the fate of the absorbed oxygen, and to determine where and how the carbon dioxide arises. SECTION V. — INTERNAL OR TISSUE RESPIRATION. Seats of Oxidation. — The suggestion which lies nearest at hand and which, as a matter of fact, was first put forward, is that the oxygen does not leave the blood at all, but that it meets with oxidizable substances in it, and unites with their carbon to form carbon dioxide. While there is a certain amount of truth in this view, oxygen, as already mentioned, being to some extent taken up by freshly shed blood, and also by blood under other conditions, to oxidize bodies, other than haemoglobin, either naturally contained 366 RESPIRATION in it or artificially added, there is no doubt that the cells of the body are the busiest seats of oxidation. This is shown by the presence of carbon dioxide in large amount in lymph and other liquids which are, or have been, in intimate relation with tissue elements ; by its presence, also in considerable amount, in the tissues themselves — in muscle, for instance ; by its continued and scarcely lessened pro- duction not only in a frog whose blood has been replaced by physio- logical salt solution, and which continues to live in an atmosphere of pure oxygen, but in excised muscles ; and by the remarkable con- nection between the amount of this production and the functional state of those tissues. In insects the finest twigs of the tracheae, through which oxygen passes to the tissues, actually end in the cells ; and in luminous insects, like the glow-worm, it has been noticed that the phosphorescence, which is certainly dependent on oxidation, begins and is most brilliant in those parts of the cells of the light- producing organ that surround the ends of the tracheae. Microscopic evidence has been obtained that intracellular oxidation proceeds most rapidly near surfaces like the nuclear and plasma membranes — e.g., in the indophenol (p. 272) and similar reactions the coloured oxidation products are deposited chiefly in and around the nuclei of such cells as liver and kidney cells and frog's red corpuscles (Lillie). The Passage of Oxygen from the Blood into the Tissues. — A fundamental fact of tissue respiration is that the amount of oxygen taken up by the cells depends essentially upon their needs, and not upon the amount of oxygen offered to them in the blood. In every case studied, an increase of functional activity on the part of an organ leads to an increased call for oxygen by that organ, and an increased consumption of oxygen in it. The cells cannot be cajoled, so to say, into consuming more oxygen merely by increasing the available supply. Nor can they be prevented from absorbing and consuming more of whatever supply is available when they are caused to work harder. For skeletal muscle at rest, Barcroft gives 0-004 c.c. per gramme per minute as the oxygen consumption; during maximal activity twenty times as much (o«o8 gramme) . In the heart of a small dog through which blood was pumped by a larger dog the oxygen intake when the heart was beating feebly was, on the average, about o-oi c.c. per gramme of heart-muscle per minute. When the heart was caused to beat very strongly under the influence of adrenalin, the oxygen intake rose in one case to 0-08, and in two others to 0-04. In the resting pancreas the oxygen intake has been found to be 0-03 to 0*05 c.c. per gramme per minute; in the active pancreas, o-i c.c. The corresponding number for the submaxillary gland at rest is 0-03, and in activity 0-09; for the kidney, 0-03 at rest or during scanty secretion, and 0-07 or even 0-09 during active secretion; for the liver in fasting animals o-oi, in fed animals 0-035. INTERNAL OR TISSUE RESPIRATION 267 We are as yet less precisely informed as to the manner in which the tissues regulate the amount of oxygen which they take up from the blood, than in the case of the passage of oxygen into the blood from the lungs. The problem, indeed, is superficially at least a different and probably a more complex one. In the lungs the task is to saturate, or nearly to saturate, the haemoglobin with oxygen, whether the blood passes fast or slow. It is a monotonous invariable process of 'oxygen-grabbing,' with no possibility of its ever being overdone. In the tissues, the task is to meet the widely varying demand from blood which is always charged with oxygen to ap- proximately the same degree. How is this thing managed ? There is little doubt that the process is again fundamentally a matter of diffusion. The oxygen dissociated from the oxyhgemoglobin in the capillaries must find its way through the plasma across the capillary walls into the tissue lymph, and thence into the interior of the cells. There is evidence that it can do so by physical diffusion. For the oxygen tension in the capillary blood is in general higher than the oxygen tension in the tissues. The common statement that the partial pressure of oxjrgen in the tissues is zero, or nearly zero, fits in very well with the conclusion that the oxygen is transported into the cells from the blood of the capillaries by a process of diffusion. For the slope of oxygen pressure thus maintained would ordinarily be greater than in the lungs, since even at the end of the capillary tract the venous blood still has a considerable partial pressure of oxygen. The fact that such liquids as lymph, bile, urine, the serous fluids, saliva, pancreatic juice and milk contain little or no oxygen was sup- posed to support the view that the pressure of oxygen must be very low in the tissues by which they are secreted, or with which they have been in intimate contact. From isolated muscles no free oxygen at all can be pumped out, and muscle being taken as a type of the other tissues in regard to the problems of internal respiration, it was concluded that the scarcity of oxygen in the parenchymatous liquids which bathe the tissues deepens in the tissues themselves into actual famine. The inference seemed plain. Either the tissues used up oxygen so rapidly that with an average blood flow their wants could just be supplied, or without actually consuming it they ' fixed ' and stowed it away in some compound in which it was still available for oxidation in the meta- bolic processes of the cell, but had lost the properties of free oxygen. On the first alternative, an increased consumption of oxygen could only be met by an increase in the blood flow; on the second assumption the stored oxygen would supply the means of temporarily increasing the metabolism even without an immediate augmentation of the blood flow. Although it was recognized that even the resting metabolism of certain organs was not inconsiderable, and required a good supply of oxygen, it was always difficult to understand why inactive tissues should have such an avidity for oxygen that practically every molecule seemed to be captured as soon as it appeared. Nor did it help much to assume that some of the tissue oxidation might really take place in the tissue lymph, oxidizable products split off in the metabolism of the cells passing out into the lymph before being burnt, and thus diminishing the oxygen tension outside the capillary walls. 268 RESPIRATION Recent work has indicated, however, that the oxygen pressure in some tissues is far from negligible. In the resting submaxillary gland, indeed, it seems to be practically equal to the oxygen pres- sure in the venous blood leaving the organ (over 40 mm. of mercury in certain experiments). In general, the same was found to be the case for the kidney. This does not mean, of course, that in such organs there is no slope of oxygen pressure from blood to tissue, in virtue of which oxygen can be moved out of the blood by diffusion. For at the beginning of the capillary tract the whole difference in oxygen pressure between the arterial and the venous blood would be available. At the end of the capillary tract, the blood by the loss of oxygen to the tissues would have come approxi- mately into equilibrium with them. These observations enable us to understand better the manner in which the tissues themselves can regulate their intake of oxygen. If the glands are using little oxygen their oxygen tension will be relatively high, and the passage of a comparatively small amount of oxygen from the blood in the capillaries will suffice to bring it into equilibrium with the tissues, after which the diffusion of oxygen will cease. If, on the other hand, the consumption of oxygen in the glands is suddenly in- creased, their oxygen pressure will fall, the pressure gradient be- tween them and the blood will become steeper, and oxygen will diffuse more rapidly into them from the capillaries. It must be remarked, also, that up to a certain point the existence of a sub- stantial oxygen pressure in the tissues permits them at once to increase their intake of oxygen from the blood without the neces- sity of an immediate increase in the blood flow. For, as the cells consume the oxygen already present in and between them, the tissue oxygen tension falls, and the gradient between the interior and the exterior of the capillaries is thus rendered more abrupt. These results were not obtained by direct tonometric measurement of the oxygen pressure in the glands investigated, but were deduced from observations on the quantities of oxygen taken up by them when different oxygen pressures were produced in the capillaries by altering the oxygen tension in the inspired air. The quantity of oxygen which diffuses out of the capillaries in a given time will be proportional to the difference of oxygen pressure on the two sides of the capillary wall. If the pressure on the side of the tissues is always zero, then the quantity of oxygen passing out will vary directly as the intracapillary oxygen pressure. Far from doing this, the amount of oxygen taken up by the glands was found to be about the same with intra-capillary oxygen pressures as different as 61 mm. and 16 mm. of mercury. It should be distinctly understood that no great importance must be attached to the actual numbers in experiments of this kind. But the general con- clusion that in some tissues, at any rate in the resting condition, a sensible partial pressure of oxygen exists, seems to be established (Yerzar). The application of the same method to muscle has led to a different result. Here, the possible partial pressure of oxygen was found decidedly less than that of the venous blood, although at times a small oxygen pressure was detected in resting muscle (25 mm. or less) 269 Muscle may accordingly be taken as a type of tissues in which the oxygen pressure is so low that the blood-flow through them cannot be much diminished without interference with their absorp- tion of oxygen, while in tissues like the resting submaxillary gland a considerable diminution of the blood-flow can occur without diminution of the oxygen consumption of the tissue. Whether this difference is merely a quantitative one, depending upon the great oxygen requirement of active muscle, or a qualitative difference connected with the peculiar relations of oxygen to the muscular contraction, is unknown. These relations have so great an interest in connection with the problem of intracellular oxidation, and have been so much more minutely studied than the relations of oxygen to the functional work of other tissues, that a few words on the respiration of muscle may fitly be introduced here. The subject must be returned to later on (Chapters XII. and XIII.). Fig. 123. — Fatigue of a Pair of Sartorius Muscles (Fletcher). A, in an atmosphere of oxygen; B, in an atmosphere of nitrogen. A is partially restored by a rest of five minutes. Respiration of Muscle. — It is a remarkable fact that an excised frog's muscle is capable of going on yielding carbon dioxide for a long time, in the entire absence of oxygen, in a chamber, for instance, filled with nitrogen or other indifferent gas. Not only so, but it can be made to contract many times in this oxygen-free atmosphere, although it loses its power of contraction sooner than in oxygen, and does not show the same capacity for recuperation during an interval of rest. In mammals the muscles can also be made to contract repeatedly when the dissociable oxygen has, as far as pos- sible, been got rid of from the blood by asphyxiating the animal, and to give off a correspondingly large quantity of carbon dioxide, although they lose their contractibility much more rapidly than the muscles of the frog. This has usually been interpreted as meaning that the carbon dioxide does not arise, so to speak, on the spot, from *70 RESPIRATION the immediate um'on of carbon and oxygen, but that a stock of it is taken up by the muscle, and stored in some compound or compounds, which are broken down during contraction, and more slowly during rest, carbon dioxide in both cases being one of the end-products. In a normal muscle with intact circulation, while carbon dioxide is given off, certain of the other decomposition products are supposed, in conjunction with oxygen and some sub- stance rich in carbon, like sugar, to be regenerated into the material which breaks down in contraction. When oxygen is not available, as in an atmosphere of nitrogen, carbon dioxide is still given off, but the other decomposition products are supposed not to be regenerated to contractile substance, but to accumulate in the muscle, producing the phenomena of fatigue, and eventually of rigor. When muscle goes into rigor (Chapter XIII.) — and this is most strikingly seen when the rigor is caused by raising the temperature of frog's muscle to about 40° or 41° C. — there is a sudden increase in the quantity of carbon dioxide given off. Moreover, in an isolated muscle the total quantity of carbon dioxide obtainable during rigor is markedly less if the muscle has been previously tetanized. From this it has been argued that the hypothetical substance (' inogen '), the decomposition of which yields carbon dioxide in contraction, is also the substance which decomposes so rapidly in rigor; that a given amount of it exists in the muscle at the time it is removed from the influence of the blood; and that this can all ' explode ' either in con- traction or in rigor, or partly in the one and partly in the other. However, according to Fletcher, there is no increase in the amount of carbon dioxide given off during tetanus by an excised frog's muscle unless the stimulation is so severe and prolonged as to hasten the onset of rigor. He therefore supposes that in the con- traction the decomposition does not proceed quite to the formation of carbon dioxide, which in the intact body is afterwards liberated from some more complex carbon-containing waste-product. He considers that the carbon dioxide yielded by excised muscles is really performed carbon dioxide, already existing in a state of loose combination, from which it is displaced by the lactic acid formed after excision. There is no reason to suppose that any independent new formation of carbon dioxide occurs within the isolated muscle in the absence of a good supply of oxygen. However this may be, there is good evidence that oxygen is used up in recovery processes after the contraction is over, and that these recovery processes are not completed when oxygen is lacking. Hill's work on the heat production of muscle (Chapter XIII.) has tended to rehabilitate the older conceptions, at least to this extent, that his results are in favour of the view that oxygen is used largely during the recovery after contraction in reactions ' whereby the molecular machine — like a steam-engine charging an accumulator — builds up bodies contain INTERNAL OR TISSUE RESPIRATION *7i ing considerable amounts of potential energy which (like the accu- mulator) can be discharged by appropriate stimuli.' The respiration of muscles in situ can be studied by collecting samples of the blood coming to and leaving them and analyzing the gases. The mere difference of colour between the venous and arterial blood of a muscle, or other active organ, is sufficient to show that oxygen is taken up and carbon dioxide given out by it to the blood. This is the case in muscle sat rest, and even in muscles with artificial circulation after they have become inexcitable. In active muscles more oxygen is used up and more carbon dioxide produced than in the resting state. Chauveau and Kaufmann, in their experiments on the levator labii superioris musck of the horse in feeding, found that the consumption of oxygen and the production of carbon dioxide might be many times as great in activity as in rest. Thus in one experiment the amount of oxygen taken in, expressed in c.c. per gramme of muscle per minute, was 0-008 during rest, and 0-14 during work; the corresponding quantities for the carbon dioxide given off were 0-006 and 0-18. The respiratory quotient rose to 1-3 in two experiments, and even to 1-7 in a third, snowing that the increase in the production of carbon dioxide was relatively greater than the increase in the intake of oxygen. These experiments were performed under conditions so normal that the animal continued to eat its hay with seeming unconcern throughout the observations, although these involved the exposure of the main bloodvessels of the muscle, and the collection of samples of blood from them. By means of the modern technique permitting the use of small quantities of blood for the gas analysis, similar experiments have been performed on a muscle as small as the cat's gastrocnemius. In one experiment it was found that as a result of stimulation of the sciatic lasting about 25 seconds the intake of oxygen was increased for at least 220 seconds. In this time the muscle used up 0-75 c.c. of oxygen, as compared with 0-26 c.c., which it would have consumed had it not been stimulated (Verzar). Nature of the Oxidative Process. — When we have recognized the cells as the seat of oxidation, the question immediately presents itself, How do they accomplish the feat of burning such masses of food substances as can only be rapidly oxidized in the laboratory at the temperature of the body by the most energetic chemical reagents ? The researches of late years have furnished a key to the solution of this long-standing puzzle by demonstrating the existence in the tissues of oxidizing ferments or oxydases. Of these, one of the most widely distributed is a ferment which splits off oxygen from hydrogen peroxide. Since any oxidation produced is only secondary to this decomposition, ferments which decompose hydrogen peroxide are often spoken of as catalases, to distinguish them from the oxydases proper. A catalase is found in practically all the tissues of the body, as well as in vegetable cells, and we have already mentioned instances of its action in connection with the oxidation of the guaiaconic acid in tincture of guaiacum in the 272 RESPIRA TION presence of the peroxide (p. 76). As regards the activity of this ferment, blood comes first; then follow spleen, liver, pancreas, thymus, brain, muscle, and ovary. It is present in the blood-free organs as well as in the blood. Some tissues, both animal and vegetable, contain a ferment, .an oxydase, which causes the oxida- tion of guaiaconic acid in the presence of atmospheric oxygen, and these do not need peroxide of hydrogen in order to render guaiacum blue. An allied ferment which also induces the blue colour in tincture of guaiacum is the so-called laccase found in the most active form in the latex of the tree from which Japanese lacquer is ob- tained, but also in many other plants. Many fungi contain a fer- ment, tyrosinase, which oxidizes tyrosin, and in certain animals tyrosinases have also been demonstrated. Another well-known oxidizing ferment in fresh animal tissues is characterized by the property of forming indophenol by oxidation in an alkaline solution of paraphenylenediamin and a-naphthol, and may therefore be termed indophenyloxydase. The colourless solution becomes reddish or violet. This ferment is contained in pancreas, salivary glands, spleen, thymus, and bone marrow, but has not been de- tected in muscle, lungs, brain, kidneys, and other organs. It is to be expected that other oxydases capable of favouring oxida- tion of specific kinds of food substances or their decomposition products will be discovered, but it ought to be remarked that those at present known are only capable of attacking relatively simple organic substances, and it would be rash to conclude that this is the only way in which living protoplasm can bring about the rapid, but at the same time the regulated, oxidation which is so characteristic a feature of its activity. Yet the capacity of the cell to regulate the intensity and the extent of the intra cellular oxidations would seem to find a simple explanation if we assign an important role to oxidizing ferments formed by the cell itself in accordance with its needs. In this connection we may mention a ferment, aldehydase, which was formerly included among the oxydases, but is now known to be a hydrolytic enzyme. It splits aldehydes so as to yield the corresponding acid — e.g., salicylic aldehyde is split into salicylic acid and saligenin. Evidence of its presence in most organs has been obtained. The Passage of Carbon Dioxide from the Tissues into the Blood. — Since nearly the whole of the carbon dioxide eventually found in the blood is formed in the tissues, and only a small amount in the blood itself, it might be supposed that the partial pressure of the gas in the tissues would necessarily be greater than in the blood. This would certainly be true if the whole of the carbon dioxide was transported in the form of dissolved gas. This, however, is not the case. Much of it is combined, and as the proportion of free to combined carbon dioxide in the blood is variable, it may be INTERNAL OR TISSUE RESPIRATION 273 assumed that it is also variable in the tissues. There is no evidence and little probability that the variations are always parallel. Quan- titative and even qualitative differences in the substances which can bind carbon dioxide are known to exist between the blood and the tissues, and it is uncertain how much interchange of such compounds carrying with them combined carbon dioxide, which may after- wards become dissociated, takes place between the blood and the tissue lymph or the cells. However probable, then, it may be that the transportation of carbon dioxide from the cells to the lymph, and from the lymph to the blood is managed in the same way as that of the carbon dioxide from the blood to the alveolar air, namely, by diffusion of the gas in solution, it cannot be said that at present clear proof of this has been obtained. Results are, indeed, on record, which purport to show that the partial pressure of carbon dioxide in various tissues or in physiological liquids which have been in contact with them is higher than that of the arterial or even than that of the venous blood. But these results are of unequal and some of them of doubtful value for the solution of the problem under discussion. Lymph, bile, urine, and the serous fluids contain, so much carbon dioxide that the pressure of that gas in all of them is greater than in arterial blood, while in lymph alone (taken from the large thoracic duct) has it been found less than that of venous blood. And it is pro- bable that lymph gathered nearer the primary seats of its production (the spaces of areolar tissue) would show a higher proportion of carbon dioxide. Strassburg found that with a pressure of carbon dioxide in the arterial blood of 21 mm. of mercury, the pressure in bile was 50 mm. , in peritoneal fluid 58 mm., in urine 68 mm., in the surface of the empty intestine 58 mm. Saliva, pancreatic juice, and milk, also contain much carbon dioxide. From muscle as much as 15 volumes per too of carbon dioxide can be pumped out, some of which is free — that is, is given up to the vacuum alone — while some of it is fixed, and only comes off after the addition of an acid. Before leaving the subject of the gaseous exchange between tissues and blood and between blood and air some important con- sequences of the form of the dissociation curve of oxyhsemoglobin, and of the modifications produced in it by change of temperature, by salts and by acids, must be alluded to. They have been developed in masterly fashion by Barcroft. The flattening of the hyperbolic dissociation curve of dialyzed oxyhaemoglobin when the tempera- ture is raised (Fig. 119) makes it obvious that for any percentage saturation which can exist in blood as it leaves the systemic capil- laries, the corresponding partial pressure of oxygen must be much greater at the body temperature (38° C.) than at 16° C. For haemo- globin which is 50 per cent, saturated it is twenty-five times as great. This favours the giving up of oxygen to the tissues by blood in which the percentage saturation is considerably reduced. On i8 274 RESPIRA TION the other hand, the absorption of oxygen in the lungs is not seriously interfered with, because the available alveolar partial pressure of oxygen is so high (100 mm. or more) that it can be considerably diminished without causing any appreciable diminution in the percentage saturation. There is evidence that in dialyzed haemoglobin solutions the haemoglobin is all in the form of single molecules. The addition of salts as they exist in blood produces a certain amount of aggrega- tion or clumping of the molecules, and as already pointed out, this alters the shape of the dissociation curve. The S-shape of the curve of blood indicates that blood will part with its oxygen at low oxygen pressures, as in passing through the tissues, much more readily than a corresponding solution of salt -free oxyhaemoglobin ; and that it will take up oxygen more easily at high pressures, as in passing through the lungs. As regards the effect of acids, it was observed by Bohr (p. 255) that an increase in the carbon dioxide tension of blood diminishes its combining power for oxygen, and therefore favours the giving up of oxygen to the lymph and tissues. This may have an important influence on internal respiration. The effect is much more marked where the oxygen tension is low than where it is high, so that in the lungs the taking up of oxygen is scarcely interfered with even by a high carbon dioxide tension. According to Barcroft, the combined effect of the increased tem- perature, the salts and the carbon dioxide of blood (at the partial pressure of 40 mm. of mercury) is to make possible a diffusion of oxygen into the tissues at 100 times the speed at which it would diffuse from a salt-free solution of oxyhaemoglobin at a tempera- ture of 16° C. This astonishing feat is accomplished without sensibly decreasing the percentage saturation with oxygen of the blood passing through the lungs. SECTION VI. — RELATION OF RESPIRATION TO THE NERVOUS SYSTEM. The Respiratory Centre and its Connections. — Unlike the beat of the heart, the respiratory movements are entirely dependent on the central nervous system. The ' centre ' which presides over them is situated in the spinal bulb. It is a bilateral centre — that is, it has two functionally symmetrical halves, one on each side of the middle line. Each of these halves has to do more particularly with the respiratory muscles of its own side, for destruction of one-half of the spinal bulb causes paralysis of respiration only on that side. Anatomically the respiratory centre has not been sharply localized, but it lies lower than the vaso-motor centre, not far from the point of the calamus scriptorius. Stimulation of this region during apncea (p. 283) is stated to cause co-ordinated inspiratory movements and RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 275 widening of the opening of the glottis through abduction of the vocal cords. The centre is brought into relation with the muscles of respiration by efferent nerves. The phrenic nerves to the dia- phragm, and the intercostal nerves to the muscles which elevate the ribs, are the most important of those concerned in ordinary breathing. The respiratory centre is further related to afferent nerves, of which the most influential are those which supply the respiratory tract itself, particularly the pulmonary fibres and superior laryngeal branch of the vagus. But almost any afferent nerve may powerfully affect the centre ; and it is also influenced by fibres pass- ing to it from the higher parts of the central nervous system. Section of the spinal cord in animals above the origin of the phrenic nerves causes complete paralysis of respiration, and con- sequent death. The phrenics arise from the third and fourth cervical nerves, and are joined by a branch from the fifth; and in man fracture of any of the four upper cervical vertebrae is as a rule instantly fatal. But in one case respiration was carried on, and life maintained for thirty minutes, merely by the contraction of the muscles of the neck and shoulders in a man entirely paralyzed below this level (Bell). Section of the cord just below the origin of the phrenics leaves the diaphragm working, although the other respiratory muscles are paralyzed. A case has been recorded of a man in whom, from disease of the spine in the lower cervical region, all the ribs became completely immovable. He was able to lead an active life, and to carry on his business, although he breathed entirely by his diaphragm and abdominal muscles. Section of one phrenic is followed by paralysis of the correspond- ing half of the diaphragm, section of both phrenics by complete paralysis of that muscle, and although respiration still goes on by means of the muscles which act upon the ribs, it is usuaUy inadequate to the prolonged maintenance of life. In the horse, however, not only has survival been seen after this operation, but the animal, after the first temporary increase in the frequency of the breathing had disappeared, could be driven in a light vehicle without any marked dyspnoea. The phrenic nuclei in the two halves of the cord are connected across the middle line. For when a semisection of the cord is made between this level and the respiratory centre in the medulla, respiratory impulses are still able to reach both phrenic nerves. In some animals both halves of the diaphragm go on con- tracting. But when, as usually happens, this is not the case, and the diaphragm on the side of the semisection has ceased to act, it at once begins to contract again when the opposite phrenic nerve is cut, and the respiratory impulse, descending from the bulb, is blocked out from the direct, and forced to follow the crossed path. It has been shown that the crossing takes place at the level of the phrenic nuclei, and nowhere else (Porter). 276 RESPIRATION The Regulation of the Respiration through the Afferent Vagus Fibres. — When one vagus is divided, there is little or no change in the respiratory movements. Half an inch of one vagus nerve has been excised in removing a tumour, and the patient showed no symptoms whatever. But section of both vagi in such animals as the dog, cat and rabbit causes respiration to become much deeper and slower, the one change for a time compensating the other, so that the total amount of air taken in and given out, the amount of carbon dioxide eliminated, and the partial pressure of that gas in the pulmonary alveoli are not greatly altered. The relative dura- tion of the two respiratory phases is completely changed, inspira- tion being much more prolonged than expiration. It has been shown that the effect is really due to the loss of impulses that nor- mally ascend the vagi, not to any irritation of the cut ends. For a nerve can be frozen without exciting it ; and when a portion of each vagus is frozen, the respiration is affected in precisely the same way as when the nerves are divided. After section of both vagi certain fibres coming from the brain above the respiratory centre appear to take a share in the regulation of the respiratory movements. The bloodvessels supplying these fibres, or the centres from which they come, can be blocked by injection of paraffin wax into the common or internal carotid, or the bulb can be severed with the knife above the level of the re- spiratory centre, without any effect being produced upon the breath- ing, except that the rate is as a rule somewhat lessened. But when both the vagi and these upper paths are cut the character of the respiration is changed, exceedingly prolonged inspiratory spasms alternating with long periods of complete relaxation of the diaphragm till the animal dies. From these facts it appears that the periodic of thft respiratory centre are being continually controlled and modi- fied by impulses passing up the vagus, and that in the absence of "tnese impulses a certain degree of control is exercised by the higher paths, which, however, do not appear to be normally in action, at any rate to the full measure of their capacity. When the vagi are severed, the control of the higher paths comes into play, and is sufficient still to keep the breathing regular, although it is slowed. When the higher paths are cut off, the vagus of itself is able to regu- late the discharge. But when both are gone, the respiratory centre, freed from nervous control, passes into a condition of alternate spasm and exhaustion. Of the central connections of these upper paths but little is surely known. The corpora quadrigemina, how- ever, seem to contain centres which can affect _the respiration. Certain areas on the cerebral cortex have also been described, the excitation of which modifies the respiratory movements. There is no question that the cortex is connected, and extensively connected, RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 277 with the respiratory centre, since the rate and depth of the co- ordinated respiratory movements, which are universally acknow- ledged to involve the activity of the centre, can be altered not only by the will, but by the most varied psychical events. The rhythmical excitation of the regulating vagus fibres must be brought about by either mechanical stimulation of the nerve- endings in the lungs, due to the alternate stretching and shrinking, or by chemical stimulation of these endings depending on the changes that occur with each respiration in the content of oxygen and carbon dioxide in the alveolar air, and therefore in their pressure (p. 260) in the blood. Both views have found advocates, but whatever influence the chemical changes in the blood "may exert, there is no doubt that the mechanical factors are the more important. That the vagus is really excited is shown by the fact that a negative varia- tion (p. 824) is set up in the nerve when the lungs are inflated. An electrical change is also observed when air is sucked out of the lungs (Alcock and Seemann, Einthoven). When the normal excitation of the vagus fibres by expansion of the lungs is exaggerated by closing the trachea at the end of in- spiration, the diaphragm immediately relaxes, and a long expira- tory pause ensues, broken at last by a series of inspirations much deeper and more prolonged than those which were taking place before occlusion. When the trachea is occluded at the end of expiration, a series of deep and long-drawn inspirations occurs, the first of which begins at the moment when the next normal inspira- tion ought to have taken place had the windpipe been left free. The most obvious explanation of these results is that the expansion of the lungs sets up impulses in the vagi which cut short the in- spiratory activity of the respiratory centre (inspiration-inhibiting fibres), while in collapse impulses are set up which excite it to re- newed inspiratory discharge (inspiration-exciting fibres). Since ordinary expiration is in the main not associated with active muscular contraction, the inspiration-inhibiting fibres would be at the same time expiration-exciting. Clearly this would constitute a so-called ' self-steering ' arrangement, each inspiration leading inevitably to the succeeding expiration, and each expiration providing the neces- sary stimulus for the succeeding inspiration. On this hypothesis section of the vagi must necessarily be followed by slowing of the respiratory movements, and we have seen that this is the case. A rival hypothesis is that the automatic activity of the respira- tory centre leads normally to the discharge of motor impulses to the inspiratory muscles, which are cut short at each expansion of the lungs by the inhibitory action of the vagus, the nerve not being excited during pulmonary collapse, and therefore carrying no in- spiratory impulses to the centre. On this assumption, we may think of the centre as being ' wound up ' like a clock, the periodic 278 RESPIRATION arrival of regulating impulses acting like an escapement movement, and allowing a certain amount of discharge. When the vagi are cut, the inspirations are greatly prolonged and deepened, because the check on the discharge of the centre has been removed. Attempts have been made by experimental stimulation of the vagus trunk to determine whether, as a matter of fact, it contains both inspiratory and expiratory fibres. But the results are neither so clear nor so constant that we can confidently appeal to them in making a decision, and even some of the investigators who main- tain the existence of but one anatomical set of fibres believe that these are affected differently by different kinds of stimulation — momentary stimuli, for example, setting up in them impulses which we may call inspiratory, and long-lasting stimuli impulses which we may call expiratory. Excitation of the central end of the cut vagus below the origin of its superior laryngeal branch, with induction shocks of moderate Fig. 124. — Respiratory Tracings: JJog. A, normal; B, ettect of stimulation of the central end of vagus; C, effect of section of both vagi. (Tracing taken as in Fig. 135, p. 301.) Time-tracing, seconds. strength, certainly causes quickening of respiration. If the excita- tion be strong, there is arrest in the inspiratory phase. A brief mechanical stimulus, or a series of such, has a similar effect. But chemical stimulation (e.g., with a strong solution of potassium chloride) or long-continued mechanical excitation like that produced by stretching or compression of the nerve, or certain kinds of elec- trical stimulation — for instance, the very weakest induction shocks, or the closure of an ascending voltaic current* — cause slowing of the respiratory movements or expiratory standstill. This is also the usual, though not the invariable result of stimulating the superior laryngeal, even when weak induction shocks are employed. With stronger stimulation energetic contractions of the expiratory muscles may occur. These facts undoubtedly suggest the existence in the vagus of two kinds of afferent nerve-fibres that affect the * I.e., a current passing towards the head in the nerve. RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 279 respiratory centre in opposite ways — inspiratory fibres, which stimulate it to greater activity of discharge, and expiratory fibres, which inhibit its action. The latter variety we may suppose to be more numerous in the superior laryngeal, the former in the pul- monary branches of the vagus. And there is nothing forced in the hypothesis that certain kinds of stimuli act particularly on the one set of fibres, and certain kinds on the other, for we have already seen an instance of this in studying the differences between the vaso- constrictor and the vaso-dilator nerves (p. 173). The most probable conclusion, and the one which best reconciles the conflicting hypotheses, is that two sets of fibres are present : (i) Fibres which inhibit inspiration (and cause expiration), and are excited in ordinary inspiration by the expansion of the lungs. (2) Fibres which Fig. 135. — Effect of Stimulation of Central End of Vagus in a Cat. Upper Traes, Respiration; Lower Trace, Blood-Pressure. At the top are the time-trace (seconds), and below it the signal line, the depression in which indicates the duration of the excitation. Practically no effect was produced on the respira- tion, but a fall of blood-pressure with slowing of the heart. cause inspiration (and inhibit expiration), and are excited in strong expiration, as in dyspnoea, by the collapse of the lungs, but are not active in ordinary expiration. However this may be, the facts we have been discussing have an importance of their own, apart from any hypothetical explanations of them. Some of them have been more than once unintentionally illustrated on man. In one case the left vagus trunk was included in a ligature with the common carotid. The respiratory move- ments immediately stopped, the pulse was slowed, and death occurred in thirty minutes (Rouse). The superior laryngeal fibres, unlike those of the vagus proper, are not constantly in action, as section of both nerves has no effect on respiration. Any source of irritation in the larynx may stimulate these fibres and produce a RESPIRA TION cough, which may also be caused by irritation of the pulmonary fibres of the vagus. Action of Other Afferent Fibres on the Respiration. — The cutaneous nerves, and especially those of the face (fifth nerve), abdomen and chest, have a marked influence on respiration. They can be easily excited in the intact body by thermal and mechanical stimulation. A cold bath, for instance, usually causes acceleration and deepening of the respiratory movements ; and the efficacy of mechanical stimu- lation of sensory nerves in stirring up a sluggish respiratory centre is well known to midwives, who sometimes slap the buttocks of a new- Fig. 126. — Effect of Stimulation of Central End of Brachial Nerve on Respiration (Upper Tracing) and Blood-Pressure (Lower Tracing) in the Cat. At the top of the figure are the time-trace (seconds) and the signal line, showing beginning and end of stimulation. born child to start its breathing. The reflex expiratory standstill caused in rabbits by inhalation of such sharp-smelling substances as ammonia, acetic acid, and tobacco-smoke is due to afferent impulses passing up the trigeminus fibres from the mucous membrane of the nose, and is still obtained after section of the olfactory nerves. Another set of afferent nerves which have been supposed by some to bear an important relation to the respiratory centre are those which supply the muscles. We have already noticed that the frequency of respiration is greatly augmented by muscular exercise. The simplest explanation would seem to be that afferent muscular nerves are stimulated either by mechanical compression of their RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 281 terminal 'spindles/ or by the chemical action on them of certain waste products produced in contraction. It is quite likely that this is one way in which the adjustment is achieved. But this is not the only, and perhaps not the most important, way. For an in- crease in the respiratory movements is caused by tetanizing the muscles of a limb whose nerves have been completely severed, and which is indeed connected with the rest of the body by no other structures than its bloodvessels. This can only be due to two things : a direct action on the respiratory centre by the blood that has p"a"ssjed through, and been altered in, the contracting muscles, or an action exerted by the blood indirectly on the centre through the excitation of afferent respiratory nerves whose connection with it is still intact — for example, the other muscular nerves or the pul- monary branches of the vagus. That the action is direct is shown by the fact that after section of the vagi, the sympathetic, and the spinal cord below the origin of the phrenics, an increase in the respiratory movements is still produced by tetanizing a limb. The Chemical Regulation of the Respiration. — However im- portant the regulation of respiration by afferent nervous impulses may be, the normal discharge of the respiratory centre is intimately associated with the gases of the blood. It is generally acknowledged that the centre may be excited both by blood that is rich in carbon dioxide and by blood that is poor in oxygen. Stimulation by deficiency of oxygen has to some minds presented a metaphysical difficulty — namely, that it is not easy to see how the absence of a thing could cause stimulation. The diffi- culty does not exist, but none the less there is some evidence that when oxygen is lacking the respiratory centre can be excited by substances like lactic acid, which are easily oxidizable and rapidly disappear from properly oxygenated blood. Be that as it may, it has been the subject of long-continued dis- cussion whether excess of carbon dioxide or deficiency of oxygen is the more potent stimulus for the respiratory centre. The best evi- dence points to the conclusion that comparatively small alterations in the amount of carbon dioxide in the inspired air cause a relatively great increase in the respiration, while in the case of the oxygen the departure from the normal proportion must be much more decided to bring about any notable effect. Nor is it at all out of harmony with this that, when very large quantities of carbon dioxide (30 per cent, and upwards in rabbits) are inhaled, a condition of narcosis comes on without any previous respiratory distress. For many substances act differently in large and in small doses. HaHane has pointed out how exquisitely sensitive the respiratory centre is to even small changes in the partial pressure of carbon dioxide in the alveo- lar air, and therefore in the arterial blood and the centre itself, and has demonstrated that this is the way in which the amount of the 282 RESPIRATION pulmonary ventilation (the volume of air breathed per unit of time) is chiefly regulated in ordinary breathing. For instance, an increase of as little as 0-2 per cent, of carbon dioxide in the alveolar air, corresponding to an increase of 1-4 mm. of mercury in the partial pressure (p. 247) of the gas, caused an increase in the pulmonary ventilation of 100 per cent. The alveolar oxygen pressure had to be diminished to 13 per cent, of an atmo- sphere before any decided increase in the respiration occurred. During moderate muscular work the percentage of carbon dioxide in the alveolar air, and therefore in the hlpn^ causing an increase in the ventilation, and this is nnp of +HP he hvperpncea associated with about. In_severe work lack of oxygen, with accumulation of lactic acid and olher metabolic products, which increase the hydrogen-ion concentration in the blood and thus stimulate the respiratory centre or render it excitable by smaller pressures of carbon dioxide, also plays a part. There is some evidence that in the case of carbon dioxide also the actual excitation of the respiratory centre is due to the increased hydrogen-ion concentration. Some physiolo- gists hold the view that the respiratory centre reacts in the same way to a given, increase in hvdro^en-ion r.nnrentra+im|_pr> matter what the~acid may be, and that there is nothing specific in the action of carbon dioxide. If it be assumed that changes in the hydrogen-ion concentration of the blood are only caused by sub- stances like carbon dioxide, which are eliminated by the lungs, or by substances like lactic acid, which are either got rid of by oxi- dation or are not liberated into the blood in the presence of a good oxygen supply, no serious theoretical objection can be urged against this conclusion. For increased activity of the respiratory centre would favour the elimination of carbon dioxide and absorption of oxygen, and in both ways would tend to maintain the normal hydrogen-ion concentration in the blood. If, however, the hydrogen- ion concentration can be influenced by substances which are not directly affected by the respiratory process, it would seem unlikely that the respiratory centre should blindly respond to such changes just as if they had been produced by substances, the amount of which it can effectively control. To sum up, the regulation of normal breathing is twofold — a chemical regulation (through the carbon dioxide, possibly by changes produced in the hydrogen-ion concentration in the blood) of the amount of air moved into and out of the lungs per unit of time ; and a nervous regu- lation (chiefly through the vagi] of the rate and depth of the movements necessary to effect the given amount of ventilation. When the vagi have been divided, an increase in the carbon dioxide pressure within certain limits is responded to by an increase in the total ventilation, just as in the normal animal, but the form of the response is different. Whereas in the normal animal both RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 283 the rate and the depth of respiration are increased, in the vagoto- mized animal there is a marked increase in depth, with little or no increase in rate (Scott). When the gaseous exchange in the lungs from any cause becomes insufficient, the respiratory movements are exaggerated, and ulti- mately every muscle which can directly or indirectly act upon the chest -wall is called into play in the struggle to pass more air into and out of the lungs. To a lesser and greater degree of this exag- geration of breathing the terms Hyperpncea and Dyspnoea have been respectively applied. If the gaseous interchange remains insuffi- cient, or is altogether prevented, asphyxia sets in. Sometimes in man impending asphyxia from loss of function by a part of the lungs (with crippling of the lesser circulation), as in pneumonia, may be warded off by inhalations of oxygen. Increase in the temperature of the blood circulating through the spinal bulb, as when the carotid arteries of a dog are laid on metal boxes through which hot water is kept flowing, also causes dyspnoea (heat-dyspncea.) (p. 3°2). But if the temperature be too high, the respiratory movements may be slowed, perhaps by a partial paralysis or inhibition of the respiratory centre. When the blood is cooled the respiration becomes deeper and slower, but if the temperature is greatly and suddenly lowered, the centre may be stimulated and the breathing quickened. In man the increased temperature of the blood in fever is a cause, though not the only one, of the increase in the rate of respiration. Apncea. — The physiological opposite of dyspnoea is apncei. This condition may be produced in an animal by rapid or prolonged artificial respiration. It is especially easy to obtain in an animal in which the circulation through the brain and bulb is interrupted for a time and then restored, while artificial respiration is being kept up. Spontaneous respiration returns after a longer or shorter interval, but if the artificial respiration be still maintained, it again ceases. In a successful experiment the animal remains without breathing for many seconds after the artificial respiration is stopped. In apnosa the chest remains at rest in the expiratory phase if the lungs have been inflated by the artificial respiration and then allowed to collapse of themselves (expiratory apnoea), but in the inspiratory phase if they have been emptied by suction and then permitted of themselves to expand (inspiratory apnosa). The apnoea is not pro- duced, as some have thought, by the accumulation of an excess of oxygen in the blood, for rapid and repeated inflation of the lungs with hydrogen may cause the condition. Indeed, towards the end of the apnosic period the venous blood may be very distinctly poorer in oxygen than normal venous blood. Apnoea is easily caused in man by a period of deep and rapid breathing and in other ways. The essential thing in this chemical or true apnoea (apncea vera) is the lowering of the partial pressure of carbon dioxide in the alveolar air, and therefore in the arterial blood and the respiratory 284 RESPIRATION centre. The carbon dioxide is washed out of the body, so to say, by the excessive pulmonary ventilation. In addition to chemical apnoea, which is obtainable whether the vagi are intact or not, a so-called mechanical apnoea, or apncez vagi, exists — -that is to say, a stoppage of the respiration due to an inhibitory effect produced through the vagi on the respiratory centre when the vagus endings in the lungs are excited mechanically by inflation. Some observers state that this vagus apncea does not outlast the inflation. Others believe that the results of successive inflations can be ' summated ' in the centre, giving rise to an apnoea which persists after stoppage of the artificial respiration. That a ' memory ' of a prolonged rhythmical inflation of the lungs can impress itself in some way on the respiratory centre is shown by the curious phenomenon that in resuscitation of the bulb after a period of anaemia the natural respiration, when it returns, may have for a short time exactly the same rhythm as the artificial respiration which has just been stopped. That the blood when the gaseous exchange in the lungs is inter- ferea with produces dyspnoea by acting on some portion of the brain may be shown in an interesting manner by establishing wrhat is called a cross-circulation in two rabbits or dogs. The vertebral arteries and one carotid are tied in both animals; the remaining carotids are divided and connected crosswise by glass tubes, or, what is better, as it avoids the risk of clotting, they are crossed by- suturing the cut ends, so that the brain of each is supplied by blood from the other. When the respiration is artificially hindered or stopped in one of the animals, it shows no dyspnoea; it is in the other, whose brain is being fed with improperly ventilated blood, that the respiratory movements become exaggerated. The point of attack of the ' venous ' blood has been further localized in the spinal bulb by the observation that when the brain has been cut away above it, the cord severed below the origin of the phrenics, and all other nerves connected with the region between the two planes of section divided, any interference with the gaseous ex- change in the lungs is at once followed by dyspnoea.* Automaticity of the Respiratory Centre. — The question has been raised whether, in the absence of this ' natural ' stimulation by the blood, and of the impulses that constantly reach the centre along its afferent nerves, it would continue to discharge itself, or whether it would sink into inaction. We have already discussed a similar question in regard to the cardiac and vaso-motor centres, and the subject must again present itself when we come to examine the functions of the central nervous system. In the meantime it is only necessary to say that there is evidence that it is not the mere * The conclusion is doubtless correct, but this experiment is not decisive. For the phrenic nerves themselves contain afferent fibres, the stimulation of •which can influence the respiration after section of the vagi. RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 285 presence of carbon dioxide (or other substances) in the blood circu- lating through the respiratory centre which determines the constant excitation of the centre, but rather the accumulation of carbon dioxide, or the increase of hydrogen-ion concentration, in the centre itself when the partial pressure of that gas in the blood is raised. The idea that the contimous excitation of the centre is ' autochthonous ' — in other words, that it is due to an internal stimulating substance or substances manufactured in the centre itself, as well as carried to it in the blood — renders it easy to under- stand that the discharge of the respiratory centre, although modified by the quality of the blood which circulates in it, is not essen- tially dependent on it. Indeed, in cold-blooded animals whose blood has been replaced by physiological salt solution, and (in frogs') even after the circulation has been stopped altogether by excision of the heart, quiet, regular breathing may be seen for a considerable time. Of course, blood is essential for the continued nutrition of the centre and its connections, and it eventually breaks down and ceases to discharge. The respiratory discharge is still less dependent for its initiation upon the arrival of afferent impulses. For after section of the bulb above the centre, of the cord below the origin of the phrenics, of the vagi and of the posterior roots of all the upper cer- vical nerves, the spasmodic respiration which we have already described as occurring when the vagi and the higher paths have been severed continues without essential modification. It has also been observed that during resuscitation of the bulb and upper cervical cord after a period of anaemia, stimulation of afferent nerves, in- cluding the vagi, is entirely without influence on the respiratory movements for some time after respiration has returned, presumably because the synapses (p. 852) on the afferent paths lying within the previously anaemic area are as yet unable to conduct the nerve impulses. Nevertheless, the respiratory centre continues steadily to discharge itself along the efferent paths, whose synapses are situated beyond the anaemic region. Section of the bulb above the level of the respiratory centre, and of the cord below the origin of the phrenic nerves, in addition to the anaemia, makes no essential difference in the result. The initial rate of discharge of the centre thus isolated from afferent impulses is approximately constant in different experiments (about four a minute in cats). Spinal Respiratory Centres. — Although the chief respiratory centre lies in the medulla oblongata, under certain conditions impulses to the respiratory muscles may originate in the spinal cord. Thus, in young mammals (kittens, puppies), especially when the excitability of the cord has been increased by strychnine, in birds and in alli- gators, movements, apparently respiratory, have been seen after destruction of the brain and spinal bulb. In adult cats, when the functions of the brain, medulla, and cervical cord have been abolished by occlusion of their vessels, similar movements of the 286 RESPIRATION thoracic and abdominal muscles may be seen, but they are not suffi cient for effective respiration. No proof has ever been given that in the intact organism the spinal cord below the level of the bulb takes any other part in respiration than that of a mere conductor of nerve impulses; and it is not justifiable to assume the existence of automatic spinal respiratory centres on the strength of such experi- ments as these. Death after Double Vagotomy. — Alterations in the rhythm of respira- tion are not the only effects that follow division of both vagi (or vago- sympathetics) in the neck. In certain animals, at least, this operation is incompatible with life. In the rabbit, as a rule, death takes place in twenty-four hours. A sheep may live three days, and a horse five or six. Dogs often live a week, occasionally a month or even two, and in rare instances they survive indefinitely. The most prominent symp- toms (in the dog), in addition to the marked and permanent slowing of respiration, quickening of the pulse and contraction of the pupils, are difficult deglutition, accompanied by frequent vomiting and pro- gressive emaciation. The appetite is sometimes ravenous, but no sooner is the food swallowed than it is rejected ; and this is particularly true of water or liquid food. Sometimes the rejected food is simply regurgitated after having reached the lower end of the oesophagus, without entering the stomach. The fatal result is usually caused, or at least preceded, by changes of a pneumonic nature in the lungs. The precise significance of the pulmonary lesion is obscure. But it would seem that paralysis of the laryngeal and cesophageal muscles, with the consequent entrance of saliva, food, or foreign bodies, carrying bacteria into the lungs, is responsible to a great extent. And when only a partial palsy of the glottis is produced, by dividing the right vagus below the origin of the recurrent laryngeal, and the left as usual in the neck, pneu- monia either does not occur or is long delayed . It may be that the tissue of the lungs is rendered particularly susceptible to such insults in conse- quence of trophic or vascular changes induced by section of the pul- monary and cardiac fibres in the vagi. It may be quite clearly demon- strated, however, in animals which live for some weeks, that, not- withstanding the paralysis of the glottis associated with aphonia, no pulmonary symptoms may be present till a day or two before death. The picture presented in these cases is that of an animal suffering, above all, from alimentary disturbances. The respiration is, to be sure, very different from the normal in frequency, depth, and type, but there is nothing to suggest that the lungs are the seat of any pathological process. Suddenly the picture changes. Pulmonary symptoms ob- trude themselves. The physical signs of consolidation of the lungs may be detected, and in a short time the animal is inevitably dead. Occasionally the determining cause of the pulmonary lesion seems to be some external circumstance, as a sudden fall of the air temperature. The idea is exceedingly apt to present itself to the observer that the pneumonia is an accident, an acute intercurrent affection breaking the course of a chronic malnutrition, which in any case must have ended in death. Of course, the vagotomized animal is predisposed to this accident, but there is no definite time after section of the nerves at which it must take place. The vomiting is certainly connected with the paralysis and consequent dilatation of the oesophagus ; and by previously making an artificial opening into the stomach or by a surgical prophy- laxis still more heroic, the establishment of a double gastric and oesophageal fistula (p. 4OT). death may be prevented for many months. Elimination of all the pulmonary fibres of the vagi, by extirpation of RELATION OF RESPIRATION TO THE NERVOUS SYSTEM 287 one lung, followed after an interval by section of the opposite vagus in the neck, is not fatal in rabbits. This is also in favour of the view that in double vagotomy the stress falls mainly on the digestive system . Innervation of the Bronchial Muscles. — Both constrictor and dilator fibres for the bronchi are contained in the vagus. They are not constantly in action, but can be reflexly excited, most easily (in the dog and cat) by stimulating the nasal mucous membrane, and particularly a small area well back upon the nasal septum. Cauterization of the corresponding area in man is said to give per- manent relief in certain cases of spasmodic asthma, a condition in which the recurrent attacks of dyspnoea seem, according to the most generally accepted view, to be associated with spasm of the bronchial muscles. Special Modifications of the Respiratory Movements. — Cheyne- Stokes Respiration is the name given to a peculiar type of breathing, marked by pauses of many seconds alternating with groups of respirations. In each group the movements gradually increase to a maximum amplitude, and then become gradually shallower again, till they cease for the next pause. The phenomenon often occurs in certain diseases of the brain and of the circulation, and pressure on the spinal bulb may produce it. In cats in which the circulation in the brain and medulla oblongata has been interrupted for a time and then restored it is often noticed at a certain stage of resuscita- tion of the respiratory centre. In frogs, Cheyne-Stokes breathing has been observed as the result of interference with the circulation in the spinal bulb, ' drowning,' or ligature of the aorta, and also as a consequence of removal of the brain, or parts of it (hemispheres and optic thalami). But it is not peculiar to pathological conditions, being also seen, more or less perfectly, in normal sleep, especially in children, in healthy men at high altitudes, in hibernating animals, and in morphine and chloral poisoning. Well-marked Cheyne-Stokes breathing can be obtained experi- mentally in normal persons in a variety of ways. If, for example, the subject is caused to breathe deeply and frequently for about two minutes, so as to produce a prolonged apncea, the respiration, when it is resumed spontaneously, is of the Cheyne-Stokes type (Haldane). The explanation given by Haldane is that the fall in the partial pressure of the oxygen in the pulmonary alveoli (p. 283) during the primary apnoea, with the consequent fall of oxygen pressure in the arterial blood and the respiratory centre, leads to the production of lactic acid in the respiratory centre and elsewhere, which stimu- lates the centre in the same way as carbon dioxide, and thus, permits it to be excited by a smaller partial pressure of carbon dioxide than that normally necessary. As soon as the pressure of carbon dioxide, which is increasing during the period of apnoea, has reached the exciting value breathing is resumed. The respirations, beginning as very feeble movements, rapidly increase in strength till the 288 RESPIRA TION breathing becomes quite deep or actually dyspnoeic. The store of oxygen is replenished by this thorough ventilation of the lungs, the changes in the excitability of the respiratory centre due to lack of oxygen disappear (perhaps by oxidation of the lactic acid), and the centre relapses into a period of repose. During this period of apnoea the oxygen pressure sinks once more to the point at which the change in the excitability of the respiratory centre by carbon dioxide occurs, and the breathing again starts. In pathological cases the want of oxygen may be associated either with deficient circulation through the bulb-centre or with deficient intake by the lungs. The adminis- tration of oxygen through a mask has been shown in such cases to abolish the periodicity in the respiration, and to render it more normal. Peculiarly modified, but more or less normal, respiratory acts are coughing, sneezing, yawning, sighing, and hiccup. A cough is an abrupt expiration with open mouth, which forces open the previously closed glottis. It may be excited reflexly from the mucous membrane of the respiratory tract or stomach through the afferent fibres of the vagus, from the back of the tongue or mouth, and (by cold) from the skin. Sneezing is a violent expiration in which the air is chiefly expelled through the nose. It is usually excited reflexly from the nasal mucous membrane through the branch of the fifth nerve which supplies it. Pressure on the course of the nasal nerve will often stop a sneeze. A bright light sometimes causes a sneeze, and so in some individuals does pressure on the supra-orbital nerve, when the skin over it is slightly inflamed. Yawning is a prolonged and very deep inspiration, sometimes accompanied with stretching of the arms and the whole body. It is a sign of mental or physical weariness. A sigh is a long-drawn inspiration, followed by a deep expiration. Hiccup, or hiccough, is due to a spasmodic contraction of the dia- phragm, which causes a sudden inspiration. The abrupt closure of "the glottis cuts this short and gives rise to the characteristic sound. The following readings of the intervals between successive spasms were obtained in one attack: 13 sees., 12 sees., 15 sees., 9 sees., 14 sees., etc. — i.e., one-fourth or one-fifth of the frequency of the ordinary respiratory movements. The mere fixing of the attention on the observations soon stopped the hiccup. Hiccup is generally considered to be a reflex movement, brought about through the respiratory centre by afferent impulses originating in the stomach. The irritation may be merely due to some slight digestive disturbance set up by overfilling of the stomach, perhaps. This is exceedingly common in infants. But persistent hiccup may also be a distressing symptom of very formidable diseases — for example, carcinoma of the pylorus. Experimentally, reflex con- tractions of the diaphragm can sometimes be elicited by stimulation INFLUENCE OF RESPIRATION ON THE, BLOOD-PRESSURE 289 of the central end of the vagus at a time when no ' spontaneous respiratory movements are going on. This has been observed, for instance, in cats during resuscitation of the brain after a period of anaemia. In man also, in a case of Cheyne- Stokes respiration accom- panied by hiccup, it was seen that the hiccup persisted during the periods of apncea. If the respiratory centre is the centre 'for the hiccup reflex, it can therefore be excited by afferent nervous im- pulses at a time when it is not excited by the normal chemical stimulus (MacKenzie and Cushny). SECTION VII. — THE INFLUENCE OF RESPIRATION ON THE BLOOD- PRESSURE. We have already stated, in treating of arterial blood-pressure (p. in), that a normal tracing shows a series of waves corresponding with the respiratory movements. The relationship between the respiratory phases and the rise and fall of the blood- pressure is not by any means a simple and invariable one. It depends upon a number of factors, which need not be equally influential under different conditions or in different animals (Lewis). Something depends upon the rate, something upon the relative preponderance of costal and abdominal respiration, and something probably upon the size of the animal. For instance, an inspiratory rise of blood-pressure occurs in man with pure dia- Insp. — Inso - (\ ' A A A Fig. 127. — Respiratory Waves in the Blood-Pressure: Simultaneous Tracings of Movements of Respiration and of Radial Pulse in Human Subject (Lewis). In A the respiration was diaphragmatic; in B, costal. In A the respiratory tracing was taken from the abdominal wall; in B, from the chest. phragmatic, and a fall with pure thoracic, breathing (Fig. 127). In cats with fairly fast and not very deep respiration the blood-pressure rises in expiration and sinks in inspiration. With deep and slow respiration the opposite effect may, upon the whole, be seen. In dogs, according to Einbrodt, although the mean blood-pressure is falling for a short time at the beginning of inspiration, it soon reaches its minimum, then begins to rise, and continues rising during the 19 RESPIRA TION rest of this period. At the commencement of expiration it is still mounting, but soon reaches its maximum, begins to fall, and con- tinues falling through the remainder of the expiratory phase. A partial explanation is afforded by a consideration of the mechan- ical changes produced in the thorax by the respiratory movements. Of these, the influence of variations in the intrathoracic pressure on the filling of the heart is of special importance. With deep abdominal breathing the changes of intra-abdominal pressure also affect the filling of the heart, an increase of pressure (in inspiration) tending to cause more blood to be squeezed from the abdominal veins towards the chest. The changes of vascular resistance in the lungs, due to the alteration in the calibre of the pulmonary vessels, may also contribute, but, for such variations of intrathoracic pressure as normally occur, only in a minor degree. The changes in the vascular capacity of the lungs — that is, in the amount of blood contained in the pulmonary vessels — are of importance espe- cially in delaying or accelerating the alterations of blood- pressure in the systemic arteries due to the other factors. The intrathoracic pressure, which, as we have seen, is always less than that of the atmosphere, unless during a forced expiration when the free escape of air from the lungs is obstructed, diminishes in inspiration and increases in expiration. The great veins outside the chest, the jugular veins in the neck, for example, are under the atmospheric pressure, which is readily transmitted through their thin walls, while the heart and thoracic veins are under a smaller pressure. The venous blood both in inspiration and ex- piration will, accordingly, tend to be drawn into the right auricle. In inspiration the ven- ous flow will be increased, since the pressure in the thorax, and therefore in the pericardial cavity, is diminished; and upon the whole more venous blood will pass into the right heart during inspiration than during expiration. Now, the right ventricle is not in general working as hard as it can work. Hence, the excess of blood which reaches it during an inspiration is at once sent into the lungs, although not even the first of it can have passed through to the left side of the heart at the end of the inspiration, since the pulmonary circulation-time (four to five seconds in a small dog, two to three seconds in a rabbit) is longer than the time of a com- plete inspiration at any ordinary rate. The increase in the quantity of blood pumped into the pulmonary artery will, if not counteracted by other circumstances, tend to raise the blood-pressure in the Fig. 128. — The upper tracing shows the respiratory movements in a rabbit with rather deep and slow diaphragmatic breathing; the lower tracing is the blood-pressure curve; /, inspiration; E, expiration, including the pause. INFLUENCE OF RESPIRATION ON THE BLOOD-PRESSURE 291 artery and its branches, and therefore at once to accelerate the out- flow through the pulmonary veins. This will be aided if at the same time the vascular resistance in the lungs is reduced, as is generally stated to be the case. The left ventricle, like the right, is capable of discharging more blood than it ordinarily receives. The excess of blood coming to it is easily and promptly ejected. The systemic arteries are better filled and the arterial pressure rises. In expiration the contrary will happen. The return of blood to the thorax will be checked. This is well shown by the swelling of the veins at the root of the neck in expiration, their shrinking in inspiration, the so-called respiratory venous pulse. Less blood being drawn into the right heart, less will be pumped into the pul- monary artery, in which the pressure will, of course, fall. The out- flow into the left auricle will thus be diminished — all the more if in the expiratory phase the vascular resistance in the lungs is increased — and the systemic arterial pressure will be lowered. In both cases, however, the change seen in the blood-pressure curve will be belated, and will not coincide exactly with the commencement of the inspira- tion or the expiration. If it is delayed for a period about equal to the length of an inspiration or an expiration, the blood-pressure will be seen to sink in inspiration and to rise in expiration. If the period of delay is less than this, the pressure will be mounting during a part of each respiratory phase and falling during the rest. As to the explanation of the delay, several factors may be concerned. The negative pressure of the thorax acts on the aorta, as well as on the thoracic veins, although, on account of the greater thickness of its walls, to a smaller extent than on the veins. The diminution of pressure in inspiration tends to expand the thoracic aorta, and to draw blood back out of the systemic arteries, while expiration has the opposite effect. And although the hindrance caused in this way to the flow of blood into the arteries during inspiration, and the acceleration of the flow during expiration may not be great, they will, of course, be better marked in small animals with compara- tively yielding arteries than in large animals. Yet, whether great or small, the tendency will be to diminish the pressure in the one phase and increase it in the other. As soon as the changes of pres- sure produced by alterations in the flow of venous blood into the chest and through the lungs are thoroughly established, the arterial effect will be overborne; but before this happens — that is, at the beginning of inspiration and expiration — it will be in evidence, and will help to delay the main change. Another factor in this delay is found in the changes of vascular capacity which take place in the lungs when they pass from the expanded to the collapsed condition. The expansion of the lungs in natural respiration causes a widening of the pulmonary capillaries, with a consequent increase of their capacity and diminution of their 2Q2 RESPIRA TION resistance. When the vessels at the base of the heart are ligatured either at the height of inspiration or the end of expiration, so as to obtain the whole of the blood in the lungs, it is found that they invariably contain more blood in inspiration than in expiration. During inspiration, as we have seen, the right ventricle is sending an increased supply of blood into the pulmonary artery; but before any increase in the outflow through the pulmonary veins can take place, the vessels of the lung must be filled to their new capacity. The first effect, then, of the lessened vascular resistance of the lungs in inspiration is a temporary falling off in the outflow through the aorta, and therefore a fall of arterial pressure. As soon as a more copious stream begins to flow through the lungs, this is succeeded by a rise. In like manner the first effect of expiration, which in- creases the resistance and diminishes the capacity of the pulmonary vessels, is to force out of the lungs into the left auricle the blood for which there is no room. This causes a rise of arterial blood- pressure, succeeded by a fall as soon as the lessened blood- flow through the lungs is established. The changes in the diastolic capacity of the chambers of the heart itself, with the changes of pericardial pressure, must also act in the Fig. 129. — Effect on Blood-Pressure of Inflation of the Lungs: Rabbit. Artificial respiration stopped in inflation at i. Interval between 2 and 3 (not reproduced) 51 seconds, during which the curve was almost a straight line. Time tracing shows seconds. same direction. It is obvious, then, how greatly the rate and depth of respiration in relation to the size of the animal and the other cir- cumstances already mentioned may influence the time relations of the respiratory oscillations in the arterial pressure curve, so that we ought not to expect them to be absolutely constant. In artificial respiration oscillations of blood-pressure, synchronous with the movements of the lungs, are also seen. During inflation (inspiration) the arterial pressure rises; during deflation (expiration) it falls. When artificial respiration is stopped at the height of inflation and the lungs kept inflated (Fig. 129), the arterial blood-pressure falls rapidly, and continues low until the rise of asphyxia begins. In the INFLUENCE OF RESPIRATION ON THE BLOOD-PRESSURE 293 fall of pressure the increased intrathoracic pressure due to the inflation is an important factor. When the respiration is stopped in collapse, instead of a fall a steady rise of pressure occurs (as in Fig. 84, p. 188). This ultimately merges in the elevation due to asphyxia, which shows itself sooner than in inflation. When the tracheal cannula is closed in natural respiration, no initial fall of pressure takes place (Fig. 130). Besides the mechanical effects of the respiratory movements on the circulation, it may be influenced by changes in the cardio- inhibitory and vaso-motor centres synchronous with the rhythm of the respiratory centre. In many animals (the dog, for instance) and in man, it can be very easily made out that the rate of the heart is greater during inspiration, especially towards its end, than in expiration. The phenomenon is especially distinct in deep and slow respiration. It is caused by a rhythmical rise and fall in the activity of the cardio-inhibitory centre, synchronous with the respiratory movements, for the difference disappears after division of both vagi. The normal respiratory oscillations of blood-pressure are not due in any great degree to such changes in the rate of the heart, for they persist after section of the vagi, and they are seen in animals like the rabbit, in which in ordinary breathing little or no variation in the rate of the heart is connected with the phases of respiration. The most probable explanation of the respiratory Fig. 130.— Blood-Pressure Tracing: Rabbit, under Chloral. Natural respiration stopped at I in inspiration, at E in expiration. The mean blood-pressure is scarcely altered; but the respiratory waves become much larger owing to the abortive efforts at breathing. Time tracing shows seconds. variations in the pulse-rate is that the respiratory centre, when it is discharging itself in inspiration, sends out impulses as a sort of over- flow along fibres connecting it with the cardio-inhibitory centre. These increase the tone of that centre, but, owing to the long latent period of the cardio-inhibitory apparatus, the inhibition does not reveal itself till the succeeding expiration. It may be, however, that the impulses discharged from the respiratory centre in inspiration diminish the tone of the cardio-inhibitory centre, and thus lead to acceleration of the heart towards the end of the inspiratory phase. In certain pathological conditions the influence of the respiration on the pulse-rate is exaggerated (so-called ' respiratory arhythmia '). Traube-Hering Curves. — Rhythmical changes in the activity of the vaso-motor centre, also associated with periodic discharges from 294 RESPIRATION the vaso-motor centre, also associated with periodic discharges from the respiratory centre, may be observed under certain conditions — e.g., when in an animal paralyzed by curara, and therefore unable to breathe spontaneously, the artificial respiration is stopped for a time. If such a dose of curara be given as will still permit slight spontaneous respiration to go on, and both vagi be cut, it can be seen on stopping the artificial respiration that the waves on the blood-pressure curve are exactly synchronous with the slow respiratory movements. The Traube-Hering waves sink in inspiration and rise in expiration. The fact that they have invariably a longer period than the natural respiratory movements indicates that they are not concerned in the production of the normal respiratory oscillations of arterial pressure. Probably the reason why the Traube waves appear after section of the vagi is the increased vigour of the slow respiratory discharges, coupled with a hyperexcitability of the vaso-motor centre, due to the long pauses in the aeration of the blood. In the asphyxial rise of pressure in a curarized dog they are constantly seen, and are often observed when the circulation in the medulla Fig. 131. — Traube-Hering Waves as the Blood-Pressure is falling during Occlusion of the Cerebral Arteries in a Cat. oblongata is in any way interfered with (Fig. 131). In addition to the true Traube-Hering waves, other and much longer periodic variations in the blood-pressure are sometimes noticed. If spon- taneous respiration is going on, their long sweeping curves then show the ordinary respiratory waves superposed on them. The normal respiratory oscillations in the veins, as might be expected, run precisely in the opposite direction to those in the arteries, and so do the Traube-Hering curves. The increased flow from the veins to the thorax during inspiration lowers the pressure in the jugular vein, while it increases the pressure in the carotid. The constriction of the small bloodvessels to which the Traube- Hering curves are due increases the blood-pressure in the arteries, because it increases the peripheral resistance to the blood- flow; in EFFECTS OF BREATHING CONDENSED AND RAREFIED AIR 295 the veins it lowers the pressure, because less blood gets through to them. Accordingly, when the Traube-Hering curve is ascending in the carotid, it is descending in the jugular. The respiratory variations in the volume of the brain, which are so striking a phenomenon when a trephine hole is made in the skull, but which can also take place, thanks to the displacement of cerebro-spinal fluid (p. 174), when the cranium is intact, have by some been attributed to interference with the venous outflow from the cranial cavity during expiration, and by others to those changes in the arterial pressure whose causes we have just been discussing. The truth is that neither factor is exclusively concerned. The ques- tion turns largely upon the time-relations of the movements. The swelling of the brain is sometimes synchronous with expiration, and the shrinking with inspiration. Here the damming back of the blood in the sinuses when the outflow is checked by the expiratory rise of pressure in the thoracic veins either conspires with an expira- tory rise of arterial pressure or is more than enough to counter- balance an expiratory fall of pressure in the cerebral arteries if the respiratory conditions are such as to lead to an expiratory fall. But sometimes the dura mater bulges into the trephine hole in inspira- tion and sinks down in expiration. Here the increase in the volume of the brain produced by the increased pressure in the arteries and capillaries in inspiration is more than sufficient to counterbalance the quickened escape of blood from the cerebral veins. SECTION VIII. — THE EFFECTS OF BREATHING CONDENSED AND RAREFIED AIR. These are— (i) mechanical, shown chiefly by changes in the cir- culation, in the blood-pressure, for instance ; (2) chemical. The mechanical effects differ according to whether the whole body, or only the respiratory tract, is exposed to the altered pressure. When the trachea of an animal is connected with a chamber in which the pressure can be raised or lowered, it is found that at first the arterial blood-pressure rises as the pressure of the air of respira- tion is increased above that of the atmosphere. But a maximum is soon reached; and when respiration begins to be impeded, the pressure falls in the arteries and increases in the veins. When the pressure of the air in the chamber is diminished a little below that of the atmosphere, there is a slight sinking of the arterial blood- pressure, which rises if the air-pressure is further diminished. It is clear that any change of the air-pressure which tends to diminish the intrathoracic pressure will favour the venous return to the heart,, and therefore, if the exit of blood from the thorax is not proportionally impeded, the filling of the arteries. An increase in the mtra-alveolar pressure must tend on the whole to increase, and a diminution in it to lessen, the pressure inside the thorax, which always remains equal to the intra-alveolar pressure, minus the elastic tension of the lungs. 296 REBP1RA TlOtf Breathing compressed air should, therefore, under the conditions described, be upon the whole unfavourable to the venous return to the heart and to the filling of the arteries, and the arterial pressure should fall ; while breathing rarefied air should have the opposite effect. But a very great diminution of the intrathoracic pressure is not necessarily favourable to the circulation, since the auricles are then unable to con- tract perfectly. Certain chest diseases have been treated by the use of apparatus by which the patient is made to breathe either compressed or rarefied air ; or to inspire air at one pressure and to expire into air at another pressure. And it has, upon the whole, been found, in agreement with theory, that condensed air cannot help the circulation however it is applied, but always hinders it; while rarefied air aids the circulation both in inspira- tion and in expiration. But the increased work of the in- spiratory muscles may coun- terbalance the advantage. Valsalva's experiment, which is performed by closing the mouth and nostrils after a previous inspiration, and Fig. 132. — Pulse Tracing in Valsalva's Experi- then forcibly trying to expire , ment (Rollett). is an imitation of breathing into compressed air. The intrathoracic pressure is raised, it may be, to considerably more than that of the atmosphere; the venous return to the heart is impeded, and may be stopped ; and the pulse curve is altered in such a way as to indicate first an increase and then a decrease of the arterial blood- pressure succeeded by a second rise (Fig. 132). Muller's experiment, which should be bracketed with Valsalva's, consists in making, after a previous expiration, a strong inspiratory effort with mouth and nostrils closed. Here the intrathoracic pressure is greatly diminished, more blood is drawn into the chest, and upon the whole effects opposite to those of Valsalva's experiment are produced (Fig. 133). Neither experiment is quite free from danger. In both the dicrotism of the pulse becomes more marked. When the whole body is sub- jected to the changed pressure, J . , ,, Fig. 133.— Pulse Tracing in Muller s as in a balloon or on a mountain, Experiment (Rollett). in a diving-bell or a caisson used in building the piers of a bridge, the conditions are very different. For the blood-pressure, the intrathoracic pressure, and the intra- alveolar pressure, all fall together when the pressure of the atmo- sphere is diminished, and all rise together when it is increased. It is possible not only to live, but to do hard manual labour, at very different atmospheric pressures. As regards the chemical effects of condensed and rarefied air, Loewy found that the quantity of oxygen absorbed by a man breath- ing air in the pneumatic cabinet remained constant at all pressures between about two atmospheres and half an atmosphere. At 440 mm. of mercury dyspnoea became evident ; but if the person was now made to work, the dyspnoea passed away, and did not again manifest itself EFFECTS OF BREATHING CONDENSED AND RAREFIED AIR 297 till the pressure was reduced to 410 mm. There are towns on the high tablelands of the Andes, and in the Himalayas, where the barometric pressure is not more than 16 to 20 inches, yet the in- habitants feel no ill effects. And in the caissons of the Forth Bridge the workmen were engaged in severe toil under a maximum pressure of over three atmospheres, while in the caissons of the St. Louis Bridge in America a maximum pressure of over four atmospheres (i.e., more than three atmospheres in addition to the ordinary air- pressure) was reached. Inside the caissons the men sometimes suffer from pain and noise in the ears, due to excessive pressure on the external surface of the tym- panic membrane. If the pressure in the tympanum is raised by a swallowing movement, which opens the Eustachian tube and permits air to enter it, the symptoms generally disappear. The suddenness of the change of pressure has much to do with its effects, and it is found that the men are most liable to dangerous symptoms while passing through the air-lock from the caissons to the external air. It may be concluded, from experiments on animals, that some of the most serious of these — the localized paralysis usually affecting the legs (paraplegia) and the circulatory disturbances — are due to the formation of gaseous emboli, by the liberation of nitrogen in the blood and other body- fluids when the pressure is abruptly reduced. And, indeed, it is found that the symptoms can often be caused to disappear, both in animals and men, by promptly subjecting them again to compressed air. To avoid gas embolism on decompression, the shift should be so short that the body-fluids do not become fully saturated with nitrogen, and the decompression should be slow. Even with a rate of decompression of twenty minutes for each atmosphere of excess pressure, the equilibrium between the dissolved and the atmospheric nitrogen is not entirely established fifteen minutes after decompression. But that the action of air under a high pressure is not merely mechan- ical follows from the singular fact that in pure oxygen at a pressure of 4 to 5 atmospheres, which corresponds to air at 20 to 25 atmospheres, convulsions are often produced in vertebrate animals, while exposure to 6 to 25 atmospheres of oxygen causes dyspnoea and coma, usually without convulsions. All animals, so far as investigated, are instantly convulsed and killed under a pressure of 50 atmospheres of oxygen (Hill and Macleod). Even seeds and vegetable organisms in general are killed in a short time in oxygen at 3 to 5 atmospheres; and an atmosphere of pure oxygen, equal to 5 atmospheres of air, hinders the development of eggs. Lorrain Smith has shown that in small birds and mice exposure for many hours to a pressure of between I and 2 atmospheres of pure oxygen causes pneumonia. He confirms Bert's observations on the acute toxic effects produced by higher pressures, and supposes that in the production of caisson disease the special action of the oxygen at high pressure may play a part as well as the rapid decompression. Even atmospheres containing 80 to 96 per cent, of oxygen under normal barometric pressure produce in rabbits, in 24 to 48 hours, congestion and oedema of the lungs, and finally a fibrinous broncho-pneumonia. Still smaller oxygen pressures may cause similar effects in a longer time (Karsner). When the air-pressure is diminished below a certain limit, death takes place from asphyxia, more or less gradual according to the rate at which the pressure is reduced. The haemoglobin cannot get 298 RESPIRA TION or retain enough oxygen to enable it to perform its respiratory func- tion; its dissociation tension is no longer balanced by an equal or greater partial pressure of oxygen in the air. The tension of carbon dioxide in the blood is also lessened, owing to the dyspnoea and the consequent increase of pulmonary ventilation. To such changes, as well as to the cold, some of the deaths in high balloon ascents must be attributed. Messrs. Glaisher and Coxwell supposed that they reached the height of 37,000 feet; the former became unconscious at 29,000 feet (8,800 metres), at which height the amount of oxygen in the arterial blood would probably not exceed 10 volumes per cent., but recovered during the descent. The symptoms of the 1 mountain sickness ' so familiar to Alpine climbers (nausea, headache, and marked depression), the undue hyperpncea produced by muscular exertion, and the sleep disturbed by irregular breathing, are also mainly due to deficiency of oxygen in the blood. The most rational prophy- laxis is to leave the high peaks severely alone. But for the enthusiasts who cannot do this a portable apparatus for generating oxygen has been devised. Experiments in the pneumatic cabinet indicate that the hyperpncea is due to the indirect action of want of oxygen already referred to in discussing the normal regulation of respiration (p. 281) — that is, to the formation, in consequence of the insufficient oxygen supply, of lactic acid or other substances which have the same influence as carbon dioxide on the respiratory centre — so that less carbon dioxide is required to excite the centre. Although the hyperpncea leads to a diminution in the partial pressure of carbon dioxide in the pulmonary alveoli, there is no evidence that lack of carbon dioxide (' acapnia ') is the primary cause of mountain sickness (Haldane). It must be remem- bered, however, that here the influence of the low barometric pressure is complicated by other conditions. For example, while in the pneu- matic cabinet, as already stated, diminution of the pressure does not affect the oxygen consumption, it is relatively much greater on the high mountain levels both during rest and during work than on the plains. This is not the case in balloon ascents. And evidence has been brought forward that changes in the mechanics as well as in the chem- istry of respiration are concerned (the breathing, for instance, taking on a periodic character, with some approach to the Cheyne-Stokes type — p. 287), and that there is something not connected with the want of oxygen which diminishes the capacity for muscular work. This ' some- thing ' is perhaps a peculiar excitation of the nervous system in the fierce light of those high levels, which acts not only on the retina, but on the skin, and may even affect the distribution of the blood. It is said that a so-called light bath, as used in the treatment of certain diseases, may increase the quantity of blood in rabbits by 25 per cent, in four hours. The shorter wave-lengths which are relatively more intense in the mountain light are most effective. Recent investigations of the effects of high mountain climates have been conducted by Haldane, Henderson, and others, on Pike's Peak, and by Barcroft and his associates on the Peak of Teneriffe and in the Alps. As already mentioned, Haldane concluded that there was evidence of oxygen secretion by the lungs which became more marked with the duration of residence on the mountain (14,000 feet above sea-level). The total oxygen capacity, and therefore the total amount of haemoglobin, gradually increased. The total volume of the blood was but slightly augmented, and the percentage of haemoglobin rose decidedly. The changes in the circulation have been especially studied by Schneider, CUTANEOUS RESPIRATION 299 who found a marked increase in the rate of blood-flow through the hands, associated with an acceleration of the heart beat, dilatation of the arterioles and a fall in the venous pressure. In ' aviator's sick- ness ' the essential factor is also oxygen deficiency. SECTION IX. — CUTANEOUS RESPIRATION. It has already been remarked that a frog survives the loss of its lungs for some time, respiration going on through the skin. Indeed, it has been calculated that in the intact frog, under ordinary conditions, as much as three-quarters of the total gaseous exchange may be cutaneous. Two frogs were seen to live thirty- three days, and one even forty days, after excision of the lungs. The effect of exclusion of the pulmonary respiration on the gaseous exchange depends on the previous intensity of the metabolism. If this is high the gaseous exchange sinks markedly ; if it is low there is scarcely any alteration. At their maximum efficient y, the frog's lungs are capable of sustaining a much greater exchange than the skin. Besides this quantitative, there is a qualitative differ- ence, the carbon dioxide passing more easily through the skin than the oxygen, so that the respiratory quotient is increased by elimination of the lungs. In mammals the structure of the skin is different, and respiration can only go on through it to a very slight extent. The amount of carbon dioxide excreted in man, although only about 4 grm. or 2 litres in twenty-four hours, is much greater than corresponds to the quantity of oxygen absorbed through the skin. It has been as- serted, and no doubt with justice, that some at least of the carbon dioxide given off is due to putrefactive processes taking place on the surface of the body. Such processes, as has already been pointed out, seem also responsible in part for the heavy odour of a ' close ' room. For no harmful products appear to be exhaled from the skin when it is properly cleansed. In spite of the romantic statements to the con- trary in ancient and modern books (for instance, the story of the child that was gilded to play the part of an angel at the coronation of a medieval pope, but died before the ceremony began), the whole of the human skin may be coated with an impermeable varnish without any ill effects. The entire surface of the body of a patient with cutaneous disease was covered with tar, and kept covered for ten days. There was not the least disturbance of any normal function. The serious effects of varnishing the skin in animals are due, not to retention of poisonous substances, but to increased heat loss. Varnishing is not so rapidly harmful in large animals like dogs as in rabbits, which have a relatively great surface and a delicate skin. The danger of widespread superficial burns is well known. But it is not due to diminished excretion by the skin, for death occurs when large cutaneous areas remain uninjured. The patient nearly always dies when a quarter of the whole skin is burnt; yet the remaining tl.ree-quarters may surely be considered capable, from all analogy, of making up the loss by increased activity. One kidney is enough to eliminate the products of the nitrogenous metabolism of the whole body. It is difficult to see why the excretion of the trifling amount of solid matter in the perspiration should be interfered with by the loss of 25 per cent, of the sweat-glands. The real explanation of the serious effects of extensive superficial burns is perhaps the excessive irritation of the sensory nerves, which may lead to changes in the nervous centres, or reflexly in other organs, or the chemical changes in the damaged tissue, for example, in the blood- corpuscles, or the transudation of lymph at the injured part, and con- sequent increase in the concentration of the blood. 300 RESPIRATION PRACTICAL EXERCISES ON CHAPTER IV. i. Tracing of the Respiratory Movements in Man. — Pass a tape through the rings B of the stethograph shown in Fig. 134, and then around the neck or over the shoulders, so as to sup- port the instrument on the chest at a convenient height. Fasten tapes to the hooks and tie them by a slip-knot 'round the chest. The tube E is connected to a recording tambour, writing on a drum. Or use the belt stethograph or spirograph of Fitz (p. 233), fastening the elastic tube round the chest with the chain, and connecting it with a tambour or the bellows recorder shown in Fig. 134. — Stethograph. Fig. 137. Compare the ex- tent of the excursion when the tube is adjusted at different levels over the thorax and abdomen. 2.* Production of Apnoea and Periodic Breathing in Man. — Arrange for taking tracings of the respiratory movements from a fellow-student as in i . Let the subj ect of the experiment recline in a perfectly easy position in an armchair. Let him then breathe deeply and frequently for about two minutes, so as to produce a prolonged apnoea of about two minutes' duration. Whenever any desire to breathe returns, the breathing is to be allowed to take its own course. It may be expected at first to be of the periodic (Cheyne-Stokes) type. 3. Tracing of the Respiratory Movements in Animals. — (a) Set up the arrangement shown in Fig. 135, and test whether it is air-tight. Have also in readiness an induction machine and electrodes arranged for an interrupted current. Anaesthetize a rabbit with chloral or ether (p. 217), or a small dogf with morphine and ether, or A.C.E. mixture. Insert a cannula into the trachea (p. 201), and connect it with the large bottle by a tube. Connect the bottle with a recording tambour adjusted to write on a drum, and regulate the amount of the excursion of the lever by slackening or tightening the screw-clamp. Set the drum off at slow speed, and take a tracing. (6) Then disconnect the cannula from its tube. Dissect out the vagus in the lower part of the neck, pass a ligature under it, but dp not tie it. Connect the cannula again with the bottle, and while a tracing is being taken ligature the vagus. Cut below the ligature and stimulate its central end with weak shocks, marking the time of stimulation on the drum. Repeat the stimulation with strong shocks, and observe the results. * This experiment is only to be attempted under the direct supervision of the demonstrator. f If a large dog is used the bottle should be omitted, the tracheal cannula being connected with the stem of a T-tube. One end of the horizontal limb of the T-tube is connected with the tambour; the other is provided with a rubber tube, which can be partially closed by a screw-clamp to regulate the excursion. Ether may be given when required by connecting the horizontal limb of the T-tube with a bottle with two glass tubes in the cork (p. 201). PRACTICAL EXERCISES 301 (c) Apply a strong solution- of potassium chloride with a .camel's- hair brush to the central end of the vagus while a tracing is being taken, and observe the effect. (d) Isolate the sciatic nerve (p. 212), ligature it, and cut below the ligature. Stimulate its central end while a tracing is being taken. The respiratory movements will be increased. (e) Disconnect the cannula, and isolate the vagus on the other side. While a tracing is being taken, divide it. The respiratory movements will probably at once become deeper and less frequent. (/) Again disconnect the cannula. Isolate the superior laryngeal branch of the vagus. This will be found entering the larynx at the point where the laryngeal horn of the hyoid bone is connected with the thyroid cartilage. If the finger is passed back along the upper border Fig. 135. — Arrangement for Respiratory Tracing. Two glass tubes are inserted through a cork in the mouth of the large bottle. One of them has a small piece of indiarubber tubing on it, which is closed or opened, as may be required, by a screw-clamp. The other is connected by a rubber tube with a recording tambour. The tubulure at the bottom of the bottle is closed by a cork, through which passes a glass tube, connected by a rubber tube with the tracheal cannula. If no bottle with tubulure is available, it is only necessary to pass through the cork, down to the bottom of the bottle, a third glass tube, which is connected with the tracheal cannula. While a tracing is being taken the animal breathes the air contained in the bottle. When this becomes vitiated the respiratory movements are exaggerated and a normal tracing is no longer obtained. For this reason the tracheal cannula must be connected with the bottle only at the moment when a tracing is to be taken. The arrangement is most suitable for a small animal. of the thyroid cartilage, this point will easily be felt. Ligature the nerve, and divide it between the larynx and the ligature. Reconnect the cannula. Take a tracing first with weak, and then with strong stimulation of the central end of the superior laryngeal. (g) Make an incision through the abdominal wall in the linea alba, and study the movements of the diaphragm. Find the nerves from which the phrenics take origin in the neck. In the dog they arise from the fifth, sixth, and seventh cervical nerves. Divide the phrenic fibres 302 RESPIRA TION on one side, and observe that the diaphragm on the corresponding side is now paralyzed. (h) Insert a cannula into the carotid artery. While a respiratory tracing is being taken, allow blood to flow from the artery. Dyspnoea and exaggeration of the respiratory movements will be seen when a consider- able quantity of blood has been lost. Mark and varnish the tracings. In the whole of this experiment the tracheal cannula is to be dis- connected, except when the lever is actually writing on the drum, in order that the period during which the animal must breathe into the confined space of the bottle may be diminished as much as possible. Instead of the method described, the stethograph shown in Fig. 136 may be used to obtain respiratory tracings from animals, a broad canvas band being put round the animal's chest. To each end of this band is clamped with sufficient tension a strong thread (F), fastened to a small metal disc on the inside of the rubber dam closing the obliquely- cut ends of the metal cylinder D. The tube G is connected with a tambour or with a bellows recorder (Fig. 13?)- 4. The Effect of Temperature on tha Respiratory Centre— Heat Dysp- noea.— Set up an arrangement for Fig. 136.— Stethograph (Crile). Fig. 137. — Bellows Recorder. B. a lead tube connected with the small bellows A, which consists of a light wooden base and top. to which is cemented very flexible (organ key) leather, properly creased for expansion and con- traction; C, writing lever. taking a respiratory tracing as in 2 (footnote, p. 300). Anaesthetize a dog, and fasten it, back downward, on a holder. Make an incision in the middle line of the neck, com- mencing a little below the cricoid cartilage, and extending down for 4 or 5 inches. Insert a cannula into the trachea. Isolate both carotid arteries for as great a distance as possible, and arrange them on the brass tubes shown in Fig. 138. Connect two adjacent ends of the tubes by a short rubber tube. Connect one of the remaining ends to a funnel, supported on a stand, and the other to a rubber tube hanging over the table above a large jar. Slip two or three folds of paper between the tubes and the vagus nerves. Heat two or three litres of water to about 65° C. (a) Now connect the tracheal cannula with the tambour. As soon as the tracing is under way, let the hot water run through the funnel and tubes into the jar. Mark on the tracing the point at which the flow of the hot water was begun, and go on passing it until it has produced an effect. Then stop the drum, and circulate water at the ordinary temperature till the breathing is again normal. Then, while a tracing is being taken, pass ice-cold water through the tubes, and again notice the effect. PRACTICAL EXERCISES 303 (6) Expose the sciatic. Pass ice-water through the tubes, and while a respiratory tracing is being taken stimulate its central end with in- duction shocks so weak as just to cause an effect. Pass water at air temperature through the tubes, and repeat the stimulation with the coils at the same distance. Do the same while hot water is being passed through the tubes, and compare the results. Always allow the water to pass for a time before making an observation. 5. Measurement of Volume of Air inspired or expired — Vital Capacity. — A spirometer (Fig. 114, p. 235) of sufficient accuracy for this experi- ment can be made by removing the bottom of a large bottle with a capacity of not less than 4 litres. A good cork, through which passes a glass tube connected with a rubber tube, is fitted into the neck. The bottle is fixed vertically, mouth downwards, the glass tube being closed for the time, and graduated, by pouring in measured quantities of water, say 100 c.c. at a time, and marking the level. The divisions are then etched in. If the cork does not fit air-tight, it is covered with wax. The bottle is swung on two pulleys, counterpoised and immersed, bottom down, in a large glass jar or a small cask nearly full of water. A smaller bottle may be used for the determination of the tidal air, so as to reduce the error of reading. (i) Submerge the bottle to the stopper, after opening Fig. 138.— Arrangement for Heating or Cooling the Blood in the Carotid Arteries. A, cylin- drical portion of tube; B, flattened portion in the groove, between which and A the artery lies; C, cross-section, showing the lumen extend- ing into B; D, rubber tube attached to a brass tube soldered into A. The other end of A has a similar brass tube soldered into it (not shown in the figure). This is connected by a rubber tube with a similar apparatus, on which the other carotid lies. D is connected with a funnel containing hot or cold water or with the outflow tube, as the case may be. the pinchcock on the rub- ber tube. Breathe into the bottle, close the cock, ad- just the bottle so that the level of the water is the same inside and outside, and then read off the level. De- termine the volume of air expired in — (a) A normal expiration after a normal inspiration (tidal air) ; (b) The greatest possible expiration after a normal inspiration (supplemental air plus tidal air) ; (c) The greatest possible expiration after the greatest possible inspiration (vital capacity). (2) Open the cock and raise the bottle till it is nearly full of air. Determine the volume of air inspired in — (a) A normal inspiration after a normal expiration (tidal air) ; (b) The greatest possible inspiration after a normal expiration (complemental air plus tidal air) ; (c)- The greatest possible inspiration after the greatest possible expiration (vital capacity). Make several observations of each quantity, and take the mean. (3) Count the rate of respiration for three minutes, keeping the breathing as nearly normal as possible; repeat the observation; and from the mean result and the amount of the tidal air calculate the quantity of air taken into the lungs in twenty-four hours (pulmonary ventilation). 6. Cardio-Pneumatic Movements. — Fill a U-tube with tobacco- smoke. One end of the tube is placed in the nostril of a fellow-student. 304 RESPIRA TION and made tight with a little cotton-wool. The other nostril and the mouth are closed, and respiration suspended. The column of smoke moves in and out at each beat of the heart. By feeling the apex-beat, try to verify the fact that during systole the cardio-pneumatic movement is inspiratory, and in diastole expiratory. 7. Auscultation of the Lungs. — This is taught in the course of physical diagnosis, but in connection with the subject the student may perform the following experiment on a dog used for some other purpose : Open the trachea as described on p. 201. Insert into it the cross-piece of a glass T-tube of as large a bore as possible, tying the trachea over it on each side of the stem. The stem projecting from the wound is armed with a short piece of rubber tubing, which can be closed at will with a clip. When the tube is thus closed the animal breathes through the glottis in the ordinary way. When the tube is open, and the mouth and nose covered tightly with a cloth, no air goes through the glottis. The tube being closed, listen with the stethoscope or the ear alone over a part of the chest where the vesicular murmur is well heard. If the Fig. 139. — Haldane's Apparatus for measuring the Quantity of CO2 and Aqueous Vapour given off by an Animal. A, chamber into which the animal is put; i and 4, Woulff's bottles filled with soda-lime to absorb carbon dioxide; 2. 3, and 5, Woulflt's bottles filled with pumice-stone soaked in sulphuric acid to absorb watery vapour; B, glass bell-jar suspended in water, by means of which the negative pressure is known; P, water-pump which sucks air through the appar- atus: i and 2 are simply for absorbing the carbon dioxide and water of the ingoing air. rubbing of the hairs below the stethoscope causes disturbing sounds, shave a portion of the skin. Continue listening while an assistant closes the tube and covers up the animal's muzzle. Determine whether any change takes place in the vesicular sound. Repeat the observation while listening over the lower part of the trachea, and determine whether any change takes place in the bronchial breathing sound. 8. Respiratory Pressure. — Connect a strong rubber tube with a glass bulb, and the bulb with a mercurial manometer provided with a scale, (i) Fasten the tube with a little cotton-wool in one nostril, breathe through the other with closed mouth, and observe the amount by which the level of the mercury is altered in ordinary inspiration and ex- piraton. (2) Repeat the observation with forced breathing, pinching the tube at the height of inspiration and expiration, and reading off the maximum inspiratory and expiratory pressure. PRACTICAL EXERCISES ^ (3) Repeat (i) with the tube connected to the mouth by a glass tube held between the lips, and the nostrils open. (4) Repeat (2) with the tube in the mouth and the nostrils closed. 9. Estimation of the Quantity of Water and of Carbon Dioxide given off by an Animal (Haldanes Method). — (i) Connect the apparatus shown in Fig. 139 with the water-pump. Allow a negative pressure of 5 or 6 inches of water to be established in it, as shown by the rise of water in the bell-jar B. Then close the open tube of carbon dioxide bottle i, and clamp the tube between the water-pump and the bell-jar. If the negative pressure is maintained, the arrangement is air-tight. Now weigh bottle 3 and bottles 4 and 5, the last two together. Place a cat in the respiratory chamber A, connect the chamber directly with the water-pump, and test whether it is tight. Then take the stopper out of bottle i, and adjust the rate at which air is drawn through the apparatus. Let the ventilation go on for a few minutes, then insert bottles 3, 4, and 5 again. Note the time exactly at this point, and after an hour dis- connect 3, 4, and 5, and again weigh. The difference of the two weighings of 3 shows the quantity of water given off by the animal in an hour ; the dif- ference in the combined weight of 4 and 5, the quantity of carbon dioxide . Weigh the cat , and calculate the amount of water and of carbon dioxide given off per kilo per hour. (2) For the student it is more convenient to use smaller animals. The mouse may be taken as an example of a warm-blooded animal, and the frog of a cold-blooded. Instead of the Woulff 's bottles use wide test - tubes connected as in Fig. 140, and for the animal chamber a small beaker, closed with a very carefully fitted cork which has been boiled in paraffin. The inlet and outlet tubes of the chamber are to be introduced through this cork. The holes for these are to be bored with the greatest care, and the tubes to be put in while the cork is still hot from boiling in paraffin. Also insert a thermom- eter about 6 inches long registering from o° C. to 45° C. Modeller's wax is to be used finally to render all the junctions air-tight. Add to the series of tubes described in the apparatus a single tube jntaining baryta- water. This is placed to the left of tube 5, and so arranged that the air-current bubbles through the water. As long as the absorption of carbon dioxide is complete, the baryta-water remains clear. Beyond this a water-bottle should be placed to act as a valve and to indicate the negative pressure in the apparatus. It can be most simply constructed by using a cylinder of stout glass tubing in a wide-mouthed bottle containing some water, the inlet and outlet tubes passing through a paraffined cork which seals the upper end of the cylinder. 20 Fig. 140. — Absorption Tubes for CO2 and Moisture. A, soda -lime tube; B, [sul- phuric acid tube; C, wooden frame, in which A and B are supported by wires d ; >, wire hook, which grips the glass tube firmly, and by means of which the tubes are lifted out of the frame in order to be weighed; a, short piece of glass tubing, by taking out which the absorption tubes are disconnected from the rest of the apparatus; e, glass tube going off to animal chamber. 3°8 RESPIRA T/O.V Before making an observation, test whether the apparatus is air- tight, as explained above, after introducing the animal into the cham- ber, sealing the latter with wax, and connecting it with the absorption- tubes. But a negative pressure of 2 or 3 inches of water is a sufficient test for the small apparatus. To make an observation, set the air-current going at the desired rate. Allow it to run for a few minutes till the carbon dioxide, which has accumulated during the testing, has been swept out. At a time which has been decided on and noted, stop the current by disconnecting the water-pump. Disconnect and stopper up the animal chamber, and weigh it as quickly as possible. Connect up again, using only recently- weighed absorption-tubes, and finally connect with the water-pump and allow the current to pass for a definite period, say an hour. The soda-lime should not be too dry, or absorption is not sufficiently rapid. The following facts are made out: (a) Loss of weight by the animal chamber (chiefly loss by the animal) ; (b) gain of the sulphuric acid tube in water; (c) gain of the soda-lime tubes in carbon dioxide. The total loss and total gain do not correspond, the gain being always greater than the loss. The surplus can only be oxygen absorbed by the animal and added to the hydrogen and carbon of its substance to form water and carbon dioxide. Calculate the respiratory quotient (p. 241). 10. Muscular Contraction in the Absence of Free Oxygen (see p. 267). — Pith a frog (brain and cord). Cut off one hind-leg at the middle of the thigh, and strip the skin from it. Pass a thread under the tendo Achillis, tie it, and divide the tendon below it. Free the tendon and the gastrocnemius muscle from the loose connective tissue lying between them and the bones of the leg, and divide the latter just below the knee. Remove superfluous thigh muscles, and fasten the gastrocnemius in a moist chamber by means of the femur. Attach the thread on the tendon to a lever. Connect the poles of the secondary coil of an induc- tion machine by fine copper wires to the femur and the tendon. Put a battery and simple key in the primary, and arrange it for single shocks. Stimulate the muscle and observe the height of the contraction. Now pass into the chamber a current of washed hydrogen gas from a bottle containing granulated zinc, upon which a little dilute sulphuric acid is poured from time to time. The air in the moist chamber will soon be entirely displaced by the hydrogen, but the muscle will contract on being stimulated, and the stimulation can be repeated many times. 11. Oxidizing Ferments. — Wash out the bloodvessels of a dog or rabbit (Practical Exercises, p. 65). Chop up finely portions of pancreas, spleen, muscle, lungs, and kidney, keeping each separate, and avoiding any contamination of one by another. Grind up half of each portion with sand in a small mortar, and extract with a small quantity of water, keeping all the extracts separate. Into each of eleven test-tubes put 10 c.c. of a colourless dilute alkaline solution of paraphenylenediamin and a-naphthol (freshly made by mixing solutions of the two sub- stances in equimolecular proportions* and adding a little sodium carbonate). To five of the tubes add the chopped organs, to five the watery extracts of the organs, and enough water to make the volume equal in all the tubes. To the remaining tube add the same amount of water. Observe in which tube a change of colour takes place (p. 272). * I.e., the weight of each of the two substances in the mixture should be proportional to its molecular weight. A convenient solution contains 0-144 per cent, of a-naphthol and o'io8 pei cent, of paraphenylenediamin. These quantities are one-hundredth-molecular. Sod urn carbonate is added to the amount of 0-25 per cent. The a-naphthol can be kept as a i per cent, solu- tion in 50 per cent, alcohol. CHAPTER V VOICE AND SPEECH Voice. — Sounds of various kinds are frequently produced by the movements of animals as a whole, or of individual organs. The muscular sound, the sounds of the heart and of respiration, we have already had to speak of. Such sounds may be considered as purely accidental as the footfall of a man or the buzzing of a fly. The wings of an insect beat the air, not to cause sound, but to produce motion; the respiratory murmur is a mere indication that air is finding its way into the lungs, it is in no way related to the oxidation of the blood in the pulmonary capillaries. But in many of the higher animals mechanisms exist which are specially devoted to the utterance of sounds as their prime and proper end. In man the voice-producing mechanism consists of a triple series of tubes and chambers : (i) The trachea, through which a blast of air is blown ; (2) the larynx, with the vocal cords, by the vibrations of which sound-waves are set up ; and (3) the upper resonance chambers, the pharynx, mouth, and nasal cavities, in which the sounds produced in the larynx are modified and intensified, and in which independent notes and noises arise. The larynx is a cartilaginous box, across which are stretched, from front to back, two thin and sharp-edged membranes, the (true) vocal cords. In front the cords are attached to the thyroid carti- lage, one a little to each side of the middle line; behind they are connected to the vocal or anterior processes of the pyramidal arytenoid cartilages. The thyroid and the two arytenoids are mounted upon a cartilaginous ring, the cricoid. The arytenoids ! can rotate on the cricoid about a vertical axis, while the cricoid can rotate on the thyroid cartilage around a transverse horizontal axis. The cricoid can thus be raised by the contraction of the crico- thyroid muscle, and the vocal cords stretched. By the pull of the posterior crico-arytenoid muscles, attached to the external or mus- i cular processes of the arytenoid cartilages, the vocal processes are rotated outwards, the cords separated from each other or abducted, and the chink between them, the rima glottidis, widened. When :i the vocal processes are approximated by contraction of the lateral 307 3oS VOICE AND SPEECH crico-arytenoid muscles and the consequent forward movement of the muscular processes, the vocal cords are brought close together, or adducted, and the rima is narrowed. The transverse or posterior arytenoid muscle, which connects the two arytenoid cartilages behind, also helps, by its contraction, to narrow the glottis by shift- ing the cartilages on their articular surfaces somewhat nearer the middle line. Running in each vocal cord, and, in fact, incorporated with its elastic tissue, is a muscle, the thyro-arytenoid, the external portion of which may to some extent cause inward rotation of the vocal processes and adduction of the cords ; but the main function, at least of its inner part, is to alter the tension of the cords. The diagrams in Figs. 141 and 142 illustrate the action of the abductors and adductors of the vocal cords. The crico-thyroid muscle and the deflectors of the epiglottis are supplied by the superior laryngeal branch of the vagus, which also Fig. 141. — Diagrammatic Hori- zontal Section of Larynx to show the Direction of Pull of the Posterior Crico-Arytenoid Muscles, which abduct the Vocal Cords. Dotted lines show position in abduction. Fig. 142. — Direction of Pull- of the Lateral Crico-Arytenoids, which adduct the Vocal Cords. Dotted lines show position in adduction. contains the sensory fibres for the mucous membrane of the larynx above the vocal cords. In the dog 'and rabbit motor fibres also reach the crico-thyroid by the so-called middle laryngeal nerve which arises from the superior pharyngeal branch of the vagus. All the other intrinsic muscles are supplied by the recurrent laryngeal branch of the vagus. It receives these motor fibres from the spinal accessory, and supplies sensory fibres to the mucous membrane of the larynx below the vocal cords and to the trachea. The voice is produced, like the sounds of a reed instrument, by the rhythmical interruption of an expiratory blast of air by the vibrating vocal cords. When a bell is struck, vibrations are set up in the metal, which are communicated to the air. It is not the same with the vibrations of the vocal cords; if they were plucked or struck, they would only produce a feeble note. The air in the mouth, pharynx, larynx, trachea, and lungs is the real sounding VOICE 309 body ; a pulse of alternate rarefaction and condensation is set up in it by the interference, at regular intervals, of the vocal cords with the expiratory blast. Forced abruptly from their position of equi- librium as the blast begins, they almost immediately regain and pass below it, in virtue of their elasticity, and continue to vibrate as long as the stream of air continues to issue in sufficient strength. Not only do they vibrate up and down, but also towards and away from the middle line, so that, at least in the chest voice, they come into contact with each other at each swing. The sound-waves thus set up spread out on every side, impinge on the tympanic membrane, set it quivering in response, and give rise to the sensation of sound. We may say, in a word, that the whole exquisite mechanism of cartilages, ligaments, and muscles, has for its object the production of a sufficient pressure in the blast of air driven through the wind- pipe by an expiratory act, and of a suitable tension in the vibrating cords. An approximation of the cords, a narrowing of the glottis, is essential to the production of voice; with a widely-opened glottis the air escapes too easily, and the necessary pressure cannot be attained. The pressure in the windpipe was found in a woman with a tracheal fistula to be about 12 mm. of mercury for a note of medium height, about 15 mm. for a high note, and about 72 mm. for the highest possible note. The period of vibration of structures like the vocal cords depends on their length, thickness, density and tension ; the shorter, thinner, more dense and less tense a stretched string is, the greater is the vibration frequency, the higher the note. In the child the cords are short (6 to 8 mm.), in woman longer (10 to 12 mm. when slack, 13 to 15 mm. when stretched), in man longest of all (14 to 18 mm. in the relaxed, and 18 to 22 mm. in the stretched position) ; and the lower limit of the voice is fixed by the maximum length of the relaxed cords. A boy or a woman cannot utter a deep bass note, because their vocal cords are relatively short, and do not vibrate with sufficient slowness. It is true that by the action of the crico-thyroid muscle the cords can be length- ened, and that the maximum length in a woman approaches or exceeds the minimum length in a man. But the lengthening of the vocal cords in one and the same individual is always accompanied by other changes — increase of tension, decrease of breadth and thickness — which tell upon the vibration frequency in the opposite way, and more than compensate the effect of the increase of length, so that for high notes the cords are longer than for low. The con- traction of the thyro-arytenoid muscle is a more influential factor in altering the tension of the cords than the contraction of the crico- thyroid. It is probable that, when the highest notes are uttered, only the anterior portions of the cords are free to vibrate, their posterior portions being damped by the approximation of the vocal processes of the arytenoid cartilages by the contraction of the 3io VOICE AND SPEECH lateral crico-arytenoid and transverse arytenoid muscles. The range of an ordinary voice is 2 octaves; by training z\ octaves can be reached; but in exceptional cases a range of 3, and even 3^, octaves (as in the celebrated singer Catalini) has been known. The development of the voice in children is of great interest. At the age of six years the boy's voice has a rather narrower range than the girl's in both directions. The boy's voice reaches its full height in the twelfth and its full depth in the thirteenth year, when the range is almost 3 octaves, its upper limit being a semitone higher than the girl's, but its lower limit a whole tone deeper. When the voice ' breaks ' in boys at the age of puberty it falls about an octave. The control of the vocal organs becomes so incomplete that only in one-fourth of the cases can notes of sufficient steadiness to be used in music be produced. The vocal cords, as may be seen with the laryngoscope, are frequently, though not always, congested. The pitch of a note, while it depends chiefly, as has been said, on the tension of the vocal cords, rises and falls somewhat with the strength of the expiratory blast ; the highest notes are only reached with a strong expiratory effort. The intensity of all vocal sounds is determined by the strength of the blast, for the amplitude of vibration of the cords is proportional to this. Besides pitch and intensity, the ear can still distinguish the quality or timbre of sounds ; and the explanation is as follows: Two simple tones of the same pitch and intensity, that is, the sounds caused by two series of air- waves of the same period and amplitude — of the same frequency and height, to use less technical terms — would appear absolutely identical to the sense of hearing; just as the aerial disturbances on which they depend would be absolutely alike to any physical test that could be applied. But no musical instrument ever produces sound-waves of one definite period, and one only; and the same is true of the voice. When a stretched string is displaced in any way from its position of rest, it is set into vibration; and not only does the string vibrate as a whole, but portions of it vibrate independently and give out separate tones. The tone corresponding to the vibra- tion period of the whole string is the lowest of all. It is also the loudest, for it is more difficult to set up quick than slow vibrations. The ear therefore picks it out from all the rest ; and the pitch of the compound note is taken to be the pitch of this, its fundamental tone. The others are called partial or overtones, or harmonics of the fundamental tone, their vibration frequency being twice, three times, four times, etc., that of the latter. Now, the fundamental tone of a compound note or clang produced by two musical instru- ments may be the same, while the number, period, and intensity of the harmonics are different ; and this difference the ear recognizes as a difference of timbre or quality. The timbre of the voice de- pends for the most part on partial tones produced or intensified in the upper resonance chambers. VOICE 3" A great deal of our knowledge as to the mode and mechanism of the production of voice has been acquired by means of the laryngo- scope (Fig. 143)- This consists of a small plane mirror mounted on a handle, which is held at the back of the mouth in such a position that a beam of light, reflected from a larger concave mirror fastened Dn the forehead of the observer, is thrown into the larynx of the patient. The observer looks through a hole in the centre of the large mirror; and an image of the interior of the larynx is seen in the small mirror, in which the parts that are anterior appear as posterior, the arytenoid cartilages in front, the thyroid behind, and the vocal cords stretching between. The small mirror is warmed to body-temperature before being introduced, so as to prevent the condensation of moisture on it. The tendency to retch, which is Concave Mirror Larynx. Fig. 143. — Diagram of Laryngoscope. caused by contact of the instrument with the soft palate, may be removed or lessened by the application of a solution of cocaine. Examined with the laryngoscope during quiet respiration, the glottis is seen to be moderately, though not widely, open, and the vocal cords almost motionless. Although the portion between the arytenoid cartilages has received the name of glottis respiratoria, in contradistinction to the glottis vocalis between the vocal cords, the rima in its whole extent from front to back is really concerned in the respiratory act. In deep expiration the vocal cords come nearer to the middle line, and the glottis is narrowed ; in deep inspiration they are widely separated, and the rings of the trachea, and even its bifurcation, may be disclosed to view. When a sound is produced a note sung, for example — the cords are approximated (Figs. 144 d 145) ; and with a high note more than with a low- 312 VOICE AND SPEECH The essential difference between the production of notes in the lower register, or chest voice, and in the higher register, or falsetto, has been much debated. The lowest notes which can be uttered by any given voice are chest notes, the highest are falsetto notes; but there is a de- batable land common to both registers, and medium notes can be sung either from the chest or from the head. Chest notes impart a vibration or fremitus to the thoracic walls, from the resonance of the lower air- chambers, the trachea and bronchi; and this can be distinctly felt by the hand. In head notes or falsetto the resonance is chiefly in the upper cavities, the pharynx, mouth, and nose. As to the mechanical conditions in the larynx, there is a pretty general agreement that during the production of falsetto notes the vocal cords are less closely approxi- mated than in the sounding of chest notes. The escape of air is conse- quently more rapid in the head voice, and a falsetto note cannot be maintained so long as a note sung from the chest. But it is only the anterior part of the rima glpttidis that is wider in the falsetto voice ; the whole of the glottis respiratoria, and even the posterior portion of the glottis vocalis, are closed during the emission of falsetto notes. Fig. 144. — Position of the Glottis preliminary to the Utterance of Sound, rs, false vocal cord; vi, true vocal cord; ar, arytenoid cartilage; b, pad of the epiglottis. fi Fig. 145. — Position of Open Glottis. I, tongue; e, epiglottis; ae, ary- epiglottidean fold; c, cartilage of Wrisberg; ar, arytenoid cartilage; o, glottis; v, ventricle of Mor- gagni; tit true vocal cord; ts, false vocal cord. Oertel has stated, and the statement has been confirmed by others, that the free edge of the vocal cord alone vibrates in the falsetto voice, one or more nodes or motionless lines parallel to the edge being formed by the contraction of the internal part of the thyro-arytenoid muscle, which thus acts like a stop upon the cord. Approximation of the vocal cords may take place in certain acts unconnected with the production of voice. Thus, a cough, as has already been mentioned, is initiated by closure of the glottis. During a strong muscular effort, too, the chink of the glottis is obliterated, and respiration and phonation both arrested. The object of this is to fix the thorax, and so afford points of support for the action of the muscles of the limbs and abdomen. But con- siderable efforts can be made even by persons with a tracheal fistula. Speech. — Ordinary speech is articulated voice — voice shaped and fashioned by the resonance of the upper air-cavities, and jointed SPEECH 313 together by the sounds or noises to which the varying form of these cavities gives rise. Here we come upon the fundamental distinction between vowels and consonants. Vowels are musical sounds; con- sonants are not musical sounds, but noises — that is to say, they are due to irregular vibrations, not to regularly recurring waves, the frequency of which the ear can appreciate as a definite pitch. This difference of character corresponds to a difference of origin: the vowels are produced by the vibrations of the vocal cords; the con- sonants are due to the rushing of the expiratory blast through certain constricted portions of the buccal chamber, where a kind of temporary glottis is established by the approximation of its walls. One of these ' positions of articulation ' is the orifice of the lips; the consonants formed there, such as p and b, are called labials. A second articulation position is between the anterior part of the tongue and the teeth and hard palate. Here are formed the dentals, t, d, etc. The ordinary English r, and the r of the Berwickshire and East Prussian ' burr,' also arise in this position through a vibratory motion of the point of the tongue. The third position of articulr.- tion is the narrow strait formed between the posterior portion of th : arched tongue and the soft palate. To the consonants arising here the name of gutturals has been given. They include k, g, the Scottish ch, and the uvular German r. The latter is produced by a vibration of the uvula. The aspirated h is a noise set up by the air rushing through a moderately wide glottis, and some have there- fore included the glottis as a fourth articulation position for con- sonants. Certain sounds like n, m, and ng, when final (as in pen, dam, ring), although produced at the glottis, are intensified by the resonance of the air in the nose and pharynx, and are sometimes spoken of as nasal consonants. As we have said, the vowels are produced by vibrations of the vocal cords, but to what they owe their special timbre or quality has been much discussed. According to the view with which Helm- holtz's name is particularly connected this is due to the reinforce- ment of certain overtones by the resonating cavities, the shape and fundamental tone of which are different for each vowel. When a vowel is whispered, the mouth assumes a characteristic shape, and emits the fundamental tone proper to the form and size of the particular ' vowel-cavity,' not as a reinforcement of a tone set up by the vibrations of the vocal cords, but in response to the rush of air through the cavity; just as a bpttle of given shape and size gives out a definite note when the air which it contains is set in vibration, by blowing across its mouth. A whisper, in fact, is speech without voice ; the larynx takes scarcely any part in the production of the soifiid ; the vocal cords remain apart and comparatively slack ; and the expiratory blast rushes through without setting them in vibration. The fundamental tone of the ' vowel-cavity ' may be found for each vowel by placing the mouth in the position necessary for uttering it, then bringing tuning-forks of different period in front of it, and noting VOICE AND SPEECH which of them sets up sympathetic resonance in the air of the mouth, and so causes its sound to be intensified. The. fundamental tone is lowest for u (as in lute). Next comes o ; then a (as in path] ; then a (as in fane) ; then i ; while e is highest of all. A simple illustration of this may be found in the fact that when the vowels are whispered in the order given, the pitch rises. When u or o is sounded, the buccal cavity has the form of a wide-bellied flask, with a short and narrow neck for u, a still shorter but wider neck for o. For e the tongue is raised and almost in contact with the palate, and the cavity of the mouth is shaped like a flask with a long narrow neck and a very short belly. For i the shape is similar, but the neck is not so narrow. For a (as in path] the vowel-cavity is intermediate in form between that of u and e, being roughly funnel-shaped, and the mouth is rather widely opened. For u (oo) the resonating cavity is made as long as possible, the larynx being depressed and the lips protruded; for e the resonating cavitj is at its shortest, the larynx being raised as much as possible and the lips retracted (Figs. 146 to 148). According to Helmholtz, all that the resonating cavity does is to strengthen certain of the partials or overtones of the laryngeal note. ou Fig. 146. Fig. 147. Fig. 148. If this is true, the partials which give a vowel-sound the timbre by which we recognize it as different from other vowel-sounds cannot preserve the same numerical relation to the fundamental tone when the pitch of the latter is altered. Suppose, for example, that a given vowel is sounded with a pitch corresponding to 100 vibrations a second, and that the partial which is particularly strengthened by the resonance of the mouth cavity is the fifth overtone, corresponding to 600 vibra- tions. Then when the same vowel is sounded with a pitch of 200 vibra- tions, the reinforced partial which will now give the quality to the sound will still correspond to 600 vibrations a second, since this is the rate which most easily elicits the resonance, but it will not now be the fifth but the second overtone. Universally accepted for a time, the Helmholtz theory has been in recent years assailed, especially by Hermann, who bases his criticism on microscopic examination of curves obtained by the Edison phono- graph, and on reproductions of such records obtained by photographing on a moving drum covered with sensitive paper a beam of light re- flected from a small mirror attached to a system of levers whose move- ments follow the curves faithfully and greatly magnify them. Hermann SPEECH 315 has come to the conclusion that the mouth does not act as a mere resonator, but that for each vowel, in addition to the fundamental note due to the vibration of the vocal cords, the pitch of which is, of course, variable, one or, it may be, two other notes (formants, as he calls them), not necessarily harmonics of the laryngeal note, but separ- ated from it by a constant or nearly constant musical interval, are directly produced by the passage of the regularly interrupted expiratory blast through the mouth, the air contained in that cavity being for an instant set into vibration at each interruption. On this view it is the musical effect produced by the oscillation or continual recurrence, in short series, of these vibrations which gives the vowels their quality. The fact that it is by no means difficult to sing (with the larynx) and whistle (with the mouth) at the same time, shows the possibility of Hermann's view, that a fixed tone can be generated in the mouth by the intermittent stream of air issuing from between the vibrating vocal cords, just as a tone is generated in a pipe by blowing into or over it, and his records do show continually recurring groups of vibrations as his theory requires. McKendrick takes up a middle position, believing that both theories are partially true, and this seems to be the best conclusion which can at present be arrived at. It seems clear, at any rate, that more than one factor is concerned in the timbre of the vowel sounds. When the vowels are being uttered, the soft palate closes the entrance to the nasal chambers completely, as may be shown by holding a candle in front of the nose, or trying to inject water through the nares. If the cavities of the nose are not completely blocked off, the voice assumes a nasal character in pronouncing certain of the vowels; and in some languages this is the ordinary and correct pronunciation. Many animals have the power of emitting articulated sounds; a few have risen, like man, to the dignity of sentences, but these only by imitation of the human voice. Both vowels and consonants can be distinguished in the notes of birds, the vocal powers of which are in general higher than those of mammalian animals. The latter, as a rule, produce only vowels, though some are able to form con- sonants too. The nervous mechanism of voice and speech will have to be again considered when we come to study the physiology of the brain and spinal cord. But the curious physiological antithesis between the functions of abduction and of adduction of the vocal cords may be mentioned here. The abductor muscles are not employed in the production of voice; they are associated with the less specialized, the less skilled and purposive function of respiration. The adductor muscles are not brought into action in respiration; they are asso- ciated with the highly specialized function of speech. Correspond- ing to this difference of function, we find that adduction is pre- ponderatingly represented in the cortex of the brain, abduction in the medulla oblongata. Stimulation of an area in the lower part of the ascending frontal convolution, near the fissure of Rolando, in the macaque monkey, causes adduction of the vocal cords, never 3*6 VOICE AN'D SPEECH abduction. In the eat, however, abduction of the cords may also be obtained by stimulation of the cortex. The same is true of the dog, but only when the peripheral adductor nerves have been divided. Stimulation of the medulla oblongata (accessory nucleus) causes abduction, never adduction. The skilled adductor function is, therefore, placed under control of the cortex. The vitally im- portant, but more mechanical, abductor function is governed by the medulla. The abductor movements are more likely to be affected by organic disease, the adductor movements by functional changes. But the distinction between the two groups of muscles is not entirely due to a difference of central connections, since by altering the strength of the stimulus and the external conditions the one or the other may be separately excited through the inferior laryngeal nerve. Thus, strong stimulation of the inferior laryngeal causes closure of the glottis, for although it supplies both abductors III Fig. 149. — Diagram of Vocal Cords in Paralyses of the Larynx, a, Paralysis of both inferior laryngeal nerves. The vocal cords have taken up the ' mean ' position. b. Paralysis of right inferior laryngeal nerve. An attempt is being made to narrow the glottis for the utterance of sound. The right cord remains in its 5 mean ' position, 'c. Paralysis of the abductor muscles only, on both sides. The cords are approximated beyond the ' mean ' position by the action of the adductors. and adductors, the latter, as the stronger muscles, prevail. With weak stimulation, and in young animals, the abductors, owing to the greater excitability of the neuro-muscular apparatus, carry off the victory, and the glottis is opened (Russell). When the nerve is cooled the abductors give way before the adductor;. The same is true when it is allowed to become dry. And after death in a cholera patient it was observed that the pos- terior cric.o-arytenoid, an abductor muscle, was the first of the intrinsic laryngeal muscles to lose its excitability. Lesions of the medulla oblongata are often accompanied by marked changes in the character of the voice and the power of articulation. Section or paralysis of the superior laryngeal nerve causes the voice to become hoarse, and renders the sounding of high notes an impossibility, owing to the want of power to make the vocal cords tense. Stimulation of the vagus within the skull causes contraction SPEECH ^17 of the crico-thyroid muscle and increased tension of the cords. Sec- tion or paralysis of the inferior laryngeal nerves leads to loss of voice or aphonia, and dyspnoea (Fig. 149). Both adductor and abductor muscles are paralyzed ; the vocal cords assume their mean position — the position they have in the dead body — -and the glottis can neither be narrowed to allow of the production of a note, nor widened during inspiration. It is said, however, that young animals, in which the structures around the glottis are more yielding than in adults, can still utter shrill cries after section of the inferior laryngeals, the contraction of the crico-thyroid muscle alone being able, while in- creasing the tension of the cords, to draw them together. Interference with the connections on one side between the higher cerebral centres and the medulla oblongata, as by rupture of an artery and effusion of blood into the posterior portion of the internal capsule (giving rise to hemiplegia, or paralysis of the opposite side of the body), is not followed by loss of voice; the laryngeal muscles on both sides are still able to act. CHAPTER VI DIGESTION IN the last chapter we have described the manner in which the interchange of gases between the tissues and the air is carried out. We have now to consider the digestion and absorption of the solid and liquid food, its further fate in relation to the chemical changes or metabolism of the tissues, and finally the excretion of the waste products by other channels than the lung. Logically, we ought to take metabolism after absorption and before excretion, tracing the food through all its vicissitudes from the moment when it enters the blood or lymph till it is cast out as useless matter by the various excretory organs. Unfortunately, however, many of the intermediate steps of the process are as yet hidden from us ; we know best the beginning and the end. We can follow the food from the time it enters the alimentary canal till it is taken up by the tissues of absorption; and we have really a fair knowledge of this part of its course. We can collect the end-products as they escape in the urine, or in the breath, or in the sweat; and our knowledge of them and of the manner in which they are excreted is considerable. But of the wonderful pathway by which the dead molecules of the food mount up into life, and then descend again into death, we catch only a glimpse here and there. Only the introduction and the conclusion of the story of metabolism are at present in our possession in fairly continuous and legible form. We will read these before we try to decipher the handful of torn leaves which represents the rest. SECTION I. — PRELIMINARY ANATOMICAL AND CHEMICAL DATA. Comparative. — In the lowest kinds of animals, such as the amoeba, there is neither mouth, nor alimentary canal, nor anus: the food, wrapped round by pseudopodia, is taken in at any part of the animal with which it happens to come in contact. A vacuole is formed around it. Acid is secreted into the vacuole, the food is digested within the cell-substance, and the part of it which is useless for nutrition is cast out again at any part of the surface. Coming a little higher, we find in the Coalenterates a mouth and alimentary tube, which opens into the body-cavity, where a certain PRELIMINARY ANATOMICAL AND CHEMICAL DATA 319 amount of digestion seems to take place, and from which the food is absorbed either through the cells of the endoderm, or, as in Medusa, by means of fine canals, which radiate from the body-cavity into its walls, and form part of the so-called gastro-vascular system. In the Echinodermata we have a further development, a complete alimentary canal with mouth and anus, and entirely shut off from the body-cavity. In many Arthropods it is possible already to distinguish parts corre- sponding to the stomach, and the small and large intestines of higher forms, the digestive glands being represented by organs which in some groups seem to be homologous with the liver, and in others with the salivary glands of the higher Vertebrates. A few Molluscs seem in addition to possess a pancreas. Among Vertebrates fishes have the simplest, and birds and mammals the most complicated, alimentary system. In the lowest fishes the stomach is only indicated by a slight widening of the anterior part of the digestive tube. In water-living Vertebrates there are no salivary glands. In birds the oesophagus is generally dilated to form a crop, from which the food passes into a stomach consisting of two parts, one pre-eminently glandular (proventriculus), the other pre-eminently muscular (ventriculus). Among mammals a twofold division of the stomach is distinctly indicated in rodents and cetaceae, but this organ reaches its greatest complexity in ruminants, which possess no fewer than four gastric pouches. The differentiation of the intestine into small and large intestine and rectum is more distinct, both anatomically and functionally, in mammals than in lower forms ; but there are marked differences between the various mammalian groups both in the relative size of the several parts of the digestive tube, and in the proportion between the total length of the alimentary canal and the length of the body. In general, the canal is longest in herbivora, shortest in carni- vora. Thus, the ratio between length of body and length of intestine is in the cat I : 4, dog i : 6, man I : 5 or 6, horse i : 12, cow i : 20, sheep i : 27. The relative capacity of the stomach, small intestine, and large intestine, is in the dog 6:2: 1-5, in the horse i : 3-5 : 7, in the cow 7:2:1. The area of the mucous surface of the alimentary canal is very considerable, in the dog more than half that of the skin, the surface of the small intestine being three times that of the stomach and four times that of the large intestine. In the horse the mucous surface has twice the area of the skin. Anatomy of the Alimentary Canal in Man. — The alimentary canal is a muscular tube, which, beginning at the mouth, runs under the various names of pharynx, oesophagus, stomach, small intestine, large intestine, and rectum, till it ends at the anus. Its walls are largely composed of muscular fibres; its lumen is clad with epithelium, and into it open the ducts of glands, which, morphologically speaking, are involutions or diverticula formed in its course. In virtue of its muscular fibres it is a contractile tube; in virtue of its epithelial lining and its special glands it is a secreting tube ; in virtue of both it is fitted to per- form those mechanical and chemical actions upon the food which are necessary for digestion. Its inner surface is in most parts richly supplied with bloodvessels, and in special regions beset with peculiarly- arranged lymphatics; by both of these channels the alimentary tube performs its function of absorption. From the beginning of the osso- phagus to the end of the rectum the muscular wall consists, broadly speaking, of an outer coat of longitudinally -arranged fibres, and a thicker inner coat of fibres running circularly or transversely around the tube. Between the layers lies a plexus of non-medullated nerves and nerve-cells (Auerbach's plexus). In the stomach the longitudinal 32* DIGESTION fibres are found only on the two curvatures, and a third incomplete coat of oblique fibres makes its appearance internal to the circular layer. In the large intestine, again, the longitudinal fibres are chiefly collected into three isolated strands. In the pharynx the typical arrangement is departed from, inasmuch as there is no regular longi- tudinal layer; but the three constrictor muscles represent to a certain extent the great circular coat. The muscles of the mouth and of the pharynx are of the striped variety. So is the muscle of the upper half of the oesophagus in man and the cat, and of the whole oesophagus in the dog and the rabbit. In the rest of the alimentary canal the muscle is smooth, except at the very end, where the external sphincter of the anus is striped. In certain situations the circular coat is de- veloped into a regular anatomical sphincter, a definite muscular ring, whose function it is to shut one part of the tube off from another (sphincter pylori, ileo-colic sphincter), or to help to close the external opening of the tube (internal sphincter of anus). Elsewhere a tonic contraction of a portion of the circular coat, not anatomically de- veloped beyond the rest, creates a functional sphincter (cardiac sphincter of stomach). Throughout the greater part of the digestive tract the peritoneum forms a thin serous layer, external to the muscular coat. Internally the muscular coat is separated from the mucous membrane, the lining of the canal, by some loose areolar tissue containing bloodvessels, lymphatics, and nerves (Meissner's plexus), and called the submucous coat. Between the mucous and submucous layers, but belonging to the former, in the whole canal below the beginning of the oesophagus, is a thin coat of smooth muscular fibres, the muscularis mucosae, con- sisting in some parts, e.g., in the stomach, of two, or even three, layers. Between this and the lumen of the canal lie the ducts and alveoli of glands, surrounded by bloodvessels and embedded in adenoid or lymphoid tissue, which in particular regions is collected into well- defined masses (solitary follicles, Peyer's patches, tonsils), extending, it may be, into the submucous tissue. In the mouth, pharynx, and oesophagus, the glands lie in the submucosa, as do the glands of Brunner in the duodenum; everywhere else they are confined to the mucous membrane proper. Between the openings of the glands the mucous membrane is lined with a single layer of columnar epithelial cells, some- times (in the small intestine) arranged along the sides of tiny projec- tions or villi. When the intestine is contracted the villi are long and cylindrical in shape, when it is relaxed or distended they are flat and conical. At the ends of the alimentary canal, viz., in the mouth, pharynx, and oesophagus, and at the anus, the epithelium is stratified squamous, and not columnar. The purpose of food is to supply the waste of the tissues, to replenish the stores of material from the oxidation of which the energy required for the running of the bodily machine is derived, and thus to maintain the normal composition of the body. In the body we find a multitude of substances marked off from each other, some by the sharpest chemical differences, others by characters much less distinct, but falling upon the whole into the few fairly definite groups already described (p. i). Now, although it is by no means necessary that a substance in the body belonging to one of these great groups should be derived from a substance of the same group in the food, it has been found PRELIMINARY ANATOMICAL AND CHEMICAL DATA 321 that upon the whole no diet is sufficient for man unless it contains representatives of all; a proper diet must include proteins, carbo- hydrates, fats, inorganic salts, and water. These proximate prin- ciples have to be obtained from the raw material of the foodstuffs — that is, as regards the first three groups, which can alone yield energy in the body, from the tissues and juices of other living things, plants or animals; it is the business of digestion to sift them out and to prepare them for absorption. This preparation is partly mechan- ical, partly chemical. The water and salts and some carbo-hydrates, such as dextrose, are ready for absorption without change. Fats are split into glycerin and fatty acids before absorption. Indiffusible colloidal carbo-hydrates, like starch and dextrin, are changed into diffusible and readily soluble sugars, and the natural proteins into diffusible peptones, and eventually into much simpler decomposition products. These changes are obviously favourable to absorption. But this is not their whole significance. For disaccharides, such as cane-sugar, maltose, or lactose, although easily soluble in the contents of the gut, and in themselves perfectly capable of being absorbed without change, are, unless present in unusually large amount, all converted into monosaccharides, such as dextrose, levulose, or galactose, either in the lumen or in the wall of the alimentary tube. The reason is that the disaccharides are unsuitable as pabulum for the cells. Digestion is not only a preparation of the food for absorption by the gut, but for assimilation by the tissues after absorption. An equally important instance of this double function is seen in the digestion of proteins. The complete shattering of the protein mole- cule into amino-acids and the other groups yielded by its decom- position (p. 360) is required, in the case of that portion of the protein which goes to build up the tissues, because of the high degree of specificity of the tissue proteins. The myosinogen of beef cannot be cobbled into the myosinogen of human muscle, still less we may suppose into the serum-albumin of human blood. It is necessary that the food protein should be completely ' wrecked ' in digestion so that protein which is to take its place in protoplasm may be built exactly to order from the bricks. A satisfactory ' fit ' cannot be obtained with ready-made protein. Mechanical division of the food is an important aid to the chemical action of the digestive juices. We shall see that this mechanical division forms a great part of the work of the stomach, but it is normally begun in the mouth, and it is of consequence that this preliminary stage should be properly performed. SECTION II. — THE MECHANICAL PHENOMENA OF DIGESTION. Mastication. — -It is among the mammalia that regular mastication of the food first makes its appearance as an important aid to diges- tion. The amphibian bolts its fly, the bird its grain, and the fish 322 DIGESTION its brother, without the ceremony of chewing. In ruminating mammals we see mastication carried to its highest point ; the teeth work all day long, and most of them are specially adapted for grinding the food. The carnivora spend but a short time in masti- cation; their teeth are in general adapted rather for tearing and cutting than for grinding. Where the diet is partly animal and partly vegetable, as in man, the teeth are fitted for all kinds of work ; and the process of mastication is in general neither so long as in the purely vegetable feeders, nor so short as in the carnivora. In man there are two sets of teeth: the temporary or milk teeth, and the permanent teeth. The milk teeth are twenty in number, and consist on each side of four incisors or cutting-teeth, two canines or tearing-teeth, and four molars or grinding-teeth. The central incisors emerge at the seventh month from birth, the other incisors at the ninth month, the canines at the eighteenth, and the molars at the twelfth and twenty-fourth month respectively. Each tooth in the lower jaw appears a little before the corresponding one in the upper jaw. Each of the milk teeth is in course of time replaced by a permanent tooth, and in addition the vacant portion of the gums behind the milk set is now filled up by twelve teeth, six on each side, three above and three below. These twelve are the permanent molars; they raise the number of the permanent teeth to thirty-two. The permanent teeth which occupy the position of the milk molars now receive the name of premolars. The first tooth of the permanent set (the first true molar) appears at the age of 6£ years; the last molar, or wisdom-tooth, does not emerge till the seventeenth to the twenty-fifth year. In mastication the lower jaw is moved up and down, so as to alternately separate and approximate the two rows of teeth. It has also a certain amount of movement from side to side, and from front to back. The masseter, temporal and internal pterygoid muscles raise, and the digastric, with the assistance of the mylo- and genio- hyoid, depresses, the lower jaw, but its downward movement is mainly a passive one. The external pterygoids pull it forward when both contract, forward and to one side when only one con- tracts. The lower fibres of the temporal muscle retract the jaw. The buccinator and orbicularis oris muscles prevent the food from passing between the teeth and the cheeks and lips. The tongue keeps the food in motion, works it up with the saliva, and finally gathers it into a bolus ready for deglutition. Deglutition. — This act consists of a voluntary and an involun- tary stage. Just before the beginning of the voluntary stage mastication is suspended, and a slight contraction of the dia- phragm generally takes place. The anterior part of the tongue is suddenly elevated and pressed against the hard palate, and the elevation travels back from the tip towards the root, as the mylo- THE MECHANICAL PHENOMENA OF DIGESTION 323 hyoid muscles in the floor of the mouth contract sharply so as to thrust the bolus through the isthmus of the fauces. As soon as this has happened, and the food has reached the posterior portion of the tongue, it has passed beyond the control of the will, and the second or involuntary stage of the process begins. This stage may be divided into two parts: (i) Pharyngeal, (2) oesophageal — both being reflex acts. During the first the food has to pass through the pharynx, the upper portion of which forms a part of the respiratory tract, and is in free communication with the larynx during ordinary breathing. It is therefore necessary that respiration should be interrupted and the larynx closed while the food is being moved through the pharynx. But that the inter- ruption may be short, the food must be rapidly passed over this perilous portion of its descent. The main propelling force under which the bolus is shot through the back of the pharynx is derived from the contraction of the mylo-hyoid muscles already mentioned, assisted to some extent by the stylo- and palato-glossi ; and that none of the purchase may be lost, the pharyngeal cavity is cut off from the nose and mouth as soon as the bolus has entered it. The soft palate is raised by the levator palati and palato-pharyngei muscles; at the same time the upper part of the pharynx, narrowed by the contraction of the superior constrictor, comes forward to meet the soft palate, closes in upon it, and so prevents the food from passing into the nasal cavities. The pharynx is cut off from the mouth by the closure of the fauces through the contraction of the palato-pharyngeal muscles which lie in their posterior pillars. The upper free end of the epiglottis (the so-called pharyngeal part) aids the back of the tongue in completing a movable partition across the pharynx, which keeps close to the bolus as it passes down between the posterior surface of the epiglottis and the posterior wall of the pharynx. Almost immediately after the contraction of the mylo-hyoids the larynx is pulled upwards and forwards by the contraction of the thyro-hyoid muscle, and the elevation of the hyoid bone by the muscles which connect it to the lower jaw. The base of the tongue is simultaneously drawn backwards by the stylo- and palato-glossus. The lower or laryngeal portion of the epiglottis is thus caused to come into contact with the upper orifice of the larynx, occluding it completely, but the pharyngeal portion projects beyond the larynx, and takes no share in its closure (Eykman). The glottis is closed by the approximation of the vocal cords and the arytenoid cartilages. The epiglottis, however, is not absolutely indispensable for closing the larynx, since swallowing proceeds in the ordinary way when it is absent. The morsel of food, grasped by the middle and lower constrictors as it leaves the back of the tongue, passes rapidly and safely over the closed larynx, the process being accelerated by the pulling up of the lower portion 324 DIGESTION of the pharynx over the bolus by the action of the palato- and stylo- pharyngei. The second or cesophageal portion of the involuntary stage is a more leisurely performance. The bolus is carried along by a peculiar ' peristaltic ' contraction of the muscular wall of the oesophagus, which travels down as a wave, constricting the tube and pushing the food before it. In front of the constricting wave moves a wave of inhibition, so that the part of the oesophagus into which the bolus is about to pass is always relaxed, while the part behind it is contracted. This exact co-ordination of inhibition and contraction is the essential thing in peristalsis. When the food reaches the lower end of the gullet the tonic contraction of that part of the tube is for a moment relaxed by reflex inhibition, and the morsel passes into the stomach. Beaumont saw, in the case of St. Martin, that the oesophageal orifice of the stomach contracted firmly after each morsel was swallowed, and so did the gastric walls in the neighbourhood of the fistula when food was introduced by this opening. In the dog the whole process of swallowing from mouth to stomach has been shown to occupy four to five seconds, but the time is by no means constant. In man the peristaltic wave requires about five to six seconds to travel from the level of the glottis to the cardiac orifice. The rate of movement is greater in the upper than in the lower portion of the oesophagus. Such is the mechanism of deglutition when the bolus is of such consistence and size that it actually distends the oesophagus. But it has been shown that liquid food is swallowed in a different way. The food lying on the dorsum of the tongue, suddenly put under pressure by the sharp contraction of the mylo-hyoid muscles, is shot rapidly down to the lower part of the lax oesophagus, or, occa- sionally, some of it even into the stomach. So far the process has only occupied one-tenth of a second. After several seconds, the food, or the portion which still remains in the oesophagus, is forced through the cardiac sphincter into the stomach by the arrival of the tardy peristaltic contraction of the cesophageal wall (Kronecker and Meltzer). Two sounds may be heard in man on listening in the region of the stomach or oesophagus during deglutition of liquids, especially when, as generally happens, they are mixed with air. The first sound occurs at once, and is due to the sudden squirt of the liquid along the gullet ; the second, which is heard after a distinct interval (about six seconds), is caused by the forcing of the fluid through the cardiac orifice of the stomach by the contraction of the oesophagus. There are certain peculiarities which distinguish this peristaltic movement of the oesophagus from that of other parts of the alimen- tary canal. It is far more closely related to the central nervous system, and, unlike the peristaltic contraction of the intestine, can THE MECHANICAL PHENOMENA OF DIGESTION 3*5 pass over any muscular block caused by ligature, section, or crush- ing, so long as the nervous connections are intact. But division of the cesophageal nerves causes, as a rule, stoppage of oesophageal movements; although an excised portion of the tube retains its vitality for a long time, and may, under certain circumstances, go on contracting in the characteristic way after removal from the body (p. 8i>). Stimulation of the mucous membrane of the pharynx will cause reflex movements of the oesophagus, while stimulation of its own mucous membrane is ineffective. From these facts we learn that although the cesophageal wall may possess a feeble power of spontaneous peristaltic contraction, yet this is usually in abeyance, or at least overmastered by central nervous control ; so that impulses discharged as a ' fusillade ' from successive portions of the vagus centre, and travelling down the oesophageal nerves, excite the muscular fibres in regular order from the upper to the lower end of the tube. Nervous Mechanism of Deglutition. — 'The centre for the whole involuntary stage (both pharyngeal and oesophageal) lies in the upper part of the medulla oblongata. When the brain is sliced away above the medulla, deglutition is not affected ; but if the upper part of the medulla is removed, the power of swallowing is abolished. In man, disease of the spinal bulb interferes far more with deglutition than disease of the brain proper. Normally, the afferent impulses to the centre are set up by the contact of food or saliva with the mucous membrane of the posterior part of the tongue, the soft palate and the fauces, the nerve- channels being the superior laryngeal, the pharyngeal branches of the vagus, and the palatal branches of the fifth nerve.* A feather has sometimes been swallowed involuntarily by a reflex movement of deglutition set up while the soft palate or pharynx was being tickled to produce vomiting. Artificial stimulation of the central end of the superior laryngeal will cause the movements of deglutition independently of the presence of food or liquid; but if the central end of the glosso- pharyngeal nerve be stimulated at the same time, the movements do not occur. The glosso-pharyngeal is therefore able to inhibit the deglutition centre, and it is owing to the action of this nerve that in a series of efforts at swallowing, repeated within less than a certain short interval (about a second), only the last is successful. It is also through the glosso-pharyngeal nerve that the respiratory movements are inhibited during deglutition. When the central end of this nerve is stimulated, respiration is stopped * It appears that the most influential reflex paths may differ in different animals. In the rabbit, e.g., the reflex is set up by excitation of the trigeminal fibres which supply the mucous membrane anterior to the tonsils, in the dog and cat by excitation of the glosso-pharyngeal fibres in the posterior wall of the pharynx, and in monkeys by excitation of the trigeminal branches dis- tributed to the mucous membrane over the tonsils (Kahn). 326 DIGESTION for four or five seconds, and this cessation is distinguished from that produced by any other afferent nerve by the circumstance that it occurs not in expiration exclusively or in inspiration ex- clusively, but with the respiratory muscles in the precise degree of contraction in which they happened to be at the moment of stimu- lation. The efferent nerves of the reflex act of deglutition are the hypoglossal to the tongue and the thyro-hyoid and other muscles concerned in raising the larynx; the glosso-pharyngeal, vagus, facial and fifth to the muscles of the palate, fauces, and pharynx; the fifth to the mylo-hyoid; and the vagus to the larynx and oesophagus. Section of the vagus interferes with the passage of food along the oesophagus; stimulation of its peripheral end causes cesophageal movements. Movements of the Stomach. — The whole of the stomach does not take part equally in the movements associated with digestion. We may divide the organ, both anatomically and functionally, into two portions — a pyloric portion, or antrum pylori, comprising about a fifth of the stomach, and a larger cardiac portion, or fundus* At the junction of the antrum and the fundus the circular muscular coat is slightly thickened into a ring called the ' transverse band/ or ' sphincter of the antrum.' In the living stomach the region of the transverse band is usually contracted so strongly and con- tinuously that a distinct groove is seen to separate the tubular antrum from the bag-like cardiac end. The suggestion of a massive constricting ring of muscle is belied by an examination of the dead viscus. The transverse band is really little more than a physio- logical sphincter. The empty stomach is contracted and at rest. A few minutes after food is taken contractions begin in the antrum, and run on in constricting undulations (in the cat at the rate of six in the minute) towards the pyloric sphincter. Each wave takes about twenty seconds (in the cat) to pass from the middle of the stomach to the pylorus. Feeble at first, they become stronger and stronger as digestion proceeds, and gradually come to involve the portion of the fundus next the sphincter of the antrum, but their direction is always towards the pylorus, never, in normal diges- tion, away from it. The food is thus subjected to energetic churn- ing movements in the pyloric end of the stomach, and worked up thoroughly with the gastric juice. Kept in constant circulation, it gradually becomes reduced to a semi-liquid mass, the chyme, which is at intervals driven against the pylorus by strong and regular peristaltic contractions of the lower end of the stomach, * Here ' fundus ' is used in the sense in which it is generally employed in speaking of the stomach of the dog 01 cat as signifying the wnole of the organ with the exception of the antrum pyiori. By the fundus of the human stomach most writers mean only the cul-'Je-sac at the cardiac end; the portion inter- vening between it and the aucrum pylori is often termei the body of the ttomach. THE MECHANICAL PHENOMENA OF DIGESTION 327 II A.M the sphincter relaxing from time to time by a reflex inhibition to admit the better-digested portions into the duodenum, but tighten- ing more stubbornly at the impact of a hard and undigested morsel. The nature, as well as the consistence of the food, influences the length of its sojourn in the stomach. Carbo-hydrate food passes more rapidly through the pylorus than fatty food, and fat more rapidly than protein. The reason is that the acidity of the gastric juice varies with the different kinds of food, hydro- chloric acid being secreted in abundance in the presence of proteins, and to a much smaller extent in the presence of fats and carbo-hydrates. Now, dilute hydrochloric acid when introduced into the stomach remains there for a much longer time than water. This depends upon the fact that such portions of the acid as get into the duodenum stimulate afferent fibres in its mucous membrane, and so cause reflex spasm of the pyloric sphincter. When the acid chyme be- comes neutralized to a certain point by the bile and pancreatic juice, inhibitory impulses pass up from the duodenum and cause the sphincter to relax. The cardiac division of the stomach, with the exception of the portion that borders the transverse band, takes no share in the peristaltic movements. And, indeed, it is far more difficult to cause such con- tractions by artificial stimulation in the fundus* than in the pylorus. The two portions of the stomach are partially, or in certain animals from time to time completely, cut off from each other by the contraction of the sphincter of the antrum. The fundus, so far as its mechanical functions are concerned, acts chiefly as a reservoir for the food, which, like a hopper, it gradually passes into the pyloric mill as digestion goes on by a tonic contraction of its walls. The existence of this reservoir enables larger quantities of food to be taken at one meal, which can then be digested gradually. * The common idea, however, that the fundus is completely inactive is erroneous. Rhythmical variations in tone with a period of i to 3 minutes have been observed during digestion. In hunger vigorous peristaltic contractions sweep over the whole stomach, beginning at the cardiac end. Fig. 150. — Cat's Stomach seen by Rontgen Rays (Cannon). The outlines of the stomach containing food mixed with bismuth subnitrate were drawn at intervals from ii a.m. to 4.30 p.m. DIGESTION These facts have been mainly ascertained by v observations on animals, such as the dog and the cat, either by direct inspection after opening the abdomen (Rossbach), or in the intact body, under absolutely physiological conditions, by means of the Rontgen rays (Cannon). In the latter method the food is mixed with subnitrate of bismuth, which is opaque to these rays, so that when the animal is looked at through a fluorescent screen the stomach appears as a dark shadow in the field (Fig. 150). This method has even been applied with success to the study of the passage of the food along the human alimentary canal from deglutition to defaecation (Hertz). It has been shown in this way that in the living body in the erect position the long axis of the stomach is much more nearly vertical than had been supposed. When food is taken it sinks into the lower (pyloric) end, and at the upper end gas collects. When the person lies down the lower end of the stomach passes more towards the left, so that the long axis lies more transversely. Other methods have thrown light on the gastric movements — e.g., direct inspection through a fistula of the stomach, and the study Fig. 151. — Human Stomach studied by Roatgen Rays, a. Empty stomach in ver- tical position ; b, shortly after a meal (peristaltic contractions are occurring at the pyloric end) ; c, full stomach in vertical position. (Halliburton after Hertz.) of records showing the changes of pressure in the viscus obtained by means of small balloons introduced into it. Such balloons attached to a rubber tube have been swallowed by normal men and kept for long periods in the stomach (Carlson). Even in the excised stomach, kept in salt solution at the body-temperature, the typical movements can be observed proceeding for some time. Movements of the Small Intestine. — In the small intestine two kinds of movements are to be seen : (i) Gentle, swaying, ' pendulum ' movements, sometimes irregular, but often recurring rhythmically at the rate (in the dog) of 10 or 12 in the minute. Both the longi- tudinal and the circular muscular coats contract, causing slight waves of constriction, which may originate at any part of the gut, but, under normal circumstances, nearly always travel from above downwards, with a velocity of 2 to 5 centimetres per second. These movements cause the coils of the intestine to sway gently from side THE MECHANICAL PHENOMENA OF DIGESTION 329 to side. Under abnormal conditions, as in the exposed ' surviving ' intestines of the rabbit, contractions, probably similar to the pendulum movements, but running indifferently in both direc- tions, can be set up by local stimulation. The function of these pendulum movements seems to be the thorough mixing of the food with the digestive juices in the intestine. When an animal is fed with food containing bismuth subnitrate and observed with the Rontgen rays, it is seen that the food in a coil is often divided into small segments, which then join together to form longer masses, these being in turn again divided. This segmentation is rhyth- mically repeated (in the cat at the rate of thirty times a minute). Although of itself it insures only the mixing of the contents of the gut, and not their onward progress, it is usually accompanied by peristalsis, so that while the food is undergoing segmentation it is also slowly passing down the intestine. Often, however, a column of food remains for a considerable time, dividing, uniting, and divid- ing again, without sensibly shifting its position. In addition to the relatively rapid pendulum movements, much slower periodic varia- tions of tone of the whole musculature may be normally observed. (2) True peristaltic movements, in which a ring of constriction, obliterating the lumen, moves slowly down the tube, with a speed, it may be, no greater than i mm. per second. The portion of the intestine immediately below the advancing constriction is relaxed and motionless, so that we may say that a wave of inhibition pre- cedes the wave of contraction. The peristaltic movements of the small intestine, the most typical of their kind, are most easily excited by mechanical stimulation of the mucous membrane, as by the contact of a morsel of food or an artificial bolus of cotton- wool. Travelling, under normal conditions, always downwards, the constriction squeezes the contents of the tube before it, and the wave usually ends at the ileo-csecal valve, which separates the small intestine from the large. The cause of the definite direction of the peristaltic wave is grounded in the anatomical relations of the intestinal wall. For when a portion of the intestine is resected, turned round in its place and sutured, so that what was before its upper is now its lower end, the contraction wave is unable to pass, and the obstruction to the onward flow of the intestinal contents causes marked dilatation of the gut, and sometimes serious disturb- ance of nutrition. The most probable explanation is that the peri- stalsis is governed by a local reflex nervous mechanism (Auerbach's plexus), the stimulation of which by the contact of the food with the mucous membrane or by the distension of the gut "causes excitation of the circular muscular fibres above the point of stimula- tion, and inhibition of them below it. The automatic pendulum movements, and also the slow, rhythmical variations of tone, have a different relation to the local nervous mechanism, for they behave differently to poisons like cocaine and nicotine, which act on that 33<> DIGESTION mechanism. The pendulum movements are, if anything, increased in intensity and made more regular. But the peristaltic waves, although they can be locally excited by direct stimulation of the muscular fibres, are no longer propagated, and a bolus introduced into the intestine remains at rest where it is placed. Some have interpreted these facts as indicating that the pendulum movements are myogenic in origin. But evidence has lately been obtained that, although they are not reflex movements elicited by afferent impulses from the mucous membrane, since they continue in unaltered in- tensity, in isolated loops of intestine immersed in Ringer's (or Locke's) solution (p. 66) after removal of both mucosa and sub- mucosa, they are nevertheless dependent upon Auerbach's plexus. For when the circular muscular coat is separated from this plexus, the automatic movements of this coat are abolished, although the excitability of the musculature to direct stimulation is not affected. The longitudinal coat, which is still in connection with Auerbach's plexus, goes on contracting spon- taneously (Magnus). Under certain conditions a movement of food or secretions in the reverse of the normal direction can be set up in the small intestine in the intact body — e.g., in the case of obstruc- Fig. 152.— Intestine Segment beating tion of the intestine leading to in Ringer's Solution. At 6 the oxy- vomiting of its Contents. But this gen stream was increased. To be does not necessarily indicate a re- read from left to right. Time trace, , , ,, J , .. ,. . half-minutes. (Reduced to one -half.) versal of the normal direction of the peristalsis. Such a reversal, if it occurs at all, is not easy to realize by artificial stimulation, and even when an antiperistaltic wave is apparently started, it travels up the intestine only for a short distance and then dies out. A third variety of intestinal movement has sometimes been described, the so-called ' peristaltic rush ' (Meltzer, etc.). It consists of a rapidly moving peristaltic contraction, preceded by relaxation of a long portion of the tube. Such a contraction may even sweep down without pause from the duodenum to the end of the ileum. The Movements of the Large Intestine differ from those of the small mainly in the great frequency of antiperistalsis. This, indeed, seems to be the usual movement of the transverse and ascending colon. The antiperistalsis recurs in periods about every fifteen minutes, and each period generally lasts about five minutes. The constrictions, running towards the caecum, thoroughly churn and mix the remnants of the food, a considerable absorption of which may take place in the upper part of the large intestine. Regurgita- tion into the ileum in man is prevented partly by the oblique entry THE MECHANICAL PHENOMENA OF DIGESTION 331 of the ileum through the wall of the colon (so-called ileo-caecal valve), but essentially by the tonic contraction of the ileo-colic sphincter. The sphincter usually permits the passage of material only in the direction from small to large intestine. But as an occasional phenomenon, a reverse movement may occur. Thus food may actually pass back through the ileo-colic sphincter into the small intestine under the action of a long-continued and vigorous antiperistalsis, and in this way a considerable portion of a bulky enema may be eventually disposed of (Cannon). This so-called antiperistalsis is not precisely the same kind of movement, leaving out of account its direction, as the peristalsis already described in the small intestine, since it is not preceded by a wave of inhibition. True peristaltic contractions preceded by relaxation of the gut may also be observed to start in the caecum, and to travel down the large intestine. They are not very frequent in comparison with those of the small intestine, and they die away before reaching the end of the colon, allowing the food to be driven back again towards the caecum by the antiperistalsis. A true downward peristalsis is more commonly seen in the descending colon, and is here asso- ciated with the propulsion and collection of the faeces, which are mainly stored in the sigmoid flexure. These peristaltic contractions do not normally reach the rectum, which, except during defaecation, remains at rest. Influence of the Central Nervous System on the Gastro- Intestinal Movements. — As we have already said, these movements are much less closely dependent on the central nervous system than are those of the oesophagus. They can not only go on, but are in general better marked when the extrinsic nervous connections are cut; they cannot spread when the continuity of the tube is destroyed, and the mere presence of food will excite them when other than local reflex action has been excluded by section of the nerves. Never- theless, the central nervous system does exercise some influence in the way of regulation and control, if not in the way of direct initiation of the movements, and the swallowing or even the smell of food has been observed to strengthen the contractions of a loop of intestine severed from the rest, but with its nerves still intact. The vagus is the efferent channel of this reflex action : stimulation of its peripheral end may cause movements of all parts of the alimentary canal from cesophagus to large intestine, and may strengthen movements already going on; but section of it does not stop them, nor hinder the food from causing peristalsis wherever it comes. The vagus also contains inhibitory fibres for the lower end of the cesophagus and the whole of the stomach. Stimulation of it is followed first by inhibition, and then, after an interval, by an increase of tone and augmentation of the contraction of the whole stomach, including the cardiac and pyloric sphincters. The 332 DIGESTION splanchnic nerves contain fibres by which the intestinal movements can be inhibited, and they appear to be always in action, for after section of these nerves the movements are strengthened. On the other hand, stimulation of the peripheral end of the cut splanchnic causes arrest of the movements. Occasionally, however, it has the opposite effect. Contractions of the small intestine are more easily caused by excitation of the vagus after the inhibitory splanch- nic nerves have been cut. The splanchnics also contain inhibitory fibres for the stomach, and it is only when these are intact that complete reflex inhibition of the organ can be obtained in the rabbit (Auer). The gastric movements are not permanently affected by section of these nerves alone, or even by simultaneous section of the splanchnics and the gastric branches of the vagi. But if the vagi are cut while the splanchnics remain intact, the peristalsis of the stomach is weakened, its onset delayed, and the proper emptying of the viscus through the pylorus interfered with. In all probability these results are due to the uncontrolled action of the inhibitory fibres. The splanchnics have a special relation to the ileo-colic sphincter, which closes when they are stimulated, and becomes in- sufficient when they are cut. The vagus does not affect it. The lower part of the large intestine is influenced by the sacral nerves (second, third, and fourth sacral in tlie rabbit), and by certain lumbar nerves, in the same way as the higher parts of the alimentary canal, and particularly the small intestine, are influenced by the vagus and the splanchnics. Stimulation of these sacral nerves within the spinal canal, or of the pelvic nerves (nervi erigentes) into which they pass, causes contraction of the parts of the large intestine concerned in defaecation — that is, in the dog, of the whole colon, with the exception of the caecum; in the cat, of the distal two-thirds of the colon. The colon first undergoes rapid shortening due to the contraction of the longitudinal fibres and the recto-coccygeus muscle. After a few seconds this is followed by contraction of the circular fibres, beginning at the lower limit of the region in which antiperistalsis can occur, and spreading downwards, so as to empty the portion of the bowel involved in the contraction. This is a very close imitation of what occurs in natural defaecation. In man the parts involved in these movements are probably the sigmoid flexure and rectum. In addition to these characteristic motor effects on the lower part of the large intestine, stimulation of the pelvic nerves causes an increase in the antiperistalsis of its upper portions. Stimulation of the lumbar nerves or of the por- tions of the sympathetic into which their visceral fibres pass (lumbar sympathetic chain from second to sixth ganglia, or the rami from it to the inferior mesenteric ganglia) causes inhibition of the movements of the caecum and the whole colon, including the antiperistaltic move- ments. Excitation of the sacral nerves initiates or increases the con- traction of both coats of the portions of the large intestine on which they act, excitation of the lumbar nerves inhibits both. And in the small intestine the same law holds good ; the two coats are contracted together by the action of the vagus or inhibited together by that of the splanchnics. Defaecation is partly a voluntary and partly a reflex act. But in the infant the voluntary control has not yet been developed; THE MECHANICAL PHENOMENA OF DIGESTION 333 in the adult it may be lost by disease; in an animal it may be abolished by operation, and in each case the action becomes wholly reflex. Owing to the tonic contraction of the rectum and the acute angle formed at the pelvi-rectal flexure, the faeces are arrested at this point. In consequence the pelvic colon becomes filled with faeces from below upwards, and the rectum remains empty till immediately before defalcation. This has been verified in man by observations with the Rontgen rays (Hertz). In persons whose bowels are opened regularly after breakfast, the passage of faeces into the rectum gives rise to the characteristic sensation which may be termed the ' call to defaecation. ' It is the distension of the rectum, and of the rectum alone, which is associated with this sensation, for in persons from whom the entire rectum has been removed for malignant disease the sensation is absent, and it may be elicited by artificially distending the rectum, though not any other part of the alimentary canal. The minimum pressure required to elicit the sensation is smaller the greater the length of the gut exposed to it, varying in one individual from 32 to 48 mm. of mercury, according to the length of a balloon introduced into the rectum. The passage of the faeces from the pelvic colon into the rectum is due to the discharge of that reflex contraction of the lower portion of the bowel already described (p. 332), of which the pelvic nerves constitute the efferent path. This reflex peristalsis is elicited by various causes, among which one of the most important is the taking of food at breakfast into the empty stomach, and another the muscular activity associated with getting up and dressing. The desire to defalcate may for a time be resisted by the will, or it may be yielded to. In the latter case the abdominal muscles, and, according to Hertz, the diaphragm also, are forcibly contracted; and the glottis being closed, the whole effect of their contraction is expended in raising the pressure within the abdomen and pelvis, and so aiding the muscular wall of the bowel itself in driving the faeces from the sigmoid flexure to the rectum. The two sphincters which close the anus — the internal sphincter of smooth muscle, and the external of striated — are now relaxed by the inhibition of a centre in the lumbar portion of the spinal cord, through the activity of which the tonic contraction of the sphincters is normally maintained. This relaxation is partly voluntary, the impulses that come from the brain acting probably through the medium of the lumbar centre. But in the dog, after section of the cord in the dorsal region, the whole act of defaecation, including contraction of the abdominal muscles and relaxation of the sphincters, still takes place, and here the process must be purely reflex. Even after complete destruction of the lumbar and sacral portions of the spinal cord the tone of the sphincters returns after a time, and defalcation is carried on as in a normal animal, the control of the sphincters 334 DIGESTION being due either to a property of the muscular tissue itself or to local ganglia. The contraction of the levatores ani helps to resist overdistension of the pelvic floor and to pull the anus up over the faeces as they escape. The nervi erigentes carry efferent constrictor fibres, and the hypogastrics, as a rule, efferent dilator fibres, to the sphincters. While the internal sphincter is by itself capable of maintaining a tonus of considerable strength, the external sphincter contributes an important share (30 to 60 per cent.) to the closure of the rectum. If the call to defaecation is neglected, the desire passes away. This is not due to the faeces being carried back into the pelvic colon by antiperistalsis, as has generally been stated. The faeces which have passed into the rectum remain there, as can be shown by examination with the finger after the desire to empty the bowels has disappeared. The reason for the disappearance of the sensation is the relaxation of tone which occurs in the muscular coat of the rectum after a period of distension. It is not till it has been again distended by the entrance of a further portion of faeces that the call to defaecation is again experienced. When the call is repeatedly neglected, the sensibility of the rectum to dis- tension becomes blunted, and this is a common cause of constipation. The time of passage of substances through the alimentary canal has been studied by administering collodion capsules filled with subnitrate of bismuth to human beings, and observing their pro- gress by taking shadow pictures of them at intervals with the Rontgen rays. During the first twenty minutes two such capsules swallowed at the same time by a healthy young man were clearly seen in the greater curvature of the stomach, but in the interval between the first half-hour and the seventh or eighth hour no further trace of them was detected. About the eighth hour they re- appeared in the caecum, where they remained with little or no onward movement till the fourteenth hour. From the fourteenth to the sixteenth hour they travelled along the ascending colon, and tarried a long time at the left angle of the colon. From the nine- teenth to the twenty-second or twenty-fourth hour they slowly passed downward in the descending colon, and stopped at the sig- moid flexure till their expulsion in defaecation. In some subjects the entire passage of the capsules was complete in sixteen hours, in others not until after thirty hours. A one cent piece swallowed by a healthy child four years old was recovered in the fasces 52 hours later, and a button, slightly larger, swallowed by the same child, appeared after almost exactly the same interval. Vomiting. — We have seen that under normal conditions the movements of the alimentary canal always tend to carry the food in one definite direction, along the tube from the mouth to the rectum. The peristaltic waves generally run only in this direction, and, further, regurgitation is prevented at three points by the THE MECHANICAL PHENOMENA OF DIGESTION 335 cardiac and pyloric sphincters of the stomach and the ileo-colic sphincter and valve. But in certain circumstances the peristalsis may be reversed, one or more of the guarded orifices forced, and the onward stream of the intestinal contents turned back. In obstruc- tion of the bowel, the faecal contents of the large intestine may pass up beyond the ileo-caecal valve, and, reaching the stomach, be driven by an act of vomiting through the cardiac orifice; in what is called a ' bilious attack,' the contents of the duodenum may pass back through the pylorus and be ejected in a similar way; or, what is by far the most common case, the contents of the stomach alone may be expelled. Vomiting is usually preceded by a feeling of nausea and a rapid secretion of saliva, which perhaps serves, by means of the air carried down with it when swallowed, to dilate the cardiac orifice of the stomach, but may be a mere by-play of the reflex stimula- tion bringing about the act, since evidence of stimulation of other bulbar centres (vaso-motor and cardio-inhibitory) has also been obtained. The diaphragm is now forced down upon the ab- dominal viscera, first with open and then with closed glottis. The thoracic portion of the oesophagus is thus placed under diminished pressure, and therefore widened, while saliva and air are aspirated into it out of the mouth. The abdominal muscles strongly con- tract. At the same time the stomach itself, and particularly the antrum pylori, contracts, the cardiac orifice relaxes, and the gastric contents are shot up into the lax oesophagus, and through it into the pharynx, and issue by the mouth or nose. The move- ments of the stomach during vomiting induced by apomorphine have been studied in the cat by the Rontgen ray method. There is first observed extreme relaxation of the cardiac end; then a deep constriction appears a little below the cardiac orifice, and runs towards the pylorus, increasing in depth as it goes. When the transverse band is reached, this contracts firmly and remains con- tracted, and the constriction passes on over the antrum pylori. Ten or twelve similar waves follow, at the end of which time the constriction in the region of the transverse band divides the stomach into the firmly-contracted antrum and the relaxed fundus. Now follows a sudden contraction of the diaphragm and abdominal muscles accompanied by the opening of the cardiac orifice. Either the diaphragm and abdominal muscles alone, without the stomach, or the diaphragm and stomach together, without the abdominal muscles, can carry out the act of vomiting. For an animal whose stomach has been replaced by a bladder filled with water -can be made to vomit by the administration of an emetic (Magendie) ; and Hilton saw that a man who lived fourteen years after an injury to the spinal cord at the height of the sixth cervical nerve, which caused complete paralysis below that level, could vomit, though with great difficulty. In a young child in which very slight causes 336 DIGESTION will induce vomiting, the stomach alone contracts during the act. But in the adult such a contraction is ineffectual, and the same la the case in animals, for a dog under the influence of a moderate dose of curara, which paralyzes the voluntary muscles but not the stomach, cannot vomit. The nerve-centre is in the medulla oblongata. It may be excited by many afferent channels : the sensory nerves of the fauces or pharynx, of the stomach or intestines (as in strangulated hernia) , of the liver or kidney (as in cases of gall-stone or renal calculi), of the uterus or ovary, and of the brain (as in cerebral tumour), are all capable, when irritated, of causing vomiting by impulses passing along them to the vomiting centre. The vagus nerve in man certainly contains afferent fibres by the stimulation of which this centre can be excited, for it has been noticed that when the vagus was exposed in the neck in the course of an operation, the patient vomited whenever the nerve was touched (Boinet, quoted by Gowers). In meningitis, vomiting is often a prominent symptom, and is sometimes due to irritation of the vagus nerve by the inflammatory process. Some drugs act as emetics by irritating surfaces in which efficient afferent impulses may be set up, the gastric mucous membrane, for example; sulphate of zinc and sulphate of copper act mainly in this way. Apomorphine, on the other hand, stimulates the centre directly, and this is also the mode in which vomiting is pro- duced in certain diseases of the medulla oblongata. The efferent nerves for the diaphragm are the phrenics, for the abdominal muscles the intercostals. The impulses which cause contraction of the stomach pass along the vagi. Dilatation of the cardiac orifice is brought about by the inhibitory fibres in the vagus already mentioned. SECTION III. — THE CHEMISTRY OF THE T IGESTIVE JUICES. Ferments. — The chemical changes wrought in the food as it passes along the alimentary canal are due to the secretions of various glands which line its cavities or pour their juices into it through special ducts. These secretions owe their power for the most part to substances present in them in very small amount, but which, nevertheless, act with extraordinary energy upon the various constituents of the food, causing profound changes with- out, upon the whole, being themselves used up, or their digestive power affected. The active agents are the enzymes, sometimes spoken of as unformed or unorganized ferments — unorganized because their action does not depend upon the growth of living cells, which was long supposed to be the case for some other fer- ments, such as yeast. Since it has been shown that specific enzymes can be separated from cells which were formerly believed to act 33? by their mere growth, the distinction between formed and unformed ferments has lost its significance, and has to a great extent been superseded by the distinction between intra- and extra-cellular enzymes (also called endo- and exo-enzymes) — i.e., between ferments which normally act in the interior of the cells where they are produced and ferments which act outside of the cells that secrete them. From yeast cultures, for instance, by crushing the cells, a ferment (zymase*) can be obtained which in the complete absence of living yeast-cells, and, indeed, of any living micro-organism, forms alcohol and carbon dioxide from sugar, just as living yeast does. There is every reason to believe that it is by the intracellular action of this endoenzyme that the yeast-cell normally causes alcoholic fermentation. The digestive ferments are typical extracellular enzymes. Their chemical nature has not been exactly made out; some of them at least do not appear to be proteins, or to contain a protein group. Many of them apparently exist in the colloidal condition, although this has not been shown for all. In certain cases the more or less stable union of a definite inorganic substance with the ferment, or its actual inclusion in the ferment molecule, seems to be a condition of its action. Thus there is reason to believe that in gastric digestion hydrochloric acid is loosely combined with the pepsin. In the plant oxydase, laccase (p. 272), manganese is present. And the fact that manganese salts oxidize certain substances as laccase does suggests that it is the manganese in combination with some protein or other organic compound in the ferment molecule which confers upon laccase its oxidizing power. A similar relation between iron and some animal oxydases is possible, though not definitely proved. But none of the ferments of the digestive juices has as yet been satisfactorily isolated, and at present it is only by their effects that we recognize them. The difficulty of isolating them is increased by the fact that, like other colloids, they readily adhere to surfaces, and are carried down by the most diverse precipitates of substances to which they are chemically indifferent. On the other hand, this very property is taken advan- tage of to procure more concentrated, although still impure, solutions of them than exist in the natural secretions. Thus in the prepara- tion of many ferments the first step is to produce an inert pre- cipitate, such as calcium phosphate, in the juice or extract. Some of the ferments act best in an alkaline, some in an acid medium. They all agree in having an ' optimum ' temperature, which is more favourable to their action than any other ; a low temperature sus- pends their activity, and boiling abolishes it for ever. The optimum temperatures of the majority of enzymes lie between * Ferments are usually designated by names with the termination ' ase,' and indicating the kind of substances on which they act, or sometimes their source. Thus proteases are ferments acting on proteins, amylases ferments acting on starch, etc. 3?* DIGESTION 37° and 53° C. ; the ' killing ' temperatures between 60° and 75* C. when they are heated in solutions, but considerably higher when they are heated dry. The action of the digestive enzymes is hydrolytic — i.e., it is accompanied with the taking up of the elements of water by the substance acted upon. The accumulation of the products of the action first checks and then arrests it. In many cases this seems to be due to combination of the ferment with one or other of the end products, and the consequent segregation of the ferment from the reaction mixture. The enzyme is not affected indiscriminately by any of the end products. On the contrary, their action is curiously selective. Thus the hydrolysis of lactose by lactase is retarded by galactose, but not by the other end product dextrose. The hydrolysis of cane-sugar by invertase is retarded by levulose, but not by dextrose. The splitting of the dipeptid (p. 2) glycyl-/-tyrosin by a ferment in the expressed juice of yeast-cells is greatly delayed by one of the products (/-tyrosin), but not by the other (glycocoll). Combination of the ferment with an end product is not, however, the only way in which the reaction may stop before the whole of the substrate, as the substance acted on by the ferment is termed, has been changed. It has been demonstrated in some cases that this is due to the action of the enzymes being reversible. For example, lipase (p. 363) not only decomposes the esters ethyl butyrate or glycerin butyrate, but also builds them up again from the decomposition products — ethyl butyrate from ethyl alcohol and butyric acid, glycerin butyrate from glycerin and butyric acid (Kastle and Loevenhart, Hanriot). Thus: C3H7COOC2H5 +H2O^=?'C3H7COOH + C2H6OH. Kthyl butyrate. Water. Butyric acid. Ethyl alcohol. The action of the enzyme is merely to accelerate the establish- ment of the proportions in which the four bodies entering into the reaction are in equilibrium, and the point of equilibrium is the same whether we start from one or the other side of the equation repre- senting the reaction. Such reversible reactions in the presence of enzymes seem to afford the key to the explanation of many of the syntheses which are known to occur in the body. Sometimes the action is not strictly reversible in the sense that precisely the original material is reconstructed, but from the products of the hydrolysis substances are synthesized or condensed, which are then incapable of being split by the ferment. When a concentrated solution of dextrose is acted on for a long time by yeast maltase, a ferment obtained from yeast which changes maltose into dextrose, some of the dextrose is reconverted into isomaltose and dextrin-like bodies. Isomaltose is not again hydrolysed by maltase. The ferment emulsin contained in almonds behaves in the converse way. It hydrolyses isomaltose so as to form dextrose, and then condenses dextrose to maltose (Armstrong). THE CHEMISTRY OF THE DIGESTIVE /VICES 339 Many of the ordinary substances of the laboratory will accelerate reaction which goes on slowly in their absence. These are called catalysers. Some writers also speak of catalysers which retard a reaction progressing quickly in their absence. The process by which the reaction is accelerated (or retarded) is termed catalysis. A typical catalyser can exert its action when it is present in ex- ceedingly small amount in comparison with the substance acted upon. However it may enter into the reaction, it does not take part in the formation of the final products nor contribute to the energy changes, and for this reason is often apparently unaltered at the end of the process. The catalysers have therefore been compared to the lubricants used for machinery as contrasted with the coal or other source of energy. If it be remembered that the expression is a purely metaphorical one, we may say that the catalyst oils the reaction so that it slips on smoothly and swiftly to an end-point which would, however, have been reached just the same in time. A classical instance of catalysis is the inversion of cane-sugar by weak acids, i.e., the change of the cane-sugar into a mixture of equal quantities of dextrose and levulose — a reaction which may be represented by the equation = C6H12O6 + C6H12O6. Cane-sugar. Water. Dextrose. Levulose. This is a reaction which occurs also when the sugar is simply dis- solved in water, but with extreme slowness at the ordinary tempera- ture, although more rapidly at 100° C. The effect of the acid is to catalyse the reaction, to markedly accelerate it. The hydrogen ions of the free acid are responsible for the catalysis, and they are not used up in the process, for the reaction at the end is unaltered. The same action upon cane-sugar is exerted by an enzyme, invertase, found in intestinal juice, although the laws governing the reaction are somewhat different. Reversibility of the reaction can be even more clearly demonstrated for catalysers than for enzymes. For example, the condensation of acetone to diacetone-alcohol, which is accelerated by hydroxyl ions (as by the addition of sodium hydroxide, ammonia, etc.), only proceeds to a certain point, at which equilibrium is established between the proportions of acetone and the condensation product. Henceforth as much of the latter is decomposed as is condensed. Thus : 2CH3.CO.CH3^=H>CH8.cb.CH2.C(CH3)2OH. Acetone. Acetone alcohol. On the other hand, the final equilibrium point need not be the same for a catalyser and an enzyme. For example, amyl butyrate is formed and decomposed according to the equation C6HUOH +C3H7COOH«=II?' CgHyCOO.CgHn +H2O. Amyl alcohol. Butyric acid. Amyl butyrate. Water. $4« DIGESTION The reaction can be accelerated either by a catalyst — e.g., H ions— as by addition of free hydrochloric or picric acid, or by pancreas lipase. When the concentrations of the reacting substances are appropriately chosen, the same equilibrium point will be reached from either side of the equation — i.e., the same percentage of the butyric acid will be converted into the ester if we start with the alcohol and acid, as will remain combined as ester if we start with the amyl butyrate. But the proportion will not be the same when the reaction is acclerated by H + as when it is accelerated by the enzyme. And although it is probable that there is no fundamental difference between the action of the digestive enzymes and that of the inorganic catalysers, it is much too early to dogmatize. Not even the m .rkedly specific action of the digestive ferments can be considered an essential distinction. It is true that invertase will act upon dextrose, and not at all upon maltose or lactase. But there are other sugars, e.g., raffinose, a trisaccharide with the formula QgHggOjg, obtained from beet-sugar residues, which it will hydrolyse. Raffinose is made up of one molecule each of dextrose, levulose, and galactose. On heating with dilute acids, it is decom- posed into these substances. Invertase, however, only splits off the levulose molecule, leaving a disaccharide isomeric, but not identical with lactose. Similarly lactase, which is without action upon cane-sugar or maltose, will hydrolyse the /3-galactosides, and maltase, inert as regards cane-sugar or lactose, will hydrolyse the a-glucosides. On the other hand, emulsin decomposes the /3-gluco- sides, to which group most of the natural glucosides belong, as well as the /3-galactosides and lactose. From raffinose emulsin splits off galactose, leaving cane-sugar. Since the a and ft compounds are isomeric, and differ not in their composition but in their struc- ture, it has been concluded that the structure of the molecule of a substance must be related to the structure of the enzyme which can act on it, in some such way as a lock is related to its proper key. Thus the key lactase fits in the lock lactose, but not in the lock dextrose or the lock maltose. Although the same specificity is not to be observed in the action of catalysers as in the action of enzymes, it is not difficult to find many instances in which inorganic substances show a marked limitation of their catalytic effects to particular reactions. Thus hydriodic acid is slowly oxidized in presence of hydrogen peroxide, with formation of iodine and water. This reaction is accelerated by the addition of many substances, e.g., tungstic acid. But tungstic acid has no catalytic effect on the oxidation of hydriodic acid by bromic acid. The existence of an optimum temperature for ferment action, above which it rapidly decreases, and eventually comes to a com- plete stop, is also in all probability only a superficial distinction between enzymes and catalysers. For enzymes are easily altered, THE CHEMISTRY OF THE DIGESTIVE JUICES 341 or even destroyed, at temperatures which very likely would favour their action were they as thermostable as the majority of catalysing agents. And inorganic catalysts are known which also show the phenomenon of an optimum temperature depending on changes produced in their physical condition when the temperature is raised above this point. Thus a colloidal solution (or ' sol/ as it is called) of platinum, prepared by passing electric sparks between two platinum electrodes immersed in distilled water, and containing the metal in the form of ultra-microscopic particles, acts as a cata- lyser of a number of reactions. As the temperature is increased up to a certain ' optimum,' the velocity of the catalysed reaction is increased. But beyond this, as the boiling-point is approached, the colloidal platinum is precipitated, and ceases to influence the reaction. As to the manner in which an enzyme increases the velocity of its appropriate reaction, it is not easy to make any very positive statement. Several possibilities are recognized, of which two have been especially discussed, (i) The existence of the enzyme in colloidal solution may be important. It is characteristic of colloidal solutions, in which the dissolved substance is present in the form of extremely fine particles, that the total surface of the particles is very great in proportion to the mass of the substance in solution. Thus, a sphere of about the same volume as the eyeball, with a diameter of, say, 2 centimetres, would have a surface of 12*5 square centimetres. If this material were subdivided into spheres of about tbe same volume as a leucocyte, with a diameter of, say, 10 p, it would form eight thousand million of these spheres, with a total surface of over i\ square metres. If the small spheres were further subdivided into spherical particles, with a diameter only the thousandth part of that of a leucocyte, say -^-, each would form a thousand million of these particles, and the total surface of all the particles would be about 2,500 square metres. Now, it is known that the intensity of action of some of the inorganic catalysers is proportional to the surface exposed. For example, hydrogen peroxide, if left to itself, is slowly decomposed into water and oxygen. The addition of finely divided v platinum, in the form of platinum black, greatly hastens the decomposition, and the oxygen bubbles off. The colloidal platinum sol is still more effective. The nature of the surface effect is not entirely clear. One factor has been thought to be an increase in the concentration of dissolved substances or condensation of gases at the surface, and the better opportunity for mutual action thus afforded to the ferment and the substrate. The great extension of the surface cannot be the only factor in the catalysis; otherwise any fine powder or suspension would have a cata- lytic action. But kaolin, or fine sand, or colloidal solutions of ordinary proteins or gelatin, have little, if any, effect on the decomposition of hydrogen peroxide. (2) Enzymes may produce their effects by contributing to the for- mation of bodies intermediate between the substrate and the end- products. If the time required for the formation of a given quantity of the intermediate compound and the time required for the decom- position of this compound into the final products of the ferment action are in sum less than the time required for the direct change of the DIGESTION substrate into the end-products, the enzyme will clearly act as a cata- lyser of the reaction. It has been shown that in the case of certain inorganic catalysers this does occur. Thus, in the oxidation of hydriodic acid by hydrogen peroxide, which has been already referred to, molybdic acid has the power of acting as a catalyser. It has been proved that the reaction occurs in two stages, permolybdic acid being first formed by the action of the peroxide on molybdic acid. The permol- ybdic acid then acts on hydriodic acid, producing iodine and water, and being itself reduced again to molybdic acid, which therefore comes out at the end of the reaction unchanged. The velocity of the double reaction is much greater than that of the direct oxidation of hydriodic acid by hydrogen peroxide. There is evidence that the ferment actually combines with the sub- strate, the combination then breaking up to form the end-products. For instance, it has been shown that the amount of lactose hydrolysed by lactase in a given time, when the ferment is present in very small quantity in comparison with the substrate, is proportional to the con- centration of the ferment, and independent of the concentration of the lactose. Also with a given small concentration of ferment the amount of lactose hydrolysed is at first the same for successive equal intervals of time. These facts can only be explained by the assump- tion that the ferment first combines with a portion of the substrate, the rest of which remains inactive as regards the reaction, and that this combination then takes up water and decomposes into the end- products, in this case dextrose and galactose, setting free the ferment to combine with another portion of the substrate. The Quantitative Estimation of Ferment Action. — Since we have as yet no certain method of freeing the digestive ferments from impurities, our only quantitative test is their digestive activity. And since a very small quantity of ferment can act upon a practically indefinite amount of material if allowed sufficient time, we can only make comparisons when the time of digestion and all other conditions are the same. If we find that a given quantity of one gastric extract, acting on a given weight of fibrin, dissolves it in half the time required by an equal amount of another gastric extract, or dissolves twice as much of it in a given time, we conclude that the digestive activity of the pepsin is twice as great in the first extract as in the second. But this does not permit us to say that the one contains twice as much pepsin as the other. For it has been found that the amount of digestion in a given time is not directly proportional to the quantity of ferment present, but to the square root of the quantity of ferment (Schutz's law). This law was deduced by Schiitz for pepsin, but is said to hold also for trypsin, steapsin, and ptyalin (Pawlow, Vernon). To determine the amount of proteolysis the nitrogen of the protein which has gone into solution may be estimated (p. 521). The following table shows the results of one experiment : Digested Nitrogen in Grammes. Pepsin Solution used inC.C. Found. Calculated. I O'O23O O-Q223 4 0*0427 0*0446 9 0-0686 0^0669 16 0-0889 0^0892 343 Or a piece of a glass capillary-tube filled with heat-coagulated egg-white may be cut off and placed in the digestive mixture (Mett's tubes). At the end of the period of digestion the length of the piece of tube and that of the undigested remnant of the column of coagulated protein are measured with a millimetre scale under a low- power microscope. The difference gives the length of the column digested. If i c.c. of gastric juice caused in a given time digestion of 2 mm. of the egg-white, 4 c.c. of the same juice would digest in the same time and under identical conditions about 4 mm., and 9 c.c. about 6 mm. As a test of the activity of a diastatic ferment, we take the amount of sugar formed in a given time in a given quantity of a standard starch solution. To determine the activity of a liquid, say, the pancreatic juice, as regards fat-splitting ferment, the acidity of the emulsion formed by the juice and fat after standing for a definite time at, a given temperature (with occasional shaking) can be estimated by titration with baryta solution. In addition to the ferments of the digestive juices which act extra- cellularly in the lumen of the alimentary canal, and those which do their work intracellularly in its walls, micro-organisms are present in the gut, and even in normal digestion contribute to the changes brought about in the food ; while under abnormal conditions they may awaken into troublesome, and even dangerous, activity. It is now known that these act by producing intracellular enzymes. It may be noted here, although the subject must be again referred to (p. 390), that specific substances capable of inhibiting the action of ferments exist. Some of these antiferments are normally present in the body — an antitrypsin, for instance, in normal blood-serum. Numerous antiferments may be artificially obtained by immunizing animals with the original ferments. Thus an antilipase is found in the serum of rabbits after injection of pancreatic lipase, and an antiemulsin after injection of emulsin. Injection of rennin causes the formation of antirennin, which can be demonstrated in the blood-serum and milk of the immunized animal. Besides the anti- ferments, bodies, sometimes spoken of as ' co-enzymes, ' are known which aid the action of certain enzymes, not in the general way in which, for instance, increase of temperature up to the optimum does, but in some quite special manner. Thus, as we shall see, bile salts greatly facilitate the fat-splitting action of lipase. This co-operation is not to be confounded with the activation of the proferment or zymogen which in many cases represents the inactive form of the enzyme, while it is still within the secreting cells. For, once activated, the fully formed enzyme cannot be made to revert to the zymogen stage. For example, the active trypsin of the pancreatic juice cannot be changed into inactive trypsinogen, whereas substances which simply co-operate or co-act with enzymes leave the latter unaltered when they are removed. Thus lipase does not preserve the increased activity conferred upon it by bile salts when the bile salts are again separated from the digestive mixture. It is now necessary to consider in detail the nature of the various? 344 DIGESTION juices yielded by the digestive glands, and the mechanism of their secretion. Since it is along the digestive tract that glandular action is seen on the greatest scale, this discussion will practically embrace the nature of secretion in general. And here it may be well to say that, although in describing digestion it is necessary to break it up into sections, a true view is only got when we look upon it as a single, though complex, process, one part of which fits into the other from beginning to end. It is, indeed, the business of the physiologist, wherever it is possible to insert a cannula into a duct and to drain off an unmixed secretion, to investigate the properties of each juice upan its own basis; but it must not be forgotten that in the body digestion is the joint result of the chemical work of five or six secretions, the greater number of which are actually mixed together in the alimentary canal, and of the mechanical work of the gastro- intestinal walls. Saliva. — The saliva of the mouth is a mixture of the secretions of three large glands on each side, and of many small ones. The large glands are the parotid, which opens by Stenson's duct opposite the second upper molar tooth; the submaxillary, which opens by Wharton's duct under the tongue ; and the sublingual, opening by a number of ducts near and into Wharton's. The small glands are scattered over the sides, floor, and roof of the mouth, and over the tongue. Two types of salivary glands, the serous or albuminous and the tmtcous, are distinguished by structural characters and by the nature of their secretion; and the distinction has been extended to other glands. The parotid of many, if not all, mammals is a purely serous gland; it secretes a watery juice with a general re- semblance in composition to dilute blood-serum. The submaxillary of the dog and cat is a typical mucous gland; its secretion is viscid, and contains mucin. The submaxillary gland of man is a mixed gland; mucous and serous alveoli, and even mucous and serous cells, are intermingled in it. The submaxillary of the rabbit is purely serous. The sublingual is, in general, a mixed gland, but with far more mucous than serous alveoli. Some of the small glands are serous, others mucous in type. The mixed saliva of man is a somewhat viscous, colourless liquid of low specific gravity (1002 to 1008, average about 1005), alkaline to litmus, acid to phenolphthalein, but when tested by the electrical method (p. 24) almost neutral. Besides water and salts, it contains mucin (entirely ^rom the submaxillary, the sublingual and the small mucous glands of the mouth), to which its viscidity is due, traces of serum-albumin and serum-globulin (chiefly from the parotid), and a ferment, which hydrolyses starch, and therefore belongs to the group of amlyases or diastases. It differs somewhat from the amylase of pancreatic juice. But the small differences THE CHEMISTRY OF THE DIGESTIVE JUICES 345 .a 42 • CO T? T3 ^ rt •^ 9 18 w o 0 rt -O M 0 § 1 § ^ & "QH'H d 'a d d ^ i £!T a! 1^1 1:1 A o 1-1 d f3 '•" d 5 co~ 'U S -2 -9 O 'd f^ u ro .3 co c3 T3 -^ y ri4^ t3O gw^d.CUCD-3 CU^03 ^oSc! 3-0 oco^itSWScS cn'PiH^^fi.art.SrirtJ Illllllt lilf!|llti| '-joPnc!^^^^ "3d.nsP;3»>!9 >> S -^ ™CcuCcpocU(i3 DC^I-i^rt^tyjH Vi; o SfLnPH^PQPlh Q<-B £>C/3il!>U <^P 1 O '-p I ' 60 .1-1 S ^ fi K 3 - CO 5 • S S Convertin :=H « _&* - ^ T3 '-S 'Sw-t2 3 "§ 3 &9 fi OT3,o3 J3 W>« ™-9 co Hcort 2'-3 J> So d-Ii-'c!^ W)^ e 11=11111 8*1 IHIHII 4«2 2rt5^S ^23 Ef^-as^^^ c^f^ p^H-lo^^i Odi^ ^ S iU^'o^ T1 *T! 55 S3 o3 w c^ co O *"i ^ fl »^ H el ^ ""^ ^ >-, 00 QH p^+j (y+JM '^'N ^ K ""* d Q "3 d 'O G « CO O 346 DIGESTION usually found between ferments of the same kind derived from different sources may be due to the presence of other substances, and do not necessarily indicate that the ferments are distinct. For the present it may be assumed that the amylase of saliva is the same ferment encountered in the pancreatic juice, and in many, or all, of the tissues. An oxydase or oxidizing ferment is also present in saliva. The salts are calcium carbonate and phosphate (often deposited as ' tartar ' around the teeth, occasionally as salivary calculi in the glands and ducts), sodium bicarbonate, sodium and potassium chloride, and almost always a trace of sul- phocyanide of potassium, detected by the red colour which it strikes with ferric chloride.* The total solids amount only to five or six parts in the thousand. A great deal of carbon dioxide can be pumped out from saliva, as much as 60 to 70 c.c. from 100 c.c. of the secretion — i.e., more than can be obtained from venous blood. Only a small proportion of this is in solution, the rest existing as carbonates. Oxygen is also present even in saliva which has not come into contact with the air, and, indeed, in somewhat greater quantity than in serum (about 0*6 volume per cent, in dog's saliva). Under the microscope epithelial scales, dead and swollen leucocytes (the so-called salivary corpuscles), bacteria, and portions of food, may be found. All these things are as accidental as the last — they are mere flotsam and jetsam, washed by the saliva from the inside of the mouth. But greater significance attaches to certain peculiar bodies, either spherical or of irregular shape, that are seen in the viscid submaxillary saliva of the dog or cat. They appear to be masses of secreted material. The quantity of saliva secreted in the twenty-four hours varies a good deal. On an average it is from i to 2 litres (Practical Exercises, p. 454). Besides its functions of dissolving sapid substances, and so allow- ing them to excite sensations of taste, of moistening the food for deglutition and the mouth for speech, and of cleansing the teeth after a meal, saliva, in virtue of its ferment, amylase, has the power of digesting starch and converting it into the disaccharide maltose, a reducing sugar (Ci^w^u)- ^n man the secretion of any of the three great salivary glands has this power, although that of the parotid is most active. In the dog, on the other hand, parotid saliva has little action on starch, and submaxillary none at all; while in animals like the rat and the rabbit the parotid secretion is highly active. In the horse, sheep, and ox, the saliva secreted by all the glands seems equally inert. When starch is boiled, the granules are ruptured, and the starch * In 100 students the saliva only once failed to give the reaction, and in this individual a trace of sulphocyanide was present 3 days later. It is absent from the saliva of many animals. In 25 dogs submaxillary saliva obtained by stimulation of the chorda tympanj only once gave the ferric chloride reaction, and then faintly. THE CHEMISTRY OF THE DIGESTIVE JUICES 347 passes into colloidal solution, yielding an opalescent liquid. If a little saliva be added to some boiled starch solution which is free from sugar, and the mixture be set to digest at a suitable temperature (say 40° C.), the solution in a very short time loses its opalescence and becomes clear. It still, however, gives the blue reaction with iodine; and Trommer's test (p. 10) shows that no sugar has as yet been formed. Later on the iodine reaction passes gradually through violet into red; and finally iodine causes no colour change at all, while maltose is found in large amount, along with some isomaltose (p. 338), a sugar having the same formula as maltose, but differing from it in the melting-point of the crystalline compound formed by it with phenyl-hydrazine (p. 525). Traces of dextrose, a sugar which rotates the plane of polarization less than maltose, but has greater reducing power, may be found among the end-products when the digestion is conducted in vitro. It is probable that this is produced from the maltose by another enzyme (maltase), present in small amount in saliva. In any case it is well known that mal- tose may be a stage in the hydrolysis of starch to dextrose, and can be detected among the intermediate products when the starch is acted on by dilute acid under conditions which permit a gradual decomposition of the fragments into which the polysaccharide molecule is successively split. But the observation has also been made that the saliva itself (in the cat) may contain a trace of dex- trose (Carlson). The red colour indicates the presence of a kind, or a group, of dextrins sometimes called erythrodextrin, because of this colour reaction; the violet colour shows that at first this is still mixed with some unchanged starch. Soon the dextrins which give the red colour disappear, and are succeeded by other dextrins, which give no colour with iodine, and are therefore called achrodextrins. These are partly, but in artificial digestion never completely, converted into maltose, and can always at the end be precipitated in greater or less amount by the addition of alcohol to the liquid. It is probable that a whole series of dextrins is formed during the digestion of starch. But little is known of their chemical nature. Recently, however, some of these intermediate bodies have been obtained in crystalline form. One of these appears to be a hexa-amylose — i.e., it consists of six C6H10O6 groups, and would therefore be capable of yielding, on the assumption of water, three molecules of maltose or six of dextrose. This is accordingly a relatively small fragment of the big original molecule of starch. When the sugar is removed as it is formed, as is approximately the case when the digestion is performed in a dialyser, the residue of unchanged dextrin is less than when the sugar is allowed to accumulate (Lea). In ordinary artificial digestion, for instance, under the most favour- able circumstances at least 12, to 15 per cent, of the starch is left 348 DIGESTION as dextrin; in dialyser digestions the residue of dextrin may be little more than 4 per cent. This goes far to explain the complete digestion of starch which takes place in the alimentary canal, a digestion so exhaustive that, although dextrins may be found in the stomach after a starchy meal, they do not occur in the intestine, or only in minute traces. Here the amylolytic ferment of the pancreatic juice, which is essentially the same in its action as the amylase of saliva, only more powerful, must effect a very complete conversion of the starch molecules accessible to its attack. It is not inconsistent with tliis, that unchanged starch granules may sometimes be excreted in the fasces, especially when imbedded in raw vegetable structures. For it may be easily shown that un- boiled starch is digested by amylase with far greater difficulty than boiled starch, an illustration of the important part played by cooking in the preparation of the food for digestion. It is a notable fact that amylases, also called diastases, are not confined to the animal body, but are widely distributed in plants. The polysaccharide starch forms the great reserve of carbohydrate material in plant nutrition, and is mobilized for the use of the vegetable cells by being hydrolysed to simple sugars under the in- fluence of these enzymes, just as the polysaccharide glycogen, the great carbohydrate reserve of animal nutrition (p. 533), is mobilized in the form of dextrose under the influence of the diastase of the liver. A diastase, which is present in all sprouting seeds, and may be readily extracted by water from malt, forms dextrin and maltose from starch. The optimum temperature of malt diastase, however, is about 55° C., while that of ptyalin is about 40° C. While a neutral or weakly alkaline reaction is not unfavourable to salivary digestion, it goes on best in a slightly acid medium. It has been shown that the activity of ptyalin on starch, both having been previously dialysed to get rid as far as possible of salts, is increased by the addition of very small amounts of acids and of the neutral salts of strong monobasic acids. The action is decreased by larger amounts of acid (0-0007 to 0-0012 per cent, of hydrochloric acid) and by neutral salts of weak acids. An acidity equal to that of a o-i per cent, solution of hydrochloric acid stops salivary digestion completely, although the ferment is still for a time able to act when the acidity is sufficiently reduced. Strong acids or alkalies permanently destroy it. These facts indicate that in the mouth, where the reaction is weakly alkaline, the conditions are comparatively favourable to the action of the ptyalin. They are still more favourable in the stomach for some time after the beginning of a meal, while the reaction is yet weakly acid. It has been observed that (in cats) salivary digestion may go on for an hour or more in the cardiac end of the stomach, since free hydro- chloric acid does not appear here before that time. Since the con- 349 tents of the cardiac end are not freely intermixed with those of the pyloric end, a greater proportion of sugar is found in the former, and the difference is more marked with solid than with liquid food (Cannon and Day). But during the greater part of gastric digestion the degree of acidity is such that the ptyalin must be hindered. Although the food stays but a short time in the mouth, there is no doubt that, in man at least, some of the starch is there changed into sugar (p. 356)- But this is not the case in all animals. Something depends on the amylolytic activity of the saliva, and something upon the form in which the starchy food is taken, whether it is cooked or raw, enclosed in vegetable fibres, or exposed to free ad- mixture with the secretions of the mouth. The fact already mentioned that hydrolytic changes of the same nature as those produced by enzymes can be brought about in other ways holds good for the salivary amylase. If starch is heated for a time with dilute hydrochloric or sulphuric acid, it is changed first into dextrin, and then into a sugar, which, however, is not the disaccharide maltose, but the monosaccharide dextrose — that is to say, the hydrolysis with acid proceeds a step farther than the hydrolysis in the presence of ptyalin. If maltose is treated with acid in the same way, it is also changed into dextrose. When glycogen (p. i) is boiled with dilute oxalic acid at a pressure of three atmospheres, isomaltose and dextrose are formed (Cremer). Facts already mentioned, and others to be cited later on, show that the action of the other digestive ferments can also be imitated by purely artificial means. Indeed, we may say that the ferments accomplish at a comparatively low temperature what can be done in the laboratory at a higher temperature, and by the aid of what may be called more violent methods. Gastric Juice. — The Abbe Spallanzani, although not, perhaps, the first to recognize, was the first to study systematically, the chemical powers of the gastric juice, but it was by the careful and convincing experiments of Beaumont that the foundation of our exact knowledge of its composition and action was laid. It is difficult to speak without enthusiasm of the work of Beaumont, if we consider the difficulties under which it was carried on. An army surgeon stationed in a lonely post in the wilderness that was then called the territory of Michigan, a thousand miles from a University, and four -thousand from anything like a physiological laboratory, he was accidentally called upon to treat a gun-shot wound of the stomach in a Canadian voyageur, Alexis St. Martin. When the wound healed, a permanent fistulous opening was left, by means of which food could be introduced into the stomach and gastric juice obtained "from it. Beaumont at once perceived the possibilities of such a case for physio- logical research, and began a series of experiments on digestion. After a while, St. Martin, with the wandering spirit of the voyagsur, returned to Canada without Dr. Beaumont's consent and in his absence. Beaumont traced him, with great difficulty, by the help of the agents of a fur-trading company, induced him to come back, provided for his 350 DIGESTION family as well as for himself, and proceeded with his investigations, A second time St. Martin went back to his native country, and a second time the zealous investigator of the gastric juice, at heavy expense, secured his return. And although his experiments were necessarily less exact than would be permissible in a modern research, the modest book- in which he published his results is still counted among the classics of physiology. The production of artificial fistulae in animals, a method that has since proved so fruitful, was first suggested by his work. Gastric juice when obtained pure, as it can be from an acci- dental fistula in man, or, better, by giving a dog with an oesophageal as well as a gastric fistula a ' sham-meal ' (p. 402), is a clear, thin, colourless liquid of low specific gravity (in the dog 1003 to 1006) and distinctly acid reaction. The total solids average about 5 parts per thousand, of which the ash (chiefly sodium and potas- sium chloride, with small quantities of calcium and magnesium phosphate) represents about a fourth, and heat-coagulable sub- stances (proteins, nucleoprotein) about a third. None of these has any special importance in digestion. Of quite a different significance are the three ferments present: pepsin, which changes proteins into peptones; rennin, which curdles milk; and a fat-splitting fer- ment or lipase which, under certain conditions at least, splits up emulsified neutral fats — e.g., the fat of milk — into the alcohol (glycerin) and the fatty acids linked with it, but has so little action upon non-emulsified fat, that when this is taken into the stomach, it eventually passes into the duodenum practically unchanged. The acidity is due to free hydrochloric acid, the other important constituent of the juice. In the dog the proportion of this acid varies from 0-46 to 0-58 per cent. In such analyses as have been made of approximately pure human gastric juice a smaller percentage of hydrochloric acid has usually been obtained (at most 0-35 to 0-4 per cent.). But there is some reason to believe that if the human juice could be collected in a faultless manner, and especially free from any admixture with saliva or with a pathological secretion of mucus, it would show as high a percentage of acid as the dog's juice. In cases of cancer, whether the growth is situated in the stomach or not, the free hydrochloric acid of the gastric juice is usually much reduced, and often absent. Under such conditions some lactic acid may be present in the stomach, being produced from the carbo-hydrates by the action of bacteria (Bacillus acidi lactici), which are normally held in check by the hydrochloric acid, although not rendered incapable of growth when they have passed on into the intestine. Even in the strength of 0-07 to 0-08 per cent, hydro- chloric acid prevents the formation of lactic acid from dextrose. Indeed, when all the hydrochloric acid of the gastric juice is com- bined with proteins, the protein-acid compound still inhibits the growth of bacteria in the stomach, although not so efficiently as the ftt£ blGbStlVE JUICES 351 Same amount of free acid. That in normal gastric juice the acidity is not due to lactic acid can be shown by shaking the juice with ether which takes up lactic acid, and then applying Uffelmann's test to the ethereal extract (Practical Exercises, p. 460)- More than this, it is not due to an organic, but to an inorganic acid, for healthy gastric juice causes such an alteration in the colour of aniline dyes like congo-red and methyl violet as would be produced by dilute mineral acids, and not by organic acids, even when present in much greater strength.* Finally, when the bases and acid radicles of the juice are quantitatively compared, it is found that there is more chlorine than is required to combine with the bases; the excess must be present as free hydrochloric acid In the pure gastric juice of fishes like the dogfish and skate, however, the acid is said not to be hydrochloric but an organic acid. The quantity of gastric juice secreted depends upon the nature and amount of the food. It has been estimated at as much as 5 litres in twenty-four hours, or several times the quantity of saliva secreted in the same time. With sham feeding a dog may yield 200-300 c.c. in an hour. The great action of gastric juice is upon proteins. In this two of its constituents have a share, the pepsin and the free acid. One member of this chemical copartnery cannot act without the other ; peptic digestion requires the presence both of pepsin and of acid ; and, indeed, an active artificial juice can be obtained by digesting the gastric mucous membrane with dilute (0-2 to 0-4 per cent.) hydrochloric acid. A glycerin extract of a stomach which is not too fresh also possesses peptic power, the zymogen or mother substance, pepsinogen, having been activated to pepsin. Free acid very readily effects this activation, but this is far from being the only function of the hydrochloric acid, for active pepsin still requires the addition of a sufficient quantity of acid to render its proteolytic power available. Well- washed fibrin obtained from blood is a convenient protein for use in experiments on digestion; although, of course, for many purposes only isolated purified proteins can be employed. Since the blood contains traces of pepsin, the fibrin should be boiled to destroy any which may be present (see also p. 453). If we place a little fibrin in a beaker, cover it with gastric juice obtained from a dog or with 0-4 per cent, hydrochloric acid, to which a small quantity of pepsin or of a gastric extract has been added, and put the beaker in a water-bath at 40° C., the fibrin soon swells up and becomes translucent, then begins to be dissolved, and in a short time has disappeared (see Practical Exercises, p. 458). If we examine the * A dilute solution of congo-red is turned violet by organic and blue by inorganic acids; the gastric juice turns it blue. Methyl violet is rendered blue by an inorganic acid like hydrochloric acid, and green if more of the acid be added. It is not altered by organic acids. Gastric juice turns it blue. 35>2 DIGESTION liquid before digestion has proceeded very far, we shall find chiefly so-called acid -albumin in solution; later on, chiefly albumoses; and of these some authors distinguish the primary albumoses (proto -albumose and hetero-albumose), the first to appear in quantity, followed by secondary or deutero-albumoses (p. 10).* Still later, peptones in large and always relatively increasing amounts will be present along with the albumoses. From this we conclude that acid-albumin is a stage in the conversion of fibrin into albumose, and albumose a half-way house between acid -albumin and peptone. It must not be supposed, however, that all the protein is first changed into acid-albumin before any of the acid-albumin is changed into albumose, or that all the protein has already reached the albumose stage before peptone begins to appear. On the contrary, a certain amount of albumoses and of peptones are present very early in peptic digestion, while the greater part of the original protein is still unaltered. • Among the somewhat vaguely characterized group of bodies comprised under the term peptones, there are no doubt decomposition products of the proteins in which the hydrolysis has been carried to different degrees. Similar, but not identical, intermediate substances occur in the digestion of the other proteins, including that of bodies like gelatin, which are not ordinary proteins, but which pepsin can digest. The generic name of proteose properly includes all bodies of the albumose type, the term ' albumose ' itself being sometimes reserved for such intermediate products of the digestion of albumin; while those of fibrin are called fibrinoses; of globulin, globuloses; of casein, caseoses; and so on. The peptones produced from different proteins are also not absolutely identical. If the digestion is prolonged, the peptones first formed are in turn further hydrolysed, so that eventually a considerable proportion of the original protein is converted into bodies which no longer give the biuret reaction. In the stomach, during the four or five hours for which gastric digestion ordinarily lasts, none of the protein passes beyond the stage of proteose and peptone, including those relatively simple ' abiuret ' compounds which still consist of several ' building-stones, ' chiefly, it would seem, the amino-acids, phenylalanin, and prolin, linked together. When precautions are taken to prevent the passage of any portion of the contents of the duodenum into the stomach, no amino-acids can be detected in the gastric contents during the digestion of protein. In this connection it is interesting to note that none of the polypeptides hitherto prepared (p. 2) are decomposed by pepsin. It is not known at what points in the link- age of the groups that compose the complex protein molecule the pepsin ruptures the chain, but the points of attack are different from those of trypsin. The pancreatic juice, as we shall see later on, not * In the light of modern investigation the results of fractional precipitation by salts of the products of proteolysis have lost a good deal of their interest, and it is seen that undue importance has often been attached to them. The student should be warned that such terms as ' albumose ' and ' peptone ' do not indicate precise chemical differences between the products separated in this way, nor even invariably such differences in molecular weight as the current schemata of the digestive processes are apt to imply. Some so-called ' peptones ' may indeed have a higher molecular weight and be more nearly related to the original protein than some so-called ' albumoses.' THE CHEMISTRY OF THE DIGESTIVE JUICES 353 only effects a more complete conversion into peptones, but can split up the whole or a very large proportion of the peptones themselves into amino-acids and the other ' building-stones ' of the original protein. Since the subject of protein digestion must come up again, it will be well to postpone any closer discussion of the process till we can view it as a whole. In the meantime it is only necessary to repeat that pepsin alone cannot digest proteins at all. Its action requires the presence of an acid; in a neutral or alkaline medium peptic digestion stops. The precise mode of action of the acid is by no means clear. Dilute acid alone does not dissolve coagulated proteins like boiled fibrin, or does so only with extreme slowness. But it causes them to swell up by imbibition of water, and probably in this way facilitates the entrance of the ferment. Uncoagulated proteins are readily changed by acid into acid-albumin ; and by the prolonged action of acids, especially at a high temperature, further changes of much the same nature as those produced in peptic digestion may be caused in all proteins. But under the ordinary conditions of natural gastric digestion, it may be said that the acid alone does little until it is aided by the ferment, just as the ferment alone does nothing without the aid of the acid. The acid enters into a temporary combination with the protein, the more highly hydrolysed proteins, such as peptones, combining with a greater proportion of acid than such proteins as fibrin or albumin. These compounds so easily undergo hydrolytic dissociation that, in spite of its union with the proteins, the hydrochloric acid is able to act along with the pepsin, so that peptic digestion goes on even when enough protein is present to combine with all the acid. There is some evidence that in the gastric juice the pepsin exists in the form of an unstable compound with hydrochloric acid, and it is probably this pepsin-hydrochloric acid compound which is the actual catalytic agent in peptic digestion. Although hydrochloric acid acts most powerfully, other acids, such as nitric, phosphoric, sulphuric, or lactic (arranged in the order of their efficacy), can replace it. The Milk-Curdling Action of Gastric Juice. — The milk-curdling ferment, rennin, or chymosin, is contained in large amount in an extract of the fourth stomach of the calf, which, as rennet, has long been used in the manufacture of cheese. It exists in the healthy gastric juice of man, but disappears in cancer of the stomach and in chronic gastric catarrh. It has been stated by a number of observers that the properties of rennin are never found in gastric juice or any preparation obtained from it or from the gastric mucous membrane unless pepsin is present. This has suggested that there is no separate milk-curdling ferment, but that the clotting or precipitation of caseinogen is merely an associated a.ction of the pepsin. Ferments of the most varied origin will curdle milk. Pawlow has maintained 354 that the milk-curdling property not only of the gastric juice, but also of the pancreatic juice and of the secretion of Brunner's glands, is associated with the proteolytic ferment. He asserts that when the comparison is instituted under proper conditions there is an exact parallelism between the proteolytic and the milk-curdling power of these secretions, no matter what the circumstances may be in which they are collected, or the influences to which they are exposed after collection. He has found it im- possible to separate from any one of them a fraction which has milk-curdling power without proteolytic power. On the other hand, the majority of investigators maintain the separate identity of rennin. Hammarsten especially states that he can destroy the peptic activity without destroying the milk-curdling power of gastric extracts, and vice versa. According to Burge, when a solution dis- playing both peptic and milk-curdling power is electrolyzed, the pepsin action is abolished at a certain stage, while the rennet action is unaffected. It would seem, then, that the balance of evidence is in favour of the separate identity of the rennet enzyme. However this may be, the curdling of milk by the gastric ferment includes two processes: (i) An action on caseinogen in the course of which it acquires new properties, becoming changed into casein. This substance is not capable of being converted into casein, and remains in solution in the whey. (2) The altered casein- ogen or casein combines with soluble calcium salts and in the presence of these is precipitated as the curd. The change which occurs in the caseinogen has been the subject of much discussion, which has not yet led to a definite conclusion. According to some observers, the change consists in a decomposition of caseinogen, in the course of which a new substance, whey-protein, not previously present in the milk, is split off. This substance is not capable of being precipitated by lime salts, and remains in solution in the whey. According to others, the molecule of caseinogen is simply changed into two molecules of casein (van Slyke). Dilute acid will of itself precipitate caseinogen, and the presence of acid, and particularly hydrochloric acid, in the gastric juice helps its milk-curdling action. But that a ferment is really concerned is indicated by the fact that the juice, after being made neutral or alkaline, still curdles milk, and that this power is destroyed by boiling. The optimum temperature is the same as that of the other ferments of the digestive tract, about 40° C. (p. 337). The persistence of the milk-curdling activity in the presence of OH ions, while for peptic activity free H ions are neces- sary, is a further and, indeed, a strong argument in favour of the separate existence of the rennet ferment. As to the exact function of the milk-curdling ferment of the gastric juice in digestion, we have no precise knowledge. It seems superfluous if we suppose that the free acid is able of itself to do all THE CHEMISTRY OF THE DIGESTIVE JUICES 355 that the ferment does along with it. But there is evidence that the curd produced by the ferment is more profoundly changed than the precipitate caused by dilute acids; for the latter may be redissolved, and then again curdled by rennin in the presence of calcium salts, while this cannot be done with the former. We may suppose, then, that the ferment is capable of effecting changes more favourable to the subsequent action of the pepsin upon the casein than those which the acid alone would effect. Or it may be that the ferment acts in the early stages of digestion before much acid has been secreted. The curdling of milk probably plays a part in ensuring the retention of this food, the proper digestion of which is all-impor- tant for the suckling, for a sufficient length of time in the stomach. Otherwise, like water and watery liquids in general, it would be rapidly passed into the duodenum. Even if this were not the case, there is another reason for early curdling. Milk is a very dilute food, and the immense proportion of water in it might weaken the gastric juice too much for rapid digestion of the proteins. The separation of the whey and its prompt escape through the pylorus would obviate this. But caution should be exercised in giving a physiological value to all the details of the milk-curdling action of the gastric juice. Milk-curdling ferments, or, at any rate, ferments with a milk-curdling influence, have an extremely wide distribution, both in secretions which in normal circumstances can never come into contact with milk, and in the tissues of animals and plants. Many bacteria produce them. And it appears that in the suckling, where it might be expected, if anywhere, to have a definite and important office, the rennet action of the gastric juice is distinctly less than in the adult. It is worthy of note that the curd formed by rennet from human milk is more finely divided than that formed from cow's milk, and therefore is more easily digested. The addition of lime-water or barley-water to cow's milk keeps the curd from adhering in large masses, and thus aids its digestion — a fact which is sometimes usefully applied in the artificial feeding of infants. Gastric Lipase. — On fats gastric juice has usually been supposed to have no action, although everybody admits that it will dissolve the protein constituents of fat-cells and the protein substances which keep the fat-globules of milk apart from each other. It has, how- ever, been recently shown that both in the stomach and in vitro (with glycerin extracts of the gastric mucous membrane) a consider- able amount of well -emulsified fat may be split up, and that this is due to a ferment which is different in several respects from the lipase of pancreatic juice. Gastric juice splits up fat, both in neutral and in weakly acid solutions. The slightest excess of alkali checks the action. The glycerin extract is much more resistant to alkali, while very sensitive to hydrochloric acid. This indicates that the fat-splitting ferment exists in the mucous membrane in a 356 DIGESTION different form from that in which it exists in the juice — namely, as a zymogen or mother-substance. But while the zymogen of the pancreatic lipase is activated by bile, this is not the case with the mother-substance of the gastric lipase. It appears that in the suckling the lipase of the gastric juice plays a more important part than in later life. This is obviously in accordance with the fact that the specific food of the suckling — milk — contains as an essential con- stituent a large proportion of emulsified fat. The conditions for the emulsification of fat do not exist in the gastric juice, and this is the reason why the gastric lipase has so slight an effect upon un- emulsified fat, which presents a surface of contact proportionally so small. In any case the amount of fat hydrolysed in the stomach under ordinary con litions is small in comparison with the amount split in the intestine, although it has been shown that with a diet rich in fat some of the intestinal contents, including pancreatic lipase, may pass back into the stomach. As regards the carbo-hydrates, the swallowed saliva will continue to act on starch in the stomach, so long as the acidity is not too great ; while the hydrochloric acid of the gastric juice is able to invert cane- sugar, changing it into a mixture of dextrose and levulose,* and also, doubtless, to hydrolyse to dextrose a portion of the maltose formed by the saliva. Altogether, there is no doubt that the pro- portion of the carbo-hydrates of the food digested in the stomach is far from insignificant. The Antiseptic Function of the Gastric Juice. — The stomach, with its acid contents, forms during the greater part of gastric digestion a valve or trap to cut off the upper end of the intestine from the bacteria-infested regions of the mouth and pharynx, and to destroy or inhibit the micro-organisms swallowed with the food and saliva. The occasional presence in vomited matter of sarcinae or regularly arranged groups of micrococci, generally four to a group, shows that under abnormal conditions the gastric contents are not perfectly aseptic ; and even from a normal stomach active micro-organisms of various kinds can be obtained. But upon the whole there is no doubt that the acidity of the gastric juice is an important check on bacterial activity during the first part of digestion, and in the upper portion of the alimentary canal. Koch has shown that the acidity of the gastric juice of a guinea-pig is sufficient to kill the comma bacillus of cholera. Normal guinea-pigs fed with cholera bacilli * These are both reducing sugars, but, as their names indicate, they rotate the plane of polarization in opposite directions. The specific rotatory power of levulose is greater than that of dextrose, so that when cane-sugar is com- pletely inverted, although equal quantities of dextrose and levulose are pro- duced, the plane of polarization is rotated to the left. Cane-sugar itself rotates it to the right. The term ' inversion ' has been extended to include the similar hydrolysis of other sugars of the disaccharide group — e.g., maltose to dextrose, and lactose to a mixture of dextrose and galactose, even although the products are not levo-rotatory. THE CHEMISTRY OF THE DIGESTIVE JUICES 357 were unaffected. But if the gastric juice was neutralized by an alkali before the administration of the bacilli the guinea-pigs died. Charrin found, too, that digestion with pepsin and hydrochloric acid causes an appreciable destruction or attenuation of diphtheria toxin. Bacteria, like the lactic acid bacillus, which form acid products, may be less profoundly affected by the acid gastric juice than the putre- factive bacteria, which, on the whole, form alkalies, and are there- fore accustomed to an alkaline medium. Yet we have seen that the growth of even the lactic acid bacillus is very strictly controlled when the gastric juice contains the normal amount of hydrochloric acid. It has been supposed by some that this bactericidal action is the chief function of the stomach, and the question has been asked why we should attribute any digestive importance to the secretion of that viscus, since the pancreatic juice can do all that the gastric juice does, and some things which it cannot do. Further, it has been shown that a dog may live five years after complete excision of the stomach, comport himself in all respects like a normal dog, and when killed for necropsy show every organ in perfect health (Czerny). In man, too, the stomach has been excised with a successful result. But if this is to be admitted as evidence against the digestive function of the stomach, it is just as good evidence against the bactericidal function, particularly as it has in addition been shown that even putrid flesh has no harmful effect on a dog after excision of the stomach, any more than on a normal dog. And, indeed, the reason- ing is fallacious which assumes that what may happen under ab- normal conditions must happen when the conditions are normal. For nothing is impressed more often on the physiological observer than the extraordinary power of adaptation, of making the best of everything, which the animal organism possesses. Doubtless a dog without a stomach will use to the best advantage the digestive fluids that remain to him ; and the pancreatic juice, with the aid of the bile and the succus entericus, may be adequate to the complete task of digestion. So, too, a man from whom the surgeon has removed a kidney, or a testicle, or a lobe of the thyroid gland, may be in no respect worse off than the man who possesses a pair of these organs. But what do we deduce from this ? Not, surely, that the excised thyroid, or testicle, or kidney was useless, or the gastric juice in- active, but that the organism has been able to compensate itself for their loss. Further, it would seem that the fate of the protein or of part of the protein digested and absorbed by the stomach is different from that digested and absorbed by the intestine. For after the operation of gastro-enterostomy (the establishment of an artificial opening between the stomach and the small intestine through which the food passes rapidly without having to submit to the challenge of the pyloric sphincter), the ingested nitrogen is more quickly eliminated than when the protein is first subjected to full gastric 358 DIGESTION digestion. So that when the quantity of protein in the food is increased above that necessary for nitrogen equilibrium (p. 602), none of the excess is assimilated and stored up, as is the case in a normal animal (Levin, etc.). Pancreatic Juice. — Pancreatic juice, bile, and intestinal juice are all mingled together in the small intestine, and act upon the food, not in succession, but simultaneously. But by artificial fistulae in animals they can be obtained separately ; and occasionally some of them can be procured through accidental fistulae in man. It is said that under certain conditions, especially when fat or oil is introduced into the stomach, the pylorus may remain open long enough to permit the passage of pancreatic juice or bile from the duodenum into the stomach, and this has been recommended as a practical method of obtaining these secretions in man. Human pancreatic juice, as obtained from a fistula, is a clear, only slightly viscid liquid of distinctly alkaline reaction to litmus. Its specific gravity is about 1007 to 1010. The total solids constitute about i'5 or 2 per cent., of which a little less than i per cent, is made up of inorganic salts, chiefly sodium carbonate, with small quantities of chlorides. The balance of the solids consists mainly of proteins. The alkaline reaction is due to the sodium carbonate, and it is worthy of remark, as showing the important part taken by this secretion in the neutralization of the chyme, that when titrated against standard acid the alkalinity of the pancreatic juice is not much less than the acidity of the gastric juice when titrated against standard alkali. The quantity of pancreatic juice secreted during the twenty-four hours in an average man has been estimated at 500 to 800 c.c. from observations on cases of fistula. Probably under perfectly normal conditions it is greater. A so-called arti- ficial pancreatic juice can be made by extracting the pancreas with water or glycerin. Since better methods of obtaining the natural juice have been developed, these extracts have lost some of their importance. Fresh pancreatic juice contains four ferments : (i) Thezymogen or mother-substance, trypsinogen, of a proteolytic or protein-digesting ferment, trypsin ; (2) an amylolytic ferment, or amylase ; (3) a fat- splitting or lipolytic ferment, steapsin ; (4) a milk-curdling ferment. The question whether the last is a different body from the proteolytic ferment has been discussed just as in the case of the gastric rennin (see p. 353). In any case, it cannot be considered as taking any practical share in digestion, since it can hardly ever happen that milk passes through the stomach without being curdled. Trypsinogen has no action upon proteins, but in normal digestion it is changed into active trypsin by the enterokinase of the intestinal juice (p. 372). Pancreatic juice collected without contact with intestinal contents or with the mucous membrane of the intestine THE CHEMISTRY OF THE DIGESTIVE JUICES 359 does not digest proteins. The same is true of extracts of perfectly fresh pancreas, but if the pancreas is allowed to stand for a time, the extracts contain active trypsin, perhaps because some decomposition product has activated the trypsinogen. Some writers, however, state that when contamination of the gland with intestinal contents or contact with the mucosa has been avoided in its removal from the body, such extracts will remain inactive for months, although the trypsinogen can at once be activated to trypsin by the addition of enterokinase. Trypsin, to a certain extent, corresponds with pepsin in its action on proteins. But it acts energetically in an alkaline as well as in a not too acid medium (a very slight amount of digestion may go on in distilled water) ; and its action, unlike that of pepsin — at least in digestions of moderate duration — does not stop at the peptone stage, but goes on rapidly to the production of the amino-acids, the basic substances arginin, lysin, and histidin, known as the hexone bases, and most of the other decomposition products obtained by boiling proteins with dilute acids. The most important of these products, so far as they have been isolated and identified, are enumerated in the table on p. 360 (see also pp. 1-3). As to the chemical nomenclature of these bodies, the student should refer to his textbook of organic chemistry. It need only be remarked here, by way of illustration, that when, e.g., leucin is designated as a-amino-isobutylacetic acid, this indicates that it can be derived from the fatty acid isobutylacetic acid by the substitution of an amino group, NH2, for a hydrogen atom in the a-carbon group (see p. 556) of the fatty acid — i.e., the group next the carboxyl (COOH) group. Thus, ~H3NcH.CH2.CH2.COOH , £§3\CH.CH«.CH.COOH. CH3/ * ^ <-H3/ ' | Isobutylacetic acid. Leucin. NH2 When norleucin, an amino-acid found especially in the proteins of nervous tissue, is termed a-amino-caproic acid, it is indicated that NH2 replaces one H in the aCH2 group of caproic acid (CH3.CH2.CH2.CH2. CH2.COOH). The long chemical name of isoleucin (a compound also derived from the proteins of nervous tissue and from some plant proteins} indicates that in propionic acid, CH3.CHa.COOH, 0 « NH2 is introduced into the a group, yielding CHg.CH.COOH, ft I NH2 or amino-propionic acid (alanin); while in the j8 group one H is replaced by a methyl (CH3) group, and another H by an ethyl (G>H5) giouy, yielding finally isonuclein : H*3 \CH.CH.COOH. NH2 DIGESTION CHIEF DECOMPOSITION PRODUCTS OF PROTEINS. MONOAMINO-ACIDS AND THEIR COMPOUNDS. /Glycin or glycocoll (aminoacetic acid), CH2(NH2)COOH. Alanin (aminopropionic acid), CH3CH(NH2)COOH. Serin or oxyalanin (oxyaminopropionic acid), CH2OH.CH(NH2).COOH. Valin or aminovalerianic acid, ^3/CHCH(NH2)COOH- Leucin (a-aminoisobutylacetic acid) £^3^)CHCH2CH(NH2)COOH. Norleucin («-amino-caproic acid), CH3.CH2.CH2.CH2.CH(NH2).COOH. Isoleucin (a-amino-/3-methyl-/3-ethyl-propionic acid), rH \ ^3^>CH.CH(NH2).COOH. Cystein (a - amino-/3 - thiopropionic acid), CH2 (SH).CH(NH2)COOH, which is unstable, two molecules of it easily yielding cystin di-(n-amino-/3-thiopropionic acid) , COOH.CH(NH2).CHa.S.S.CH2.CH(NHj).COOH. fAspartic or aminosuccinic acid, CH(NH2).COOH. Jen I CH2COOH. Q rt I Glutamic or glutaminic acid, C *—~ •a -si Prolin (pyrrolidin carboxylic acid). Oxyprolin (oxypyrrolidin carboxylic acid). Tryptophane (a-amino-j3-indol-propionic acid), yNH— CH \N — C— CH2.CH(NHa).COOH. Histidin (/J-imidazol-a-aminopropionic acid), C6H9N3O2. DlAMINO-ACIDS AND THEIR COMPOUNDS. Lysin (a-amino-€-amino-caproic acid, CH2NH2(CH8)3CH(NH2)COOH. Arginin (guanidinaminovalerianic acid), ufa — p/ NH2 \NH CH2 (CH2)2CH (NH2) COOH . Ammonia (representing the so-called ' amide-nitrogen,' and liberated from the products of acid hydrolysis of proteins by heating the mixture after addition of alkali). It has not been shown that ammonia is itself one of the ' Bansteine ' of the proteins. • THE CHEMISTRY OF THE DIGESTIVE JUICES 361 In the artificial compounds of two or more amino-acids which have been synthesized by Fischer and named by him polypeptides (p. 2), the carboxyl group of one amino-acid is linked with the amino group of another. For example, a molecule of alanin and a molecule of glycin form, with loss of a molecule of water, a molecule of alanyl-glycin, according to the equation NH2.CHa.CO|pH +H!NH.CH.CH3— H2O= NH2.CH2.CO.NH.CH.CH3. COOH COOH Glycin. Alanin. Glycyl-alanin. Two or more molecules of the same amino-acid can be linked in the same way; e.g., two molecules of glycin yield a molecule of glycyl-glycin, and so on. It has been proved that polypeptides identical witn some of these synthetic bodies are present in the peptone mixtures derived from the native proteins, so that it must be assumed that one of the ways at least in which amino-acids are linked in the protein molecule is that described. It has been suggested that the early appearance of some of these amino-acids in pancreatic digestion is not really due to trypsin, but to other ferments, peptases, which act upon the peptones formed by the trypsin. There is, however, no clear evidence of the existence of a separate peptone-splitting enzyme in pancreatic juice, like the erepsin of intestinal juice, and it is therefore most natural to suppose that under the influence of trypsin the protein molecule breaks at different points from those at which it ruptures under the influence of pepsin. After the most prolonged artificial digestion with trypsin, a residue of the protein remains unconverted into these relatively simple substances. But even this small portion of the original protein has undergone a great change, for it no longer gives the biuret reaction. It can be split into amino-acids, etc., by heating with acid, and also by the action of the erepsin of the intestinal juice, and then yields mainly prolin and phenylalanin, substances which are generally not to be detected among the decomposition products of protein after digestion with pancreatic juice. This illustrates the important fact that some of the ' building stones ' of the protein molecule can be separated from it with far greater ease than others. Tyrosin, tryptophane, and cystin appear very early in the digestive fluid, and tyrosin, as shown in the following example from Abder- halden, may be completely liberated at a time when glutaminic acid is scarcely more than beginning to appear. The plant protein edestin, obtained from cottonseed, was digested with pancreatic juice or an extract containing trypsin. The quan- tities of tyrosin and glutaminic acid liberated at different periods of the experiment are expressed as percentages of the total amounts of these substances contained in the edestin. 362 DIGESTION Duration of Digestion : i Day. 2 Days. 3 Days. : 7 Days, j 16 Days. Tyrosin Glutaminic acid 78-4 4' 3 97'6 7*4 97'6 io'9 100 3ri IOO 60' 2 When trypsin ads upon protein already digested by pepsin, this partially hydrolysed residue is smaller than when the trypsin acts alone, no matter for how long a time. Also the decomposition of a given quantity of protein by trypsin is accomplished in a notably shorter time if it has been previously subjected to the action of pepsin. This illustrates the co-operative relation of these two ferments — a relation still more clearly implied in the fact that, although trypsin readily forms albumoses and peptones from native protein when such is offered to it, yet in natural digestion the great albumose- and peptone-forming ferment is pepsin. In the lumen of the intestine the trypsin is confronted mainly with protein already hydrolysed to the albumose and peptone stage in the stomach. In other words, instead of the very large molecules of the original protein food with a weight of perhaps 5,000 to 7,000, the trypsin begins its action on a much larger number of much smaller molecules of only one-twentieth the initial weight or less. The statement is sometimes made that trypsin is a stronger proteolytic ferment than pepsin. This may be true in the sense that trypsin carries the de- composition down to bodies of smaller molecular weight than pepsin. But within the range of its hydrolytic action pepsin decomposes certain proteins and allied bodies more readily than trypsin — e.g., the serum proteins, and especially elastin and the constituents of connective tissue. In all that we have hitherto said regarding tryptic digestion we have supposed that putrefaction has been entirely prevented — e.g., by the addition of toluol. If no antiseptic is added to a tryptic digest, it rapidly becomes filled with micro-organisms, and emits a very disagreeable faecal odour ; and now various bodies which are not found in the absence of putrefaction make their appearance, such as indol, skatol, and other substances, to which the fascal odour is due. They are not true products of tryptic digestion, but are formed by the putrefactive micro-organisms, which can themselves split off from proteins numerous decomposition products, including tyrosin, and change tyrosin into indol. Amylase or pancreatic amylase, the diastatic or sugar- forming ferment of pancreatic juice, changes starch into dextrin and maltose, just as the ptyalin of saliva does. The two ferments are possibly identical, but under the conditions of action of the pancreatic juice its diastatic power is greater than that of saliva, and it readily acts THE CHEMISTRY OF THE DIGESTIVE JUICES 363 on raw starch as well as boiled. Pancreatic amylase is mainly, perhaps entirely, present in the juice in the form of active ferment. If a zymogen stage exists, the mother-substance is less stable or less easily extracted from the gland than is trypsinogen. In this respect amylase also resembles ptyalin. A small amount of maltase is contained in pancreatic juice, and further hydrolyses to dextrose a portion of the maltose formed by the amylase. Steapsin or pancreatic lipase splits up neutral fats into glycerin and the corresponding fatty acids. The latter unite with the alkalies of the pancreatic juice and the bile to form soaps. In this important process bile acts as the helpmate of pancreatic juice; together they effect much more than either or both can accomplish by separate action. Many tissues contain fat-splitting ferments or Upases, some of which are perhaps identical with the pancreatic lipase. The lipase exists as active ferment in the pancreatic j uice, but there is reason to believe that a portion of it may be present as a zymogen in the gland, and probably in the secretion as well. It is changed Jnto active ferment by the bile salts. Active lipase can also be extracted from the pancreas by glycerin or water. It is to be noted that it is only the proteolytic enzyme which is totally inactive till it reaches the intestine. The significance of this will be discussed later on. Bile. — Bile is a liquid the colour of which varies in different groups of animals, and even in the same species is not constant, depending on the length of time the fluid has remained in the gall-bladder and other circumstances. When it is recognized that the colour is due to a series of pigments, which are by no means stable, and of which one can be caused to pass into another by oxidation or reduction, this want of uniformity will be easily intelligible. The fresh bile of carnivora is golden-red. The bile of herbivorous animals is in general of a green tint, but, when it has been retained long in the gall-bladder, may incline to reddish-brown. Fresh human bile, as it flows from a fistula just established, is of a reddish-brown, golden- yellow or yellow colour. Beaumont speaks of the yellowish bile which he could press into the stomach of St. Martin by manipulating the abdomen. In a case observed by the writer, it was seen that when the bile flowing from a fistula was allowed to spread out in a dressing, it became greenish, because of oxidation of a part of the bilirubin to biliverdin, although as it actually escaped from the fistula it was yellow. The bile of a monkey taken from the gall-bladder immediately after death is dark green, but if left a few hours in the gall-bladder it is brown, the green pigment having been reduced. It should be remembered that human bile from the post-mortem room may alter its colour in the interval which must elapse before it can usually be procured after death. Bile, as obtained from fistulse in otherwise healthy persons, has a specific gravity of about 1008 to lo 10. In the gall-bladder water is absorbed from the bile and DIGESTION mucin added to it, so that the specific gravity of bladder bile is as high as 1030 to 1040. The reaction is feebly alkaline to litmus. The composition of two specimens of human bile — one from a fistula, the other from the gall-bladder — is shown in the following table : Bladder Bile. Fistula Bile. Water - 898-1 977*4 Solids - 101-9 22'6 Mucin and other substances insoluble in alcohol I4'5 2-3 Sodium taurocholate and sodium glycocholate 56-5 IO-I Inorganic salts ... 6-3 8-5 Fat ) Lecithin }- - 30-9 0-05 CholesterinJ 0-56 The substance which renders bladder bile viscid, but which is present in much smaller amount in bile from a fistula, and is probably entirely absent from the fluid as it is secreted by the liver-cells, is commonly termed 'mucin.' It has been shown, however, that in many animals — for example, the ox, dog, sheep, etc. — the substance is not a true mucin. It does not yield, like mucin, on boiling with dilute acid, a carbo- hydrate group (viz., glucosamine, C6HuO5Nll2, corresponding to dextrose in which OH is replaced by NH2). It is relatively rich in phos- phorus, and consists — mainly, at any rate — of a phospho-protein (p. 2). The mucilaginous substance of human bile consists largely of true mucin. Mucin is scarcely to be looked upon as an essential constituent of bile ; it is not formed by the actual bile-secreting cells, but by mucous glands in the walls and goblet-cells in the epithelial lining of the larger bile-ducts, and especially of the gall-bladder. Bile-Pigments. — It has been said that these form a series, but only two of the pigments of that series are present in normal bile, bilirubin, and biliverdin. In human bile, the former, in herbivorous bile and that of some cold-blooded animals, such as the frog, the latter is the chief pigment. But bilirubin can be extracted in large amount from the gall-stones of cattle ; while the placenta of the bitch contains bili- verdin in quantity, although, as in all carnivora, it is either absent from the bile or exists in it in comparatively small amount. These facts show that the two pigments are readily interchangeable, but there is no question that bilirubin is the pigment which is formed by the liver-cells. Bilirubin (C32H36N4O6) can be prepared from powdered red gall- stones by dissolving the chalk with hydrochloric acid, and extracting the residue with chloroform, which takes up the pigment. From this solution, on evaporation, or from hot dimethyl anilin, beautiful rhombic tables or prisms of bilirubin separate out. Biliverdin (C32H36N4O8) can be obtained from the placenta of the bitch by extraction with alcohol. It is insoluble in chloroform, and by means of this property it may be separated from bilirubin when the two happen to be present together in bile. Biliverdin can also be formed from bilirubin by oxidation. By the aid of active oxidizing agents, THE CHEMISTRY OF THE DIGESTIVE JUICES 365 such as yellow nitric acid (which contains some nitrous acid), a whole series of oxidation products of bilirubin is obtained, beginning with biliverdin, and passing through bilicyanin, a blue pigment, and other intermediate bodies, to choletelin, a yellow substance. It is question- able whether these are all definite compounds. This is the foundation of Gmeliris test for bile-pigments (see Practical Exercises, p. 462)- The same colours are produced, and in the same order, when a solution of bilirubin in chloroform is treated with a dilute alcoholic solution of iodine . The positive pole of a galvanic current causes the same oxidative changes, the same play of colours, while the reducing action of the negative pole reverses the effect, if the action of the positive electrode has not gone too far. These reactions can also be used for the detection of bile -pigments. By the reducing action of sodium amalgam on bilirubin, hemi- bilirubin (C33H44N4O6) is obtained. It gives a beautiful red colour with ^-dimethylaminobenzaldehyde (Ehrlich's reaction). Hemibilirubin is identical with the urobilinogen of urine from which urobilin is derived. Urobilinogen and urobilin (often called in this connection stercobilin) are also found in the faeces from birth onwards, although not in the meconium (p. 424). Urobilinogen is derived from the normal bile- pigment by reduction in the intestine itself, where reducing substances due to the action of micro-organisms are never absent in extra -uterine life . The bile of most animals shows no characteristic absorption spectrum. But the fresh bile of certain animals, the ox, for instance, does show bands. These, however, are not due to the normal bile-pigment, and they are not essentially changed when this is oxidized or reduced by electrolysis. MacMunn attributes the spectrum of the bile of the ox and sheep to a body which he calls cholohaematin, and which does not belong to the bile-pigments proper. The Bile-Salts. — These are the sodium salts of certain acids, of which glycocholic and taurocholic are the chief. In the bile of omnivora, including man, both are in general present, and in various proportions; in human bile there is more glycocholic than taurocholic acid; some- times taurocholic acid is entirely absent. In the bile of many carnivora — e.g., the dog and cat — only taurocholic acid is found; in that of the carnivora generally it is by far the more important of the two acids. In the bile of most herbivora there is much more glycocholic than taurocholic acid. The bile acids are paired acids: glycocholic acid (better named cholyl-glycin) formed by the union of glycin and cholic acid, and taurocholic acid (or cholyl-taurin), consisting of cholic acid united with taurin. The decomposition of the bile-acids into these substances is effected by boiling them with dilute acid or alkali, a molecule of water being taken up; thus — 06 + H20 = CH2 (NHa) .COOH + C24H40O6 ; Glycocholic acid. Glycin. Cholic acid. C26H45NS07 + H20 = CH2(NH2).CH2.S02.OH + C24H40O5. Taurocholic acid. Taurin. Cholic acid. A notable difference between glycocholic and taurocholic acid is that the latter contains sulphur. The whole of this belongs to the taurin. Both glycin and taurin are derived from the disintegration of proteins. We have already seen that among the products of protein hydrolysis a sulphur-containing body, cystein, which is readily changed into cystin, is found, and there is good evidence that taurin is derived from cystein or cystin. In certain pathological conditions cystui appears in the urine (cysfcinuria). The source of the chohc acid which goes to form the bile acids is unknown, but it has been surmised that it may be derived from cholesterin. Thus, [CH2.SH+3O CH2.SO2.OH CH2.SOa.OH CH.NH2 = CH.NH2 = CH.NH2 COOH COOH— CO2 Cystein. Cysteinic acid. Taurin. Traces of cholic acid, formed by hydrolysis from the bile-acids by the action of putrefactive bacteria, are found in the intestines, especially in the lower portion. Pettenkofer s test for bile-acids (Practical Exercises, p. 462), acciden- tally discovered in examining the action of bile upon sugar, depends upon three facts: (i) That cholic acid and furfuraldehyde give a purple colour when brought together; (2) that the bile-salts yield cholic acid when acted upon by sulphuric acid ; (3) that when cane-sugar is decomposed by strong sulphuric acid, furfuraldehyde is formed. Since a similar colour is given when the same reagents are added to a solution containing albumin, it is necessary to remove this, if present, from any liquid which is to be tested for bile-acids. Lecithin and cholesterin, or cholesterol, are by no means peculiar to bile (p. 4). They are very widely distributed in the body. Lecithin (C44^oNPO9) belongs to the group of phosphatides, fat-like phosphorus- containing substances present in all cells. It is a compound of glycerin with two molecules of fatty acid and one of phosphoric acid. The phosphoric acid is at the same time united with a base cholin (C6H15NO2) . The fatty acid (stearic, palmitic, oleic, etc.) varies in different varieties of lecithin . Heated with baryta-water, lecithin yields the corresponding fatty acid in the form of a soap, along with cholin and glyceryl-phos- phoric acid. Glyceryl-phosphoric acid can be further split so as to yield a molecule of glycerin and one of phosphoric acid. Cholesierin is a substance with the empirical formula C^I^gO. It contains an alcohol group in virtue of which fatty acids can be linked to it, forming esters. It is best obtained from white gall-stones, of which it is the chief, and sometimes almost the sole constituent (see Practical Exercises, p. 463). All the compounds related to cholesterin are grouped together under the name of sterins. The sterins are very widely distributed both in animals (zoosterins) and in plants (phytosterins) . Every cell seems to contain sterins and sterin esters (compounds of the same nature as the compounds of fatty acids with the alcohol glycerin which constitute the neutral fats) . In the vertebrates cholesterin and its product: constitute the chief, perhaps the only sterins, but in invertebrate animals and plants there is a much greater variety of these substances. The chief inorganic salts of bile are sodium chloride, sodium carbonate, and alkaline sodium phosphate. The phosphoric acid of the ash comes partly from the phosphorus of organic compounds (lecithin and bile- mucin), the sulphuric acid from the sulphur of taurocholic acid, the sodium largely from the bile-salts. Iron is a notable inorganic con- stituent of bile, although it exists only in traces, in the form of phosphate of iron. Manganese is also present in minute amount. 100 c.c. of fresh bile yields 50 to TOO c.c. of carbon dioxide, part of which is in solution and part combined with alkalies. THE CHEMlSTRy OF THE DIGESTIVE J VICES 367 The quantity of bile secreted in twenty-four hours in an average man is probably from 750 c.c. to a litre. In nine cases of fistula ol the gall-bladder in patients operated on for gall-stones or echino- coccus the daily quantity varied from 500 to 1,100 c.c. (Brand). Digestive Functions of Bile. — The great action of the bile in digestion is undoubtedly the preparation of the fats for absorption. In this preparation four processes are important: two chemical actions, hydrolysis of neutral fats to glycerin and fatty acids, and saponification, or the formation of soaps by the union of fatty acids with bases, especially sodium ; and two physical processes, emulsifi- cation, or the formation of a mechanical suspension of such fine globules of unaltered neutral fat as exist in milk, and solution of soaps and fatty acids. While there has been much discussion as to the relative share taken by these processes, and especially by saponi- fication and emulsification in the absorption of fat (p. 441), there is no doubt that they are all concerned in the digestion of fat or the preparation of it for absorption and assimilation. In this, indeed, the processes are complementary to each other, for an essential pre- liminary to emulsification in the intestine seems to be the format-ion of a certain amount of soaps, soluble in the intestinal contents, while the formation of an emulsion enormously increases the surface of contact between the unaltered fat and the digestive juices, and so favours more rapid hydrolysis, saponification, and solution. In the whole series of changes the bile plays a part, though not an indepen- dent one; it acts always in conjunction with the pancreatic juice. While no complete explanation has been given of the precise nature of this partnership, it is certain that the fat-splitting ferment of the pancreatic juice on the one hand, and the bile-salts on the other, contribute largely to the total action. An alkaline solution, a solution of sodium carbonate, e.g., is unable of itself to emulsify a perfectly neutral oil ; but if some free fatty acid be added, emulsifi- cation is rapid and complete (p. 12). Now, there is no doubt that here a soap is formed by the action of the alkali on the fatty acid, and there is equally little doubt that the formation of the soap is an essential part of the emulsification. But it is not clear in what manner the soap acts, whether by forming a coating round the oil- globules, or by so altering the surface-tension, or other physical properties of the solution in which it is dissolved, that they no longer tend to run together. However this may be, in pancreatic juice we have the two factors present which this simple experiment shows to be necessary and sufficient for emulsification ; we have a ferment which can split up neutral fats and set free fatty acids, and an alkali which can combine with those acids to form soaps. Accordingly, pancreatic juice is able of itself to form emulsions with perfectly neutral oils. It is possible that the protein constituents of pancreatic juice may have a share in emulsification, since the addition of protein 368 DIGESTION — e.g., egg-white — to a soap solution increases the stability of the emulsions formed by the soap. In bile, on the contrary, although the alkali is present, there is no fat-splitting ferment, and, according to the best experiments, bile alone has no emulsifying power on perfectly neutral fat. But we now come to a remarkable fact: this inert bile when added to pancreatic juice greatly intensifies its emulsifying action, and a solution of bile-salts has much the same effect as bile itself. The fact is undoubted, but the explanation is obscure. What it is that the bile or bile-salts can add to the pancreatic juice which so increases its power of emulsification, we do not know. It has been surmised that a characteristic physical property of bile, the diminution of the surface-tension of watery liquids to which it is added, may play an important part, perhaps, in enabling the fat-splitting ferments or the emulsifying soaps to get into closer contact with the unaltered fat. It is also true that bile, presumably in virtue of the chemical action of its alkaline salts, can, in presence of a free fatty acid, rapidly form an emulsion. But the pancreatic juice itself contains so considerable a quantity of sodium carbonate that it would scarcely seem to require the rela- tively feeble reinforcement of the alkaline salts of the bile.* An important part of the effect of the bile is certainly due to its favouring the fat- splitting action of the pancreatic juice. By the addition of bile, the quantity of fat split up by a definite amount of dog's pancreatic juice may be increased two to threefold. It has been shown that this is an action of the bile-salts. The sodium salts of synthetically-obtained glycocholic and taurocholic acids produce the same effect. It is in virtue of this action that the bile- salts are sometimes spoken of as the co-ferment of the lipase. As already pointed out, this action is exerted in presence of the fully formed enzyme, and should not be confounded with the effect of the bile-salts in activating the lipase zymogen. The capacity of dis- solving soaps, which is a property of the bile-salts, is also of great importance in supplementing the solvent power of the intestinal liquids for the products formed by the pancreatic juice. The solution of soaps in the bile- salts has the power in its turn of dis- solving free fatty acids. The significance of this in fat absorption will be referred to again. A further illustration of the mutual adaptation of the various digestive juices, of the remarkably precise manner in which the action of each dovetails into the action of others, is afforded by the facts already mentioned in connection with the lipase of the stomach. It is highly probable that the fatty acids formed by the gastric lipase, even if formed only in small amount, may exertan important influence in emulsifying the fat as soon as it enters the intestine. The intestinal juice itself also unquestionably takes a share in the digestion of fat along with the pancreatic * It has lately been shown that bile-salts accelerate the formation of soap from oleic acid by sodium bicarbonate (Kingsbury). THE CHEMISTRY OF THE DIGESTIVE JUICES 369 secretion and the bile. There exists also, as will be seen later on, a certain adaptation between the food and the digestive secretions. Not the best illustration of this, but one which suits the present topic, is the fact that the food itself probably always contains some free fatty acids when it contains fat at all. Although our knowledge of the mutual action of the pancreatic juice and the bile on the digestion of fats is still incomplete, there is no doubt that they are equally necessary. For in some diseases of the pancreas fat or fatty acid often appears in the stools, and this token of imperfect digestion of the fatty food may be confirmed by the wasting of the patient. The same may occur when the bile is prevented by obstruction of the duct or by a biliary fistula from entering the intestine. Yet in some cases of fistula, where there is every reason to believe that all the bile is escaping externally, the nutrition of the patient — at any rate, on a diet not abnormally rich in fat — is unaffected. The mere deficiency of bile in the intestine is, of course, complicated in obstructive jaundice by the harmful effects of the biliary constituents circulating in the blood. The white stools of jaundice owe their colour, not merely to the absence of bile -pigment, but also to the presence of fat. Their highly offensive odour used to be adduced as evidence that bile is the ' natural antiseptic ' of the intestine. It seems rather to be due to the coating of the particles of food with undigested fat, which shields the proteins from the action of the digestive juices, while permitting the putrefactive bacteria to revel in them unchecked. As a matter of fact, the bile itself has little, if any, power of hindering the growth of micro-organisms, although the free bile-acids are tolerably active antiseptics. In suckling children it is not uncommon to see the faeces white with fat. This is a less serious symptom than in adults, and perhaps betokens merely that the milk in the feeding-bottle is undiluted cow's milk, which is richer in fat than human milk, and ought to be mixed with water. Bidder and Schmidt found that the chyle in the thoracic duct of a normal dog contained 3-2 per cent, of fat. In a dog with the bile-duct ligatured the proportion fell to 0-2 per cent. It is an instance of the extraordinarily exact adaptation of the digestive juices to the nature of the food, the mechanism of which will present itself for discussion later on, that the reinforcing action of the bile upon the fat-splitting ferment of the pancreatic juice is said to be greater when the food is rich in fat (p. 414). Bile has been credited with a physical power of aiding the passage of fat through membranes moistened with it by diminishiiig the surface tension, and it has been inferred that this has an important bearing on the absorption of fat from the intestine. But the inference does not follow from the statement, and the statement has been itself denied. There is at present no evidence that the digestive function of the bile extends beyond the preparation of the food for absorption to the preparation of the mucosa for absorbing it. On proteins bile has either no digestive action, or only a feeble one. Fibrin is slightly digested by the bile of the dog and of man. But the addition of it to fresh pancreatic juice considerably increases the proteolytic power of that secretion (Rachford), although not so decidedly as in the case of the fat-splitting action. The amylolytic 370 DIGESTION action of the pancreatic juice is also favoured by the bile, and in about the same degree as its proteolytic effect. Although bile some- times exerts by itself a feebly amylolytic action, this is not to be included among its specific powers, for a diastatic ferment in small quantities is widely diffused in the body. The addition of bile or bile-salts to a gastric digest causes the precipitation of any unaltered native protein, acid-albumin, albu- mose, and pepsin. The precipitate, which is a salt-like compound of protein with taurocholic acid, is redissolved when the liquid is rendered alkaline, and therefore in excess of bile, or of a solution of bile-salts, but the pepsin has no longer any power of digesting proteins. Pait of the bile-acids and bile-mucin is also thrown down by the acid of the digest. It has been suggested that by thus precipitating the constituents of the chyme which have not been carried to the peptone stage bile prepares them for the action of the pancreatic juice. But it is difficult to see how the precipitation of a substance can prepare the way for its digestion, and it is more probable that if any physiological value is to be given to this reaction, it has the function of preventing the absorption of proteins which have not been sufficiently split up. There is little doubt, however, that the rendering of the pepsin inactive has physiological signifi- cance, for pepsin exerts an injurious influence upon the ferments of the pancreatic juice. In digestion, then, the bile has a twofold func- tion, favouring greatly the activity of the pancreatic ferments, especially the fat-splitting ferment, and aiding in establishing the conditions necessary for the transition of gastric into intestinal digestion. Succus Entericus. — This is the name given to the special secretion of the small intestine, which is supposed to be a product of the Lieberkuhn's crypts or intestinal glands. In order to obtain it pure, it is of course necessary to prevent admixture with the bile, the pan- creatic juice, and the food. This can be done by dividing a loop of intestine from the rest by two transverse cuts, the abdomen having been opened in the linea alba. The continuity of the digestive tube is restored by stitching the portion below the isolated loop to the part above it. One end of the loop is sewed into the lips of the wound in the linea alba, and the other being closed by sutures, the whole forms a sort of test-tube opening externally (Thiry's fistula). Or both ends are made to open through the abdominal wound (Vella's fistula). Another method is to make a single opening in the intes- tine, and by means of two indiambber balls, one of which is pushed down, and the other up through the opening, and which are after- wards inflated, to block off a piece of gut from communication with the rest. Or several openings may be made at different levels in the intestine, each being allowed to heal into a wound in the abdominal wall. When pure juice is required it is collected from the lower fistulae, while the upper fistulae are opened to permit the escape of the THE CHEMISTRY OF THE DIGESTIVE JUICES 371 secretions which enter the higher portions of the alimentary canal (gastric juice, pancreatic juice, and bile). The intestinal juice so obtained is a thin yellowish liquid of alkaline reaction, generally somewhat turbid from the presence of a certain number of leucocytes and epithelial cells. Its specific gravity is about 1010, the total solids about 1-5 per cent. It contains a small amount of proteins, including serum albumin and serum globulin, and about the same proportion of inorganic salts as most of the liquids and solids of the body, namely, 0-7 or 0-8 per cent., chiefly sodium carbonate and sodium chloride; but, like the other digestive liquids, it is adapted to the nature of the food, and therefore its composition is not quite constant. Like bile, intestinal juice acts but feebly on the food substances by itself, and if we contented ourselves with examining the pure and isolated secretion, we should greatly underestimate its importance. The sodium carbonate, in which it is relatively rich, will, to be sure, form soaps with fatty acids produced by the action of the pancreatic juice or of the fat-splitting bacteria in which the intestine abounds, and thus aid in the digestion of fats. A lipase, feebler than that of the pancreatic juice, or present in smaller con- centration, is also a constituent of the succus entericus. That a great deal of fat may be split up in the alimentary canal in the absence both of bile and pancreatic juice is well ascertained. The alkali of the succus entericus must at the same time aid in neutraliz- ing the original acidity of the chyme, and in preserving the proper reaction of the intestinal contents. A ferment called invertase, or sucrase — which is not introduced with the food or formed by bacterial action as has been suggested, since it occurs in the aseptic intestine of the new-born child — will invert cane-sugar. The ferments maltase and lactase will cause a corresponding change in maltose and lactose (see footnote, p. 356). It is worthy of remark that these inverting enzymes are present in the intestinal mucosa as well as in the intestinal juice, and extracts of the mucosa are usually distinctly more active than the juice itself. So that there is reason to believe that hydrolysis of the disaccharides may take place both in the lumen of the gut before absorption and in the wall of the gut during absorption. Inverting enzymes appear in the intestine early in embryonic life. Maltase is the most generally distributed of all these enzymes, and it is found along with lactase in the intestine of the embryo pig, while invertase is missing till after birth (Mendel). On native proteins and starch the isolated succus entericus has little or no action. But it contains a ferment, erepsin, which, although it does not affect native proteins like serum- and egg-albumin (fibrin and caseinogen may be slightly digested), exerts a powerful action on the first products of protein hydrolysis, albumoses, and peptones, breaking them up into bodies which no longer give the biuret re- action (ammonia, mono-amino acids, hexone bases, etc.). It 37* DIGESTION destroys the diphtheria toxin, which is also rendered innocuous by trypsin. Erepsin, however, is not specific to the secreted intestinal juice, for it occurs also, not only in the mucous membrane of the intestine, which, indeed, contains a greater quantity of it than the succus entericus, but in all animal tissues hitherto investigated. It is said even to be sometimes present in pancreatic juice, since in- activated pancreatic juice, which does not digest other proteins, will sometimes digest casein. But the matter is far from being settled, and the presence of erepsin in the pancreatic tissue is a complicating circumstance. For under abnormal conditions, most glands pro- vided with artificial fistulae have an increased liability to injuries of various kinds, which might permit constituents not normally present in the secretion to pass into it from the cells. The kidney in mammals is even richer in erepsin than the intestinal mucous membrane. Next to these come the pancreas, spleen, and liver, then at a long interval the heart muscle, while skeletal muscle and brain-tissue are poorest of all in the ferment. The intestinal mucosa varies in its erepsin content at different levels and on different diets. In cats on a meat or a mixed diet the duodenum is about five times richer in the ferment than the stomach. The ileum is about half as rich as the duodenum, and the jejunum occupies an intermediate position between the duodenum and ileum (Vernon). The secretion of Brunner's glands in the duodenum, which resemble in structure the pyloric glands of the stomach, digests coagulated albumin, although its proteolytic powers are feebler than those of the pan- creatic juice. Enterokinase. — The most characteristic constituent of succus entericus is a ferment, enterokinase, which differs from all the fer- ments we have hitherto described in acting not directly upon the foodstuffs, but upon the trypsinogen of the pancreatic juice, chang- ing it into the active enzyme trypsin. It may therefore be spoken of as a ferment of ferments. It has been previously stated that freshly secreted pancreatic juice is without action upon proteins. The addition of succus entericus immediately confers upon it a high degree of proteolytic power. In one experiment pancreatic juice, obtained by a temporary fistula, required four to six hours to dissolve fibrin, and did not attack coagulated albumin even in ten hours. On addition of succus entericus, the same pancreatic juice dissolved fibrin in three to seven minutes, and rapidly digested coagulated albumin (Pawlow). In like manner a glycerin extract of a fresh pancreas has hardly any effect on proteins; a similar extract of a stale pancreas is active. The fresh pancreas contains trypsinogen, which is soluble in glycerin, for the inert extract becomes active when it is treated with dilute acetic acid, or even when it is diluted with water and kept at the body-temperature. If the fresh pancreas be first treated with dilute acetic acid, and then with glycerin, the THE CHEMISTRY OF THE DIGESTIVE JUICES 373 extract is at once active. The trypsinogen can therefore be activated within the pancreatic cells, gradually when the pancreas is simply allowed to stand after excision, more rapidly in presence of the dilute acid. The ordinary tests for ferment action (destruction by boiling, activity in very small amounts, etc.) have shown that this property of the intestinal juice is due to a ferment, although it differs in certain respects from most ferments — -for instance, in requiring a relatively high temperature to inactivate it. The smallest trace of enterokinase will convert a large quantity of trypsinogen into trypsin if time be given. At the same time, although to a much smaller extent, the fat-splitting and starch-digesting activity of the pancreatic juice is increased. The secretion of the duodenum causes a greater increase in the proteolytic power than that of the other portions of the small intestine, while no such difference has been made out in the case of the amylolytic and lipolytic functions. It is probable that the enterokinase, which is secreted mainly in the upper two -sevenths of the small intestine, and solely by the intestinal epithelium, acts only on the trypsinogen, and that the amylopsin and steapsin are aided in some other way. Enterokinase is only found in the intestinal juice when pancreatic juice is present in the gut. It is therefore secreted in response to the presence of tryp- sinogen or of some other constituent of the pancreatic juice. Delezenne has attempted to explain the interaction of enterokinase and trypsinogen as an adaptive phenomenon of the same kind as the formation of antitoxins and hsemolysins (p. 31). According to him, enterokinase acts like a complement in haemolysis, while trypsinogen plays the part of an intermediary body or amboceptor which enables the enterokinase to attack the protein molecule. He asserts that enterokinase, or a substance which produces a similar effect on tryp- sinogen, is contained not only in the mucous membrane of the intestine, but also in leucocytes, in fibrin (one of whose properties it is to pick out ferments from liquids containing them), in lymph-glands, in snake venom, and even in certain anaerobic bacteria. On this view trypsin would not be a definite substance produced by the interaction of enterokinase and trypsinogen, but only an expression for these two bodies acting together. Strong evidence against this view, and in favour of the independent existence of trypsin, haj been brought forward by Bayliss and Starling, and it does not seem to merit further con- sideration. According to Mellanby, enterokinase is really a proteolytic ferment, and trypsinogen contains a protein moiety with which trypsin is firmly combined. The conversion of trypsinogen into trypsin depends on the digestion of this protein moiety, and the consequent liberation of trypsin. Vernon has put forward the view that, while enterokinase starts the activation of trypsinogen in the intestine,- and can no doubt in time complete it, the .trypsin a.s it is formed aids in the activation of more trypsinogen to trypsin, and so on by a process of so-called auto-catalysis of the trypsinogen. This idea can be har- monized with Mellanby's conception by assuming that the trypsin formed from trypsinogen can itself digest the protein moiety of a further portion of trypsinogen. 374 DIGESTION According to Pawlow, the reason why the trypsin is not secreted in the active form is that active trypsin readily destroys the amylo- lytic and lipolytic ferments. In the intestine, where trypsin is rendered active by enterokinase, these ferments are protected from its attack by the proteins of the food and by the bile. Enterokinase is itself immediately destroyed in the presence of free acid (centi- normal hydrochloric acid). Having now finished our review of the chemistry of the digestive juices, our next task is to describe what is known as to their secre- tion— the nature of the cells by which it is effected and their histo- logical appearance in activity and repose, and the manner in which it is called forth and controlled. SECTION IV. — THE SECRETION OF THE DIGESTIVE JUICES — MICROSCOPICAL CHANGES IN THE GLAND CELLS. The digestive glands are formed originally from involutions of the mucous membrane of the alimentary canal, the salivary glands from the ectoderm, the others from the endoderm (Chap. XIX.). Some are simple unbranched tubes, in which there is either no distinction into body and duct, as in Lieberkuhn's crypts in the intestines, or in which one or more of the tubes open into a duct, as in the glands of the f undus of the stomach. Some are branched tubes, several of which may end in a common duct ; such are the glands of the pyloric end of the stomach and the Brunner's glands in the duodenum. In others the main duct ramifies into a more or less complex system of small channels, into each of the ultimate branches of which one or more (usually several) of the secreting tubules or alveoli open. The salivary glands and the pancreas belong to this class of compound tubular or racemose glands, and so does the liver of such animals as the frog. But in the latter organ the typical arrangement is obscured in the higher vertebrates by the pre- dominance of the portal bloodvessels over the system of bile-channels as a groundwork for the grouping of the cells. In every secreting gland there is a vascular plexus outside the cells of the gland-tubes, and a system of collecting channels on their inner surface ; and in a certain sense the cells of every gland are arranged with reference to the bloodvessels on the one hand, and the ducts on the other. But in the ordinary racemose gland the blood-supply is mainly required to feed the secretion ; the cells of the alveoli have either no other function than to secrete, or if they have other fxmctions, they are* not such as to entail a great disproportion between the size of the cells and the lumen of the channels into which they pour their products. For both reasons the relation of the grouping of the cells to the duct- system is very obvious, to the blood-system very obscure. In the liver, the conditions arc precisely reversed. We cannot suppose that the manufacture of a quantity of bile less in volume than the secretion of the salivary glands, though doubtless containing far more solids, requires an immense organ like the liver, and a tide of blood like that which passes through the portal vein. And, as we shall see, the liver has other functions, some of them certainly of at least equal importance with the secretion of bile, and one of them evidently requiring from its very nature a bulky organ. Accordingly, both the richness of the blood-supply and the size of the secreting cells are out of proportion to THE SECRETION OF THE DIGESTIVE JUICES 375 the calibre of the ultimate channels that carry the secretion away. The so-called bile-capillaries, which represent the lumen of the secreting tubules, are mere grooves in the surface of adjoining cells; and the architectural lines on which the liver lobule is built are: (i) the inter- lobular veins which carry blood to it; (2) the rich capillary network which separates its cells and feeds them; and (3) the central intra- lobular vein which drains it. Thus a network of cells lying in the meshes of a network of blood-capillaries takes the place of a regular dendritic arrangement of ducts and tubules; and in accordance with this the bile-capillaries, instead of opening separately into the ducts, form a plexus with each other within the hepatic lobule (see also foot- note, p. 14). The ducts and secreting tubules of all glands are lined by cells of columnar epithelial type, but the type is most closely preserved in the ducts. In none of the digestive glands is there more than a single complete layer of secreting cells. But the alveoli of the mucous salivary glands show here aiul there a crescent-shaped group of small deeply-staining cells (crescents of Gianuzzi) outside the columnar layer (Fig. 158, A", B"), and between it and the basement membrane, while the gland-tubes of the f undus of the stomach have in the same situation a discontinuous layer of largo ovoid cells, termed parietal from their position, oxyntic (or acid -secreting) from their supposed function (Figs. 155-157). Access to the lumen of the glands is provided for these deeply-placed parietal cells and for the cells of the crescents by fine branching channels which enter and surround the cells. The serous salivary glands, the pyloric glands of the stomach, and the Lieberkiihn's crypts, have but a single layer of epithelium; and since there is no hepatic cell which is not in contact with at least one bile capillary, the liver may be regarded as having no more. The same is true of the pancreatic alveoli, except that in the centre of many of the acini a few spindle-shaped cells (centro-acinar cells), apparently con- tinued from the lining of the smallest ducts, may be seen. Remarkable histological changes, evidently connected with changes in functional activity, have been noticed in most of the digestive glands. In dis- cussing these, it will be best to omit for the present any detailed reference to the liver, since, although there are histological marks of secretive activity in this gland as well as in others, and of the same general character, they are accompanied, and to some extent overlaid, by the microscopic evidences of other functions (p. 534). The serous salivary glands and the pancreas can be taken together; so can the glands of the various regions of the stomach; the mucous salivary glands must be considered separately. Changes in the Pancreas and Parotid during Secretion.— The cells of the alveoli of the pancreas or parotid during rest, as can be seen by examining thin lobules of the former between the folds of the mesentery in the living rabbit, or fresh teased preparations of the latter, are filled with fine granules to such an extent as to obscure the nucleus. In the parotid the whole cell is granular, in the pancreas there is still a narrow clear zone at the outer edge of the cell which contains few granules or none; in both, the divisions between the cells are very indistinct, and the lumen of the alveolus cannot be made out. During activity the granules seem to be carried from the outer portion of the cell towards the lumen, and 376 there discharged. The clear outer zone of the pancreatic cell grows broader and broader at the expense of the inner granular zone, until at last the granular zone may in its turn be reduced to a narrow contour line around the lumen (Fig. 153). In the uniformly clouded parotid cell a similar change takes place; a transparent outer zone arises; and, after prolonged secretion, only a thin edging of granules may remain at the inner portion of the cell (Fig. 154). In both glands the outlines of the cells be- come more clearly indicated, and a distinct lumen can now be recognized. The cells are smaller than they A B are during rest. A , alveolus of rabbit's pancreas, (resting) ; B, ' discharged ' Fig. 153-- ' loaded (active), observed in the living animal (Kiihne and Lea). The disappearance of granules from without in- wards during activity sug- gests that these are manu- factured products eliminated in the secretion, and they are generally spoken of as zymogen granules. Bensley, who has made a careful study of the pancreas in the guinea-pig, has been able to distinguish, even in fresh preparations examined in the animal's own serum, but better after staining with such a dye as neutral red, another kind of granules, which he regards Fig. 154. — Alveoli of Parotid Gland: A, at Rest; B, after a Short Period of Activity; C, after a Prolonged Period of Activity (Fresh Preparations) (Langley). In A and B the nuclei are obscured by the granules of zymogen. as zymogen granules in the course of formation, and therefore designates prozymogen granules. The resting acini show a clear basal zone which is unstained, and a zone next the lumen containing coarse zymogen granules which are faintly stained. In the active gland — e.g.. after a meal or after the injection of secretin (p. 407) — prozymogen granules which stain much more intensely than the zymogen granules with neutral red make their appearance between the zymogen granules, now much reduced in number and size, and THE SECRETION OF THE DIGESTIVE JUICES 377 the clear outer zone. After prolonged secretion the zymogen granules may be entirely absent from the cells, and only a narrow rim of prozymogen granules can be seen around the lumen. In one respect the pancreas differs remarkably from the salivary glands — namely, in the presence of the islets of Langerhans — characteristic groups of small polygonal cells, richly sup- plied with bloodvessels, but not arranged in the form of alveoli. Some observers state that they are remarkably in- creased in size, and even in number when the pancreas is caused to secrete actively by repeated injections of secretin, and also in starvation. But it has been shown that this conclusion was based upon faulty methods of counting the islets, and even of identi- fying the islet cells. There appears to be no foundation for the view that they are derived from the ordinary secreting cells, and that they can, in turn, give rise to new alveoli by a process of pro- liferation. It is far more probable that they are inde- pendent structures, with a different function from the pancreatic alveoli (p. 6^,8). Changes in the Glands of the Stomach during Secre- tion.— The mucous membrane of the stomach is covered with a single layer of colum- nar epithelium, largely con- Fig. 156.— A Fundus Gland prepared by sisting of mucigenoUS goblet- GolSi>s Method, showing the Moda of Com- ,, J T ° j A ,° . , munication of the Parietal Cells with the IS Studded Wltn Gland-Lumen (Schafer, after E. Muller). minutt pits, into which open the ducts of the peptic and pyloric glands, the ducts being lined with cells just like those of the general gastric surface. Three varieties of gastric glands have been distinguished: (i) The glands of the cardia. In man these occupy a small portion of the mucous membrane at the cardiac end, near the orifice of Fig. 155. Fig. 156. Fig. 155. — A Fundus Gland of Simple Form from the Bat's Stomach (Osmic Acid Pre- paration) (Langley). c, Columnar epithe- lium of the surface; n, neck of the gland with chief or central and parietal cells; /, base, occupied only by chief cells, which show the granules accumulated towards the lumen of the gland. 378 DIGESTION the oesophagus. Some of the glands are single tubules, but others have two or more tubules opening into a common duct. Both are lined by a single layer of short columnar epithelium, which contains granules. (2) The glands of the pyloric canal or antrum. These consist of short, branched tubules, which open by twos and threes into long ducts. (3) The glands of the fundus or oxyntic glands, occupying the intermediate and greater portion of the organ. The gland tubules are long and seldom branched, and the ducts, into each of which open from one to three tubules, are relatively short. The secreting parts of both kinds of glands are lined by short columnar granular cells; and in the pyloric tubules no others are present. But, as we have said, in the glands of the fundus there are besides large ovoid cells scattered at intervals like beads between the basement membrane and the lining or chief cells. The cells of the pyloric glands have a general resemblance to the chief cells of the fundus glands, but they are not quite the same. For example, the granules are less distinct in the pyloric glands. In the human stomach it is only quite near the pylorus that the parietal cells disappear altogether. The parietal cells also contain granules, but they are smaller and less numerous than those of the chief cells, so that the deeper portions of the fundus glands are much darker in appearance than the more superficial portions, since the oxyntic or parietal cells are more numerous in the neighbourhood of the ducts (Bensley). The histological changes connected with gastric secretion do not differ essentially from those described in the pancreas and the parotid, but there is much greater difficulty in making observations on the living, or at least but slightly altered, cells. For the mammal the best method is to use animals with a permanent gastric fistula, and to remove from time to time small portions of the mucous membrane for examination in the fresh condition. During digestion the granules disappear from the outer part of the chief cells of the fundus glands, leaving a clear zone, the lumen being bordered by a granular layer. Or, more rarely, there may be a uniform decrease in the number of granules throughout the cell. The total volume of the cell is less than in the fasting condition. The parietal cells, which are small in the fasting animal, swell up, so as to bulge out the membrana propria. They reach their maximum size (in the dog) very late in digestion (the thirteenth to the fifteenth hour). No such definite changes in their contents as those observed in the other cells have been made out. The granules in the ovoid cells during and after activity seem to be as large and as numerous as in the resting cell, or even larger. After sham feeding in dogs the histo- logical changes in the gastric glands are very slight, even when con- siderable amounts of gastric juice have been secreted (Noll and Sokoloff). THE SECRETION OF THE DIGESTIVE JUICES 379 The chief cells of the oxyntic, and the similar if not identical cells of the pyloric glands, are believed to manufacture the pepsin-form- ing substance. The ovoid cells of the former are supposed to secrete the hydrochloric acid. The evidence on which this belief is based is as follows: The glands of the antrum pylori in the dog, in which in most situations no ovoid cells are to be seen, secrete pepsin, but no acid. The pyloric end of the stomach or a portion of it has been isolated, Fig- I57- — The Gastric Glands (Ebstein). On the left, oxyntic; right, pyloric. the continuity of the alimentary canal restored by sutures, and the secretion of the pyloric pocket collected. It was found to be alka- line, and contained pepsin. The glands of the frog's oesophagus, which contain only chief cells, secrete pepsin, but no acid. It seems fair to conclude that the chief cells of the fundus glands in the mammal secrete none of the free hydrochloric acid, but certainly some of pepsin. But it does not follow that all the pepsin is formed by these cells, although it would seem that all the hydrochloric acid 380 DIGESTION must be secreted by the only other glandular elements present, the parietal or ' border ' cells. And, indeed, the glands in the fundus of the frog's stomach, which are composed only of ovoid cells, whilst secreting much acid, also form some pepsin, although far less than the cesophageal glands. During winter sleep (in the marmot) the production of hydrochloric acid in the parietal cells stops altogether, while the chief cells continue to accumulate granules of pepsinogen. That some pepsin is secreted by the pyloric end of the stomach is not difficult to prove. The secretion collected from the isolated pyloric portion is, indeed, like the secretion of the Brunner's glands in the duodenum, quite unable to digest protein until dilute hydro- chloric acid is added. But this is because in both cases the juice as it flows from the glands is slightly alkaline, and, as we have already seen, pepsin only acts in the presence of an acid. The milk-curdling action of these two juices also unfolds itself only when the secretions are first acidulated, and later on again neutralized ; in other words, the ferments must be activated by the addition of an acid. In normal digestion the pepsin of the (in itself) alkaline secretion of the pyloric end of the stomach becomes a constituent of the acid gastric juice; and it may perhaps be considered a morphological accident, so to speak, that the oxyntic cells of the fundus should mingle their acid products with the (presumedly) alkaline secretion of the chief cells in the lumen of each gland-tube, instead of being massed as a separate organ with a special duct. We are not without other examples of digestive juices fitted or destined to act in a medium with an opposite reaction to their own. The ' saliva ' of the cephalopod Octopus macropus, strongly acid though it is, contains a proteolytic ferment which in vitro acts, like trypsin, better in an alkaline than in an acid solution. And trypsin, whose precursor is a constituent of the alkaline pancreatic juice, while the enterokinase which activates it is a constituent of the alkaline succus entericus, performs a part at least of its work in an acid medium. Attempts made to demonstrate an acid reaction in the border cells have hitherto failed. Harvey and Bensley on the basis of experiments with dyes (cyanimin and neutral red), which give different colours according to whether the reaction is acid, alkaline, or neutral, have concluded that free acid exists only on the internal surface of the stomach, or at most also in the necks of the glands. The parietal cells they find alkaline. They suggest that these cells form in some way, of course ultimately from the chlorides of the blood, a chloride of an organic base which does not decompose so as to yield free hydrochloric acid until it reaches the mouth of the gland. The nature of this decom- position, if it occurs, is unknown. It may be mentioned, although only as a matter of historical interest, that some observers have denied that the acid is secreted in the depths of any cell from the chlorides of the blood, and have asserted that it is formed at the surface of contact of the stomach-wall with the gastric contents from the sodium chloride of the food by an exchange of sodium ions (p. 428) for hydrogen ions from THE SECRETION OF THE DIGESTIVE JUICES 381 the blood or lymph. It was pointed out in favour of this view that when, instead of sodium chloride, sodium bromide is given in the food, the hydrochloric acid in the stomach is to a large extent replaced by hydrobromic acid. And it was argued that this cannot be due to the decomposition of the bromide by hydrochloric acid, since it occurs in animals deprived for a considerable time of salts, and in ' salt-hunger ' the stomach contains no acid (Koeppe). It may be, however, that even in ' salt -hunger ' the presence of sodium bromide in the stomach stimulates the secretion of hydrochloric acid, which then decomposes the bromide, with the formation of hydrobromic acid. The sodium chloride formed in the double decomposition might be re-absorbed, and the stock of chlorides in the blood remain undiminished. It is in any case a decisive objection to this now defunct theory that a copious secretion of gastric juice, containing hydrochloric acid in abundance, can be obtained, without the introduction of any food into the stomach, either by the process of sham feeding (p. 401) or by psychical simulation of the gastric glands when food is shown to an animal. Changes in Mucous Glands during Secretion. — In the mucous salivary and other mucous glands similar, but apparently more complex, changes occur. During rest the cells which line the lumen may be seen in fresh, teased preparations to be filled with granules or ' spherules.' After active secretion there is a great diminution in the number of the granules. Those that remain are chiefly collected around the lumen, although some may also be seen in the peripheral portion of the cell; and there is no very distinct differentiation into two zones. That a discharge of material takes place from these cells is shown by their smaller size in the active gland. That the material thus discharged is not protoplasmic is indicated by the behaviour of the cells to proto- plasmic stains such as carmine. The resting cells around the lumen stain but feebly, in contrast to the deep stain of the demilunes, while the discharged cells take on the carmine stain much more readily. Further, when a resting gland is treated with various reagents (water, dilute acids, or alkalies), the granules swell up into a transparent sub- stance identical with mucin, which fills the meshes of a fine protoplasmic network. In ordinary alcohol-carmine preparations only the network and nucleus are stained ; the nucleus, small and shrivelled, is situated close to the outer border of the cell. When a discharged gland is treated in the same way there is proportionally more ' protoplasm ' (or ' bioplasm ') and less of the clear material, what remains of the latter being chiefly in the inner portion of the cell, while the nucleus is now large and spherical, and not so near the basement membrane (Fig. 158). Everything, therefore, points to the granules in what we may now call the mucin-forming cells as being in some way or other precursors of the fully-formed mucin ; manufactured during ' rest ' by the proto- plasm and partly at its expense, moved towards the lumen in activity, discharged as mucin in the secretion. It has been asserted that not only is the protoplasm lessened in the loaded cell and re- newed after activity, but that many of the mucigenous cells may be altogether broken down and discharged, their place being supplied by proliferation of the small cells of the demilunes. This conclusion, however, is not supported by sufficient evidence. The cells of the crescents contain fine granules, but none which can be changed into 382 DIGESTION nmcin. They are of serous and not of mucous type. But the fact on which we would specially insist is that the granules of the resting mucigenous cell may be looked upon as a mother-substance from which the mucin of the secretion is derived ; they are not actual, but potential, mucin. So in the pancreas, the serous or albuminous salivary glands, and the glands of the stomach, there is every reason to believe that the granules which appear in the intervals of rest, and are moved towards the lumen and discharged during activity, are the pre- cursors, the mother-substances, of important constituents of the secretion. These granules are sharply marked off from the proto- plasm in which they lie and by which they are built up. By every mark, by their reaction to stains, for instance, they are non-living substance, formed in the bosom of the living cell from the raw material which it culls from the blood, or, what is more likely, formed from its own protoplasm, then shed out in granular form and Fig. 158. — Mucous Cells (from Submaxillary of Dog) in Rest and Activity (Langley). A, B, fresh; A', B', after treatment with dilute acetic acid; A", B*, alveoli hard- ened in alcohol and stained with carmine. A, A', and A* represent the loaded; B, B', and B", the discharged condition. secluded from further change. The granules in the ferment-forming glands are not in general composed of the actual ferments, and, indeed, in several instances it has been shown that the actual fer- ments are not present in the secreting cells at all. We have already seen that the pancreas and even the fresh pan- creatic juice are devoid of active trypsin. Similarly, a glycerin extract of a fresh gastric mucous membrane is inert as regards proteins, or nearly so. But if the mucous membrane has been pre- viously treated with dilute hydrochloric acid, the glycerin extract is active, as is an extract made with acidulated glycerin. Here we must assume the existence in the gastric glands of a mother-sub- stance, pepsinogen, from which pepsin is formed. The rennin of the gastric juice, which is formed in the chief cells, also has a precursor, which, if the ferment is identical with pepsin (p. 353), must be pepsinogen. The proteolytic power of an extract of the pancreas, THE SECRETION OF THE DIGESTIVE JUICES 3.83 when the trypsinogen has been activated into trypsin, or of the gastric mucous membrane, when the pepsinogen has been changed into pepsin, seems to be, roughly speaking, in proportion to the quantity of granules present in the cells. Therefore it is concluded that the granules represent mother-substances of the ferments or zymogens. Some observers believe they have obtained evidence of stages in the elaboration of the ferments still further back than the mother-substances, grandmother-substances so to speak, or pro- zymogens. Bensley, e.g., concludes that the nuclei of the chief cells in the fundus glands of the stomach take part in the formation of a prozymogen, the precursor of the zymogen or pepsinogen, as pepsino- gen is the precursor of the enzyme pepsin. A glycerin or watery extract of the salivary glands always con- tains active amylolytic ferment, if the natural secretion is active. So that if ptyalin is preceded by a zymogen in the cells, it must be very easily changed into the actual ferment. But we should greatly deceive ourselves if we supposed that granules of this nature in gland-cells are necessarily related to the production of ferments. The mucigenous granules have no such significance. Most digestive secretions contain protein constituents, with which the granules may have to do as well as with ferments. And bile, a secretion which contains no mucin, no proteins, and either no ferments or mere traces, as essential and original constituents, is formed in cells with granules so disposed and so affected by the activity of the gland as to suggest some relation between them and the process of secretion. In the liver-cells of the frog, in addition to glycogen, and oil-globules small granules may be seen, especially near the lumen of the gland tubules ; they diminish in number during digestion, when the secretion of bile is active and increase when food is withheld and secretion slow. And in fasting dogs the secreting cells of Brunner's glands, the pyloric glands and the pancreas, as well as the lining epithelium of the bile-ducts, have been found to contain many fatty granules. Possibly some of these represent the fat which is known to be excreted into the alimentary canal (pp. 443, 444). The Nature of the Process by which the Digestive Secretions are Formed. — We have spoken more than once of the gland-cells as manufacturing their secretions. It is an idea that rises naturally in the mind as we follow with the microscope the traces of their functional activity. And when we compare the composition of the digestive juices with that of the blood-plasma and lymph, the suggestion that the glands which produce them are not merely passive niters, but living laboratories, acquires additional strength. It is evident that everything in the secretion must, in some form or other, exist in the blood which comes to the gland, and in the lymph which bathes its cells. No glandular cell, if we except the leucocytes, which in some respects are to be considered as unicellular glands, dips directly into the blood ; everything a gland-cell receives must pass through the walls of the bloodvessels. (But see footnote 384 DIGESTION on p. 14). So that anything which we find in the section and do not find in the blood must have been elaborated by the gland epithelium (or by the capillary endothelium) from raw material brought to it by the blood. Take, for example, the saliva or gastric juice. These liquids both contain certain things that also exist in the blood, but in addition they contain certain things specific to themselves: mucin in saliva, hydrochloric acid in gastric juice, ferments in both. It is true that a trace of 'pepsin and a trace of a diastatic ferment may be dis- covered in blood; but there is no reason whatever to believe that this is the source of the pepsin of the gastric juice, or the ptyalin of the salivary glands, except, perhaps, in animals like the cat, whose saliva contains a diastase in still smaller concentration than the serum (Carlson). On the contrary, it is possible that the fer- ments of the blood may be in part absorbed from the digestive glands, the rest being formed by the leucocytes and liberated when they break down. Formation of Bile. — The liver affords an even better example of this ' manufacturing ' activity of gland-cells, and many facts may be brought forward to prove that the characteristic constituents of the bile, the bile-pigments and bile-acids, are formed in the liver, and not merely separated from the blood. Bile-pigment has indeed been recognized in the normal serum of the horse, and bile-acids in the chyle of the dog, but only in such minute traces as are easily accounted for by absorption from the intestine. Frogs live for some time after excision of the liver, but no bile-acids are found in the blood or tissues. But if the bile-duct be ligatured, bile-acids and pigments accumulate in the body, being absorbed by the lymphatics of the liver (Ludwig and Fleischl). If the thoracic duct and the bile-duct are both ligatured in the dog, no bile-acids or pigments appear in the blood or tissues. Wertheimer and Lepage state that bile or bilirubin injected into a bile-duct appears sooner in the urine than in the lymph of the thoracic duct, and therefore conclude that the bloodvessels are the most important channel of absorption. This conclusion, however, cannot be accepted until it is shown that in these experiments the injection did not cause rupture of some of the hepatic capillaries and direct entrance of the bile-pigment into the blood. It is not improbable that the pressure attained by the bile in the bile-capillaries is a factor in determining the path by which it is absorbed, and that when the pressure rises beyond a certain limit it may pass both into the bloodvessels and into the lymphatics. In mammals life cannot be maintained for any length of time after ligature of the portal vein, since this throws the whole intestinal tract out of gear. But after an artificial communication has been made between the portal and the left renal vein or the inferior cava, the portal may be tied and the animal live for months THE SECRETION OF THE DIGESTIVE JUICES 385 (Eck). The liver can- now be completely removed, but death follows in a few hours. A good method of establishing an- Eck's fistula i's to make a longitudinal incision in the inferior vena cava and the portal or superior mesenteric vein, and to suture the edges of the two openings together with a very fine sewing-needle and thread (Carrel and Guthrie). In birds there exists a communicating branch between the portal vein and a vein (the renal-portal) which passes from the posterior portion of the body to the kidney, and there breaks up into capillaries ; and not only may the portaL be tied, but the liver may be completely destroyed without immedi- ately killing the animal. In the hours of life that still remain to it no accumulation of biliary substances (acids or pigments) takes* place in the blood or tissues. A further indication that bile-pig- ment is produced in the liver is the fact that the liver contains iron in relative abundance in its cells (p. 21), and eliminates small quantities of iron in its secretion. Now, bile-pigment, which con- tains no iron, is certainly formed from blood-pigment, which is rich in iron. For haematin, when injected under the skin, has been found to appear almost quantitatively in the form of bile-pigment in the bile, and haematoidin (Fig. 159), a crystalline derivative of haemoglobin found in old ex- travasations of blood, especially in the brain and in the corpus luteum, is identical with bilirubin. The fact that one of the derivatives of haematin, haematoporphyrin (C33H38N4O6), contains no iron, and is probably nearly related to bilirubin (C32H36N4O6), suggests that haematoporphyrin may be an inter- pig I59Z^raatoidin. mediate step in the formation of bile-pigment from blood- pigment. In any case, the seat of formation of bile- pigment might be expected to be an organ peculiarly rich in iron. The existence of haematoidin, however, shows that bile-pigment may, under certain conditions, be formed outside of the hepatic cells. The occurrence of biliverdin in the placenta of the bitch points in the same direction. But the pathological evidence in favour of the pre-formation of the biliary constituents tends rather to shrink than to increase. For many cases of what used to be considered ' idiopathic ' or ' haematogenic ' jaundice, i.e., an accumu- lation of bile-pigments and bile-acids in the tissues, due to defective elimination by the liver, are now known to be caused by obstruction of the bile-ducts and consequent re-absorption of bile (' obstructive ' or ' hepatogenic ' jaundice). But if substances such as the ferments, mucin, hydrochloric acid, the bile-salts and bile-pigments, are undoubtedly manufactured in the gland-cells, it is different with the water and inorganic salts which form so large a part of every secretion. No tissue lacks them ; no 25 386 DIGESTION physiological process goes on without them ; they are not high and special products. As we breathe nitrogen which we do not need because it is mixed with the oxygen we require, the secreting cell passes through its substance water and salts as a sort of by-play or adjunct to its specific work. But this is not the whole truth. The gland-cell is not a mere filter through which water and salts pass in the same proportions in which they exist in the liquids that the cell draws them from. When, e.g., the salivary glands secrete against the resistance of an abnormally high pressure in the ducts, the percentage of salts in the saliva increases. The secretions of different glands differ in the nature, and especially in the relative proportions, of their inorganic constituents. They differ also in their osmotic pressure and electrical conductivity, which depend so largely upon those constituents, notwithstanding the fact that the osmotic pressure and conductivity of the blood-serum (p. 26) vary only within narrow limits. Even the secretion of one and the same gland is by no means constant in this respect, as we shall have to note more especially when we come to deal with the in- fluence of the nervous system on secretion (p. 395). The following tables illustrate this point : Dog. Blood-Serum.* Filtrate of Gastric Contents. At KJ(5°C.)xio«. A KI(5°C.)xio«. I. II. III. 0-643° 0-628° O'6O2° 92-0 87-6 87-7 0-5850 0-585° 0-642° 312-5 179-4 351-7 Vomit of man with complete intes- tinal obstruction - 0-433° 84-7 Pancreatic Juice of Dog (Pincussohn) . Diet. A Milk o-57°— < j-630 Cauliflower 0-58°— < 3-63° Horseflesh 0-62°— < >-63° * The blood and gastric contents were obtained from the animals twenty- four hours after the last meal. f The depression of the freezing-point below that of distilled water. J See footnote on p. 27. The number in brackets is the temperature at which the measurement was made. THE SECRETION OF THE DIGESTIVE JUICES 387 Gastric Juice from Miniature Stomach in a Dog in Different Experiments (Bickel) . Milk Diet. Meat Diet. A K(25oC.)Xio*. A K(2s°C.)xio«. 0-52° 195*9 0-60° 310-3 0-65° 402'6 O-yi0 473-5 0-64° 436-5 1-21° 483*3 0-69° 404-2 o-79° 5i4'i 0-81° 436-5 0-70° SH'i A of Blood and Saliva Compared (Jappelli}. A of Blood. A of Submaxillary Saliva of Dog. 0-570° 0-610° 0-600° 0-590° 0-580° 0-605° 0-650° 0-610° 0-410° 0-350° 0-430° 0-410° 0-450° 0-425° 0-380° 0-475° A of Human Fistula Bile. A of Human Bladder Bile. 0-56° 0-547° 0-615° 0-60° 0-545° 0-65° 0-865« 0-78° 0-92° A of Dog's Submaxillary Saliva. Chorda stimulated : Left submaxillary - 0-293* Both glands - 0-408° Spontaneous secretion : Right submaxillary - - - -0-195° A of dog's serum - - 0-590° The protein substances, such as serum-albumin and globulin, common to blood and to some of the digestive secretions, take a middle place between the constituents that are undoubtedly manu- 338 DIGESTION factured in the cell and those which seem by a less special and laborious, though a selective, process to be passed through it from the blood. Their practical absence from bile, and, as we shall see, from urine, their relative abundance in pancreatic and scantiness in gastric juice, point to a closer dependence upon the special activity of the gland-cell than we can suppose necessary in the case of the salts. Although it is in the cells of the digestive glands that the power of forming ferments is most conspicuous, it is by no means confined to them. It seems to be a primitive, a native power of protoplasm. Lowly animals, like the amoeba, lowly plants, like bacteria, form ferments within the single cell which serves for all the purposes of their life. TTie ferment-secreting gland-cells of higher forms are perhaps only lop- sided amoebae, not so much endowed with new properties as dispro- portionately developed in one direction. The contractility has been lost or lessened, the digestive power has been retained or increased; just as in muscle the power of contraction has been developed, and that of digestion has fallen behind. The muscle-cell and the cartilage- cell are parasites, if we look to the function of digestion alone. They live on food already more or less prepared by the labours of other cells; and it is a universal law that in the measure in which a power becomes useless it disappears. But the presence of pepsin in the white blood- corpuscles, the parasites as well as the scavengers of the blood, and of amylolytic, proteolytic and lipolytic ferments in many tissues, should warn us not to conclude that the power of forming ferments belongs exclusively to any class of cells. There is good and growing evidence that food-substances absorbed from the blood are further decomposed and, in turn, elaborated by ferment action within the tissues them- selves ; while many facts show that the power of contraction is widely diffused among structures whose special function is very different, and a few point to its possession in some degree even by glandular epithelium. On the other hand, it must be remembered that none of the digestive glands absorb food directly from the alimentary canal to be then digested within their own cell-substance; the ferments which they form do their work outside of them ; their cells feed also upon the blood. Why are the Tissues of Digestion not affected by the Digestive Ferments ? — This is the place to mention a point which has been very much debated. Why is it that the stomach or the small intes- tine does not digest itself ? This is really a part of a wider question : Why is it that living tissues resist all kinds of influences, which attack dead tissues with success ? And we have to inquire whether the immunity of the alimentary canal to the digestive juices is an example of a general resistance of all living tissues to destructive agencies, or a specific resistance of certain tissues to certain in- fluences. That all living tissues cannot withstand the action of the gastric juice has been shown by putting the leg of a living frog inside the stomach of a dog; the leg is gradually eaten away (Bernard). It is true that it has first been killed and then digested, but the question is, why the stomach-wall is not first killed and then THE SECRETION OF THE DIGESTIVE JUICES s8g digested ? When the wall has been injured by caustics or by an embolus, the gastric juice acts on it. But the living epithelium that covers it is able to resist the action of the acid and pepsin, which destroys the tissues of the frog's leg. The explanation is not to be found in the alkalinity of the blood, for the frog's blood is also alkaline, and the cells that line the intestine are preserved from the pancreatic juice, which is intensely active in an alkaline medium, while the living frog's leg is not harmed by a weakly alkaline pan- creatic extract, which does not digest the epithelium because it cannot kill it. A certain amount of protection may be afforded to the walls of the stomach by the thin layer of mucus which covers the whole cavity, for mucin is not affected by peptic digestion. And a mucous secretion seems in some other cases to act as a protective covering to the walls of hollow viscera, whose contents are such as would certainly be harmful to more delicate membranes, e.g., in the urinary bladder, large intestine, and gall-bladder. Still, how- ever important such a mechanical protection may be, it does not explain the whole matter, and it is necessary to suppose that the gastric epithelium has some special power of resisting the gastric juice, either by turning any of the ferment which may invade it into an inert substance and neutralizing any intrusive acid, or by opposing their entrance as the epithelium of the bladder opposes the absorption of urea. There is reason to believe that, as a matter of fact, free hydrochloric acid cannot penetrate the living cells, and it is to be noted that both active pepsin and free acid must be present at the same point within the cells before digestion of them can take place. In the gland-cells of the pancreas the protoplasm is, no doubt, shielded from digestion by the existence of the ferment in an inert form as zymogen ; and it is possible that this is one of the reasons for the existence of the mother-substance. But no such explanation is, of course, available for the intestinal epithelium. Trypsin when injected below the skin causes the tissue to break down and ulcerate. And while an active solution of trypsin can be allowed to remain a long time in an isolated loop of small intes- tine without producing any ill effect, damage is soon caused not only to the intestinal wall, but also to the liver, when the mucous membrane of the loop has been injured before the introduction of the trypsin. We must suppose, then, that the normal mucous membrane of the intestine prevents the absorption of trypsin, or, if it absorbs any of it, renders it harmless. On the other hand, .the intestinal mucosa is injured by the natural gastric juice when intro- duced directly into it unless the animal takes food simultaneously or a little earlier. But for reasons already given (p. 370) injury to the intestine cannot be produced in this way in normal digestion. It is impossible to escape the conclusion that each membrane becomes accustomed, and, so to speak, ' immune,' to the secretion normally 39° DIGESTION in contact with it, although not necessarily to other secretions. It is easy to multiply illustrations of this principle. The mucosa of the dog's urinary bladder is digested by the natural activated pancreatic juice of the dog, and still more readily by the natural gastric juice. Yet few tissues but the lining of the urinary tract or of the large intestine could bear the constant contact of urine or faeces. When urine is extravasated under the skin, or the contents of the alimentary canal burst into the peritoneal cavity, they come into contact with tissues which, although alive, are much less fitted to resist them than the surfaces by which they are normally enclosed; and the consequences, which are not wholly due to infection, are often disastrous. Leucocytes thrive in the blood, but perish in urine. Blood does not harm the endo- thelial cells of the vessels, but kills a muscle whose cross-section is dipped into it. The defensive or, in some cases, offensive liquids secreted by many animals are harmless to the tissues which produce and enclose them. A caterpillar investigated by Poulton secretes a liquid so rich in formic acid that the mere contact of it would kill most cells. The so-called saliva of Octopus macropus contains a substance fatal to the crabs and other animals on which it preys. The blood of the viper contains an active principle similar to that secreted by its poison-glands, but its tissues are not affected by this substance, so deadly to other animals. A step in the solution of our problem has been taken by Wein- land. Starting with the idea that if special protective mechan- isms against the digestive juices were anywhere to be found, it would be in the intestinal parasites whose whole existence is passed among them, he has made the important discovery that in these parasitic worms specific antiferments exist — i.e., substances which inhibit the action either of pepsin or of trypsin or of both. These substances can be precipitated from the expressed juice of the worms by alcohol, without completely losing their activity. Fibrin can be impreg- nated with them, and it is then, just like the ' living tissue,' rendered for a longer or shorter time unassailable by the proteolytic ferments. These facts are full of suggestion for future work, although the sup- posed proof that similar antiferments are contained in the cells of the mucous membrane of the stomach and intestines of the higher animals appears to have broken down. Substances can indeed be obtained by Weinland's method from the gastric and intestinal mucosa which, when added to a digestive mixture, strongly inhibit the digestion of proteins. But there is no clear proof that these sub- stances are specific antiferments. They are probably merely some of the split products of protein (Langenskjold). There is, however, some evidence of the existence of an antipepsin in many tissues including the mucous membrane of the stomach. As already men- tioned, it is known that an antitrypsin exists in the blood, with the INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 391 same properties as the antitrypsin in the intestinal worms (Hamill). This explains the resistance of blood-serum to the digestive action of trypsin. In addition to this body, which hinders the action of fully-formed trypsin, and has no effect upon enterokinase, the serum of some animals contains an antikinase — i.e., a substance which hinders the action, not of trypsin, but of enterokinase, pre- venting it from activating the trypsinogen into trypsin. SECTION V. — THE INFLUENCE OF THE NERVOUS SYSTEM ON THE DIGESTIVE GLANDS. The Influence of Nerves on the Salivary Glands. — All the salivary glands have a double nerve-supply, from the medulla oblongata through some of the cranial nerves, and from the spinal cord through the cervical sympathetic (Fig. 160). In the dog the chorda tympani branch of the facial nerve carries the cranial supply of the sublingual and submaxillary glands. It joins the lingual branch of the fifth nerve, runs in company with it for a little way, and then, breaking off, after giving some fibres to the lingual, passes, as the chorda tympani proper, along Wharton's duct to the submaxillary gland. In the hilus of this gland most of its fibres break up into fibrils around nerve-cells situated there, and lose their medulla in doing so. A few fibres terminate in a similar manner before entering the hilus, and a few deeper in the gland. The nervous path is continued by the axis-cylinder processes (p. 851) of these nerve-cells, which, passing in as non-medullated fibres, end in a plexus on the basement membrane of the alveoli. From the plexus fibrils run in among the gland-cells, but do not seem to penetrate them. The lingual, the chorda tympani proper, and Wharton's duct form the sides of what is called the chordo-lingual triangle. Within this triangle are situated many ganglion cells, a special accumulation of which has received the name of the submaxillary ganglion. This, however, should rather be called the sublingual ganglion, since its cells, as well as the others in the chordo-lingual triangle, are the cells of origin of axons which proceed as non-medullated fibres to the sublingual gland. The sublingual gland receives its cerebral fibres partly from branches given off from the lingual in the chordo-lingual triangle after the chorda tympani proper has separated from it, and ending around the nerve-cells within that triangle, partly from the chorda itself in the terminal portion of its course. These statements rest on anatomical and physiological evi- dence. The latter we shall return to. The cerebral fibres for the parotid (in the dog) pass from the tympanic branch of the glosso-pharyngeal (Jacobson's nerve) through connecting filaments to the small superficial petrosal branch of the facial, with this nerve to the otic ganglion, and thence by the auriculo-temporal branch of the fifth to the gland. The sympathetic fibres for all the salivary glands appear to arise from nerve-cells in the upper dorsal portion of the spinal cord. Issuing from the cord in the anterior roots of the upper thoracic nerves (first to fifth, but mainly second thoracic for the submaxillary), they enter the sympathetic chain, in which they run up to the superior cervical ganglion. Here they break up into terminal twigs, and thus come into DIGESTION VJ1 relation with ganglion cells, whose axons pass out as non-medullated fibres, and, surrounding the external carotid, reach the salivary glands along its branches. Langley has shown, by means of nicotine (p. 182), that the sympathetic fibres for the submaxillary and sublingual, and, indeed, for the head in general in the dog and cat, are connected with nerve-cells in this ganglion, but not between it and their termination, or between it and their origin from the spinal cord. Stimulation of the Cranial Fibres. — When in a dog a cannula is placed in Wharton's duct, and the saliva collected (p. 456), it is found that stimulation of the peripheral end of the divided chorda causes a brisk flow of watery saliva, and at the same time a dila- tation of the vessels of the gland, which we have already described in dealing with vaso- motor nerves (p. 179). Notwithstanding the vaso - dilatation, the volume of the gland is in general dimin- ished, owing to the rapid passage of water into the duct (Bunch). The blood has been shown to lose water in making the circuit of the submaxillary gland during excita- tion of the chorda, but doubtless some of the water of the saliva comes directly Fig. 160 — Nerves of the Salivary Glands. SM and SL, submaxillary and sublingual glands; P, parotid; V, fifth nerve; VII, facial; GP, glosso-pharyngeal; L. lingual; CT, chorda tympani; CL, chordo-lingual ; D, submaxillary (Wharton's) duct; C, ganglion cell of so-called submaxillary ganglion in the chordo- lingual triangle, connected with a nerve fibre going to sublingual gland; C", ganglion cell in hilus of sub- maxillary gland; SSP, small superficial petrosal branch of the facial; OG, otic ganglion; IM, inferior maxillary division of fifth nerve; AT, auriculo- temporal branch of fifth; JN, Jacobson's nerve; C', gangUon cells in superior cervical ganglion (SG) connected with sympathetic fibres going to parotid, submaxillary and sublingual glands. The figure is schematic. from the cells or from the lymph. That the increased secretion is not due merely to the greater blood- supply, and the consequent increase of capillary pres- sure, is shown by the injection of atropine, after which stimulation of the nerve, although it still causes dilatation of the vessels, is not followed by a flow of saliva. This can be shown fully as well by injecting a small quantity of yohimbin into the submaxillary artery. Great dilatation of the vessels is produced, but no saliva is secreted; INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 393 nor is the amount of oxygen consumed by the gland increased. Mere increase of pressure could not in any case of itself account for the secretion, since it has been found that the maximum pressure in the salivary duct when the outflow of saliva from the duct is prevented may, during stimulation of the chorda, much exceed the arterial blood-pressure (Ludwig). In one experiment, for example, the pressure in the carotid of a dog was 125 mm., in Wharton's duct 195 mm. of mercury. Even in the head of a decapitated animal a certain amount of saliva may be caused to flow by stimulation of the chorda, but too much may easily be made of this. And since the blood is the ultimate source of the secretion, we could not expect a permanent or copious flow in the absence of the circulation, even if the gland-cells could continue to live. In fact, when the circulation is almost stopped by strong stimulation of the sympathetic, the flow of saliva caused by excitation of the chorda is at the same time greatly lessened or arrested, even though the sympathetic itself possesses secretory fibres. So that, while there is no doubt that the chorda tympani contains fibres whose function is to increase the activity of the gland-cells, its vaso-dilator action is, under normal conditions, closely connected with, and, indeed, auxiliary to, its secretory action, although the dilatation of the vessels does not directly produce the secretion. This is only a particular case of a physiological law of wide application, that an organ in action in general receives more blood than the same organ in repose, or, in other words, that the tissues are fed according to their needs. The contracting muscle, the secreting gland, is flushed with blood, not because an increased blood- flow can of itself cause contraction or secretion, but because these high efforts require for their continuance a rich supply of what blood brings to an organ, and a ready removal of what it takes away. Evidence exists that in the salivary glands, as in muscle (p. 178), metabolic products given off during functional activity contribute to the dilatation of the vessels. This is the simplest explanation of the fact that the dilatation caused by chorda stimulation lasts longer when saliva is being secreted than when the secretion has been abolished by atropine. The quantity of blood passing through the parotid of a horse when it is actively secreting during mastication may be quadrupled (Chauveau). The parallel between the muscle and the gland is drawn closer when it is stated that electrical changes accompany secretion (p. 838), and that the rate of production of carbon dioxide and consumption of oxygen (in the submaxillary gland) is three or four times greater during activity than during rest. The temperature of the saliva flowing from the dog's submaxillary during stimulation of the chorda has been found to be as much as 1-5° C. above that of the blood of the carotid, although with the gland at rest no con- stant difference could be detected between the arterial blood and 394 DIGESTION the interior of Wharton's duct. But such measurements are open to many fallacies; and while there is no doubt that more heat is produced in the active than in the passive gland, it will not be surprising, when the vastly-increased blood-flow is remembered, that no difference of temperature between the incoming and out- going blood has been satisfactorily demonstrated. It has already been mentioned that most of the fibres of the chorda tympani proper become connected with ganglion-cells, and lose their medulla inside the submaxillary gland, only a few having already lost it by a similar connection with ganglion-cells in the chordo-lingual triangle. These facts have been made out by means of the nicotine method previously described (p. 182). Thus, it is found that, after the injection of nicotine (5 to 10 mg. in a rabbit or cat, 40 or 50 mg. in a dog), stimulation of the chorda tympani proper or of the chordo- lingual nerve causes no secretion from the submaxillary gland; but stimulation of the hilus of the gland is followed by a copious secretion — as much, if the stimulation is fairly strong, as was caused by excitation of the nerve before injection of nicotine. That this is due neither to any direct action on the gland-cells, nor to stimulation of the sympa- thetic plexus on the submaxillary artery, but to stimulation of chorda fibres beyond the hilus, is shown by the fact that after atropine has been injected in sufficient amount to paralyze the nerve endings of the chorda, but not of the sympathetic, stimulation of the hilus causes little or no flow of saliva. The application of nicotine solution to the chordo- lingual triangle does not affect the submaxillary secretion caused by stimulation of the chordo-lingual nerve, even in cases where a few secretory fibres for the submaxillary do not leave the chordo-lingual nerve in the chorda tympani proper, but are given off to the chordo- lingual triangle. This shows that none of the ganglion-cells in the triangle are connected with the secretory fibres of the submaxillary gland. By observations of the same kind they are known to be con- nected with fibres going to the sublingual. In a similar way, by observ- ing the effects of stimulation of the chorda on the bloodvessels before and after the application of nicotine, it has been found that the vaso- dilator fibres are connected with ganglion-cells in the same positions as the secretory fibres (Langley). Stimulation of the Sympathetic Fibres. — The sympathetic, as has been already indicated, contains both vaso-constrictor and secretory fibres for the salivary glands. If the cervical sympathetic in the dog is divided, and the cephalic end moderately stimulated, a few drops of a thick, viscid and scanty saliva flow from the submaxillary and sublingual ducts, while the current of blood through the glands is diminished. As a rule, no visible secretion escapes from the parotid, but microscopic examination shows that many of the ductules are filled with fluid, which is apparently so thick as to plug them up (Langley) ; while the cells show signs of ' activity ' (p. 376). Simultaneous Stimulation of Cranial and Sympathetic Fibres. — When the chorda and sympathetic are stimulated together, the former prevails so far, with moderate stimulation of the iatter, that the submaxillary saliva is secreted in considerable quantity, and is not particularly viscid. It is, however, richer in organic matter than is the chorda saliva itself. When the chorda is weakly, and INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 395 the sympathetic strongly, excited, the scanty secretion (if there is any) is of sympathetic type, thick and rich in organic matter. With strong stimulation of both nerves, the secretion, at first plentiful and watery, soon diminishes, even below the amount obtained by stimulation of the chorda alone, because of the diminution in the blood-flow, and therefore in the oxygen-supply, produced by the vaso -constrictors of the sympathetic (Heidenhain). With stimula- tion just strong enough to cause secretion when applied separately to either nerve, there is no secretion when it is applied simul- taneously to both. All this refers to the dog. In this animal, then, there seems to be a certain amount of physiological antagonism between the secretory action of the two nerves. But it differs in one respect from the antagonism between their vaso-motor fibres; for with strong stimu- lation the constrictors of the sympathetic always swamp the dilators of the chorda, while the secretory fibres of the chorda appear upon the whole to prevail over those of the sympathetic. And in all probability this apparent secretory antagonism is very superficial, and is due largely to the difference in the vaso-motor effects of the two nerves. For it has been shown that, when the blood- flow through the submaxillary gland is artificially diminished by gradu- ated compression of its artery, stimulation of the chorda gives rise to a thick, viscid and scanty saliva, relatively rich in organic solids (Heidenhain). When the amount of blood passing through the gland is made approximately the same as during stimulation of the sympathetic, the chorda saliva becomes practically identical in composition with the sympathetic saliva. This is one reason, perhaps the chief one, why the sympathetic, when both nerves are stimulated together, without artificial interference with the blood- supply, always appears to add something to the common secretion when there is a secretion at all, this something being represented by an increase in the percentage of organic matter. The observation that the sympathetic effect persists after stimulation has been stopped, and that excitation of the chorda after previous stimula- tion of the sympathetic causes a flow of saliva richer in organic matter than would have been the case if the sympathetic had not been stimulated, has long been considered a proof that the secretory fibres of the two nerves are widely different in function. To explain this result, Heidenhain assumed the existence in the sympathetic of a preponderance of fibres concerned in the building up in the cells of the organic constituents of the saliva (so-called ' trophic/ or, better, since the word ' trophic ' is usually associated with the building up of the bioplasm itself, ' trophic-secretory ' fibres). It would seem, however, that the increase in organic constituents is only realized when a sufficient time has not been allowed, after stimulation of the sympathetic, for the normal circulation to become re-established in the gland. The saliva obtained by stimulation of 396 DIGESTION the chorda immediately after a period of artificially diminished blood-flow, without any previous excitation of the sympathetic, also contains a surplus of organic matter (Carlson). Indeed, the distinction between chorda and sympathetic saliva, which, by taking account of the parotid as well as the submaxillary and sublingual glands, has been generalized into a distinction between cerebral and sympathetic saliva, and which, when the vaso-motor conditions are left out of account, appears to hold good in the dog and the rabbit, breaks down before a wider induction. For in the cat the sympathetic saliva of the submaxillary gland, although more scanty, is more watery than the chorda saliva (Langley), which, however, is by no means viscid; and the two secretions differ far less than in the dog. The discovery of Carlson that the usual effect of stimulation of the cat's cervical sympathetic with a weak interrupted current is a marked augmentation in the blood-flow* through the submaxillary gland affords an explanation. In accordance with this functional similarity, there is a much smaller difference in the action of atropine on the two sets of fibres in the cat than in the dog, although even in the cat the sympathetic is less readily paralyzed than the chorda. In their secretory action there is not even an apparent antagonism in the cat, with minimal stimulation of both nerves, which causes as much secretion as would be produced if both were separately excited. Further, even in the dog, after prolonged stimulation of the sympathetic, the submaxillary saliva is no longer viscid, but watery, the proportion of solids, and especially of organic solids, being much lessened, as it is also in chorda saliva after long excita- tion. When the cerebral nerve of the resting gland is strongly excited, it is found that up to a certain limit the percentage of organic matter in a small sample of saliva subsequently collected during a brief weak excitation increases with the strength of the previous stimulation; this is also true of the inorganic solids. But there is a striking difference when the experiment is made on a gland after a long period of activity; here increase of stimulation causes no increase in the percentage of organic material, while the inorganic solids are still increased. In both cases the absolute quantity of water, and therefore the rate of flow of the secretion, is augmented. All this pojnts to the same conclusion as the microscopic appear- ances in the gland-cells, that the cells during rest manufacture the organic constituents of the secretion, or some of them, and store them up, to be discharged during activity. The water and the inorganic salts, on the other hand, seem rather to be secreted on the spur of the moment, so to speak, and not to require such elaborate preparation. And it has been stated that when the * The increased blood-flow, which can be even better elicited by injection of adrenalin, immediately succeeds the secretion of saliva, and seems to be due not to the presence of vaso-dilator fibres in the sympathetic, but to metabolic product*. INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 397 chorda tympani is stimulated with currents of varying strength, the quantity of organic substances in small samples of saliva collected from a fresh gland is more nearly proportional to the rate of secretion than is the quantity of water and salts, which varies also with the blood-supply. Lest the apparently insignificant result of artificial stimulation of the sympathetic in such animals as the dog should cause its secretory action to be aopraised at too low a value, it should be remembered that in me intact body the sympathetic secretory fibres, when they are excited, are, it may be assumed, excited independently of the vaso-constrictors, and even in conjunction with the vaso- dilators'of the salivary glands. It is conceivable that such differences between chorda and sympathetic saliva as are not accounted for by the differences in the blood-flow during their stimulation are due, not to the nerve fibres, but to the end organs with which they are connected; that is, the two nerves may supply, not the same, but different gland- cells. And it is well known that even after prolonged stimulation of the chorda or chordo-lingual alone, some alveoli of the dog's submaxillary gland remain in the ' resting ' state ; after stimulation of the sympathetic alone, the number of unaffected alveoli is much greater; while after stimulation of both nerves, few alveoli seem to have escaped change. If there is no essential difference between the cranial and sympathetic secretory fibres, it is easy to understand that they will be distributed to different secreting elements. The supposed proof that there must be some overlapping in the nerve- supply — i.e., that some cells must be supplied from both sources, since excitation of the sympathetic influences the amount of organic material in the saliva obtained by subsequent stimulation of the chorda — is, as we have just seen, by no means so cogent as has been assumed. And, indeed, we know nothing of a division of labour between the cells of a gland, except when there are obvious anatom- ical distinctions. Thus, the submaxillary gland in man contains both serous and mucous acini, and mucin-making cells are scattered over the ducts of most glands, and, indeed, on nearly every surface which is clad with columnar epithelium. In these cases we cannot doubt that one constituent — mucin — of the entire secretion is manu- factured by a portion only of the cells. In the cardiac glands of the stomach, too, the ovoid cells, in all probability, yield the whole of the acid of the gastric juice. But, so far as we know, every hepatic cell is a liver in little. Every cell secretes fully-formed bile ; every cell stores up, or may store up, glycogen. So it is with the secretory alveoli of the pancreas, if we consider the islands of Langerhans as having no connection with the alveoli; one cell is just like another; all apparently perform the same work; each is a unicellular pan- creas. (See p. 638.) Paralytic Secretion. — When the chorda tympani is divided, a slow ' paralytic ' secretion from the submaxillary gland begins in a few 398 DIGESTION hours, and continues for a long time accompanied by atrophy of the gland. There is also a secretion of the same kind from the submaxillary on the opposite side, but it is less copious. This is called the ' antilytic ' secretion, which is most pronounced in the first few days after the operation, and seems to be a transient phenomenon. It can be at once abolished by section both of the chorda and the sympathetic on the corresponding side, ana is therefore due to impulses arising in the central nervous system. The cause of the paralytic secretion has not been fully made out. If within two or three days of division of the chorda the sympathetic on the same side is cut, the secretion is greatly diminished or stops altogether; and it is concluded that up to this time it is maintained by impulses passing along the sympathetic to the gland from the salivary centre, the excitability of which has been in some way increased by division of the chorda, possibly by some such degenerative process in the cells as the changes seen in cerebro-spinal motor cells whose axons have been divided (p. 858). This may also account for the antilytic secretion. But if section of the sympathetic is not performed for several days, it has no effect on the paralytic secretion, which at this stage seems to depend on local changes in or near the gland itself, leading to a mild continuous excitation of those nerve-cells on the course of the fibres of the chorda to which reference has already been made. Section of the sympathetic alone causes neither secretion nor atrophy, nor does removal of the superior cervical ganglion. The histological characters of the gland-cells during paralytic secretion are those of ' rest.' Reflex Secretion of Saliva. — The reflex mechanism of salivary secretion is very mobile, and easily set in action by physical and mental influences. It is excited normally by impulses which arise in the mouth, especially by the contact of food with the buccal mucous membrane and the gustatory nerve-endings. The mere mechanical movement of the jaws, even when there is nothing between the teeth, or only a bit of a non-sapid substance like india- rubber, causes some secretion. The vapour of ether gives rise to a rush of saliva, as does gargling the mouth with distilled water. The smell, sight, or thought of food, and even the thought of saliva itself, may act on the salivary centre through its connections with the cerebrum, and make ' the teeth water.' A copious flow of saliva, reflexly excited through the gastric branches of the vagus, is a common precursor of vomiting. The introduction of food into the stomach also excites salivary secretion. The researches of Pawlow and his pupils have shown that the salivary glands are not excited indifferently by everything which comes into contact with the buccal mucous membrane. A remark- able adaptation exists between the properties of food or foreign bodies introduced into the mouth and their effects upon the secre- tion of saliva. When solid dry food is given to a dog saliva is copiously poured out; much less is secreted when the food is moist. Acids or salts induce an abundant flow, in order that they may be neutralized, diluted or washed out of the mouth. In this case a watery liquid, poor in mucin, flows from the mucous glands. Mucin is a lubricant to facilitate the swallowing of solid food, and INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 399 here it could be of no use. When clean pebbles are put in the dog's mouth the animal may try to chew them, but eventually ejects them. Either no saliva or very little is secreted, since it could not aid in their expulsion. If, however, the very same stones are reduced to sand and again introduced into the animal's mouth, saliva is plentifully secreted to wash it out. The serous and mucous salivary glands are not necessarily excited by the same food materials, and here again we can trace an astonish- ingly exact adaptation. A permanent parotid or submaxillary fistula can easily be made in a dog by freeing Stenson's or Wharton's duct from the surrounding mucous membrane for a little distance, bringing the natural orifice of the duct out through a small wound in the cheek, and stitching it in position there. When it is desired to collect saliva, the wide end of a funnel-shaped tube, whose stem is bent so as to hang vertically, can be attached by a little shellac of low melting-point to the skin around the orifice of the duct and at some distance from it, and on the narrow end can be hung a small graduated tube, into which the saliva drops. When fresh meat is given to the animal little or no parotid saliva is secreted, while a copious flow takes place from the submaxillary gland, mucin being required to lubricate it for deglutition, while water is not specially needed. But if the meat is in the form of a dry powder the parotid pours out a plentiful secretion, while the submaxillary also secretes a fluid relatively rich in mucin. The same difference is seen between fresh moist bread and dry bread. The afferent nerve-endings from which impulses are carried to the reflex centres (or the portions of the salivary centre) which preside over the various salivary glands must possess the power of very delicate selection as regards the kinds of stimulation by which they are affected. The mere relish of the animal for the different kinds of food plays but a small part. Most dogs display a much livelier interest in a piece of meat than in a piece of dry biscuit, yet it is the biscuit which excites the parotid to activity. The sight of dry food causes an abundant flow of watery saliva from the parotid, and a flow of fluid rich in mucin from the sub- maxillary. Various uneatable substances, including substances which in contact with the mucous membrane of the mouth produce strong and disagreeable stimulation of it, and excite disgust, cause also, when viewed from a distance, secretion by all the salivary glands; but the submaxillary saliva, as ought to be the case for substances unfit for food, and therefore not destined to be swallowed, is poor in mucin. When the animal is shown pebbles and sand the phenomena are qualitatively the same as when they are put into its mouth — the glands remaining inactive in presence of the pebbles, but secreting plentifully at sight of the sar^d. In short, the same adaptation is observed in the case of the so-called psychical secretion as when the stimulating substances act directly 4oo DIGESTION upon the endings of the afferent salivary nerves in the buccal mucous membrane. It is further worthy of note that when the animal is hungry the psychical secretion is most copious and most easily obtained. After a full meal it cannot be excited at all. When food (or other exciting substance) is repeatedly shown to a fasting animal the reaction becomes each time weaker, and finally the glands cease to respond. All that is then necessary to restore the reaction is to put into the animal's mouth a little of the food (or other object) . When it is now shown it at a distance the ordinary effect follows promptly. This indicates that the condition of the salivary centre exercises an important influence upon the psychical secretion, its excitability to the weaker stimulus set up by the sight of the object being increased by the stronger reflex stimulation coming directly from the mouth. In the condition of satiety the inexcitability of the centre may be due to the action of food- products in the blood. In most animals and in man the activity of the large salivary glands is strictly intermittent. But the smaller glands that stud the mucous membrane of the mouth never entirely cease to secrete, and the same is the case with the parotid in ruminant animals. The centre is situated in the medulla oblongata, stimulation of which causes a flow of saliva. The chief afferent paths to the salivary centre are the lingual branch of the fifth and the glosso- pharyngeal ; but stimulation of many other nerves may cause reflex secretion of saliva. In experimental reflex stimulation, the sole efferent channel seems to be the cerebral nerve-supply of the glands. After section of the chorda, no reflex secretion by the submaxillary gland can be caused, although the sympathetic remains intact. It was alleged by Bernard that, after division of the chordo- lingual, a reflex secretion could be obtained from the submaxillary gland by stimulating the central end of the cut lingual nerve between the so-called submaxillary ganglion and the tongue, the ganglion being supposed to act as ' centre.' It has been shown, however, that this is not a true reflex effect, but is due to the excitation of certain (recurrent) secretory fibres of the chorda that run for some distance in the lingual, then bend back on their course and pass to the gland. It may be in part a pseudo- or axon-reflex (p. 913), elicited by excitation of efferent fibres, which send branches to some of the ganglion-cells. The salivary centre can also be inhibited, especially by emotions of a painful kind — for instance, the nervousness which often dries up the saliva, as well as the eloquence, of a beginner in public speaking, and the fear which sometimes made the medieval ordeal of the consecrated bread pick out the guilty. In rare cases the reflex nervous mechanism that governs the salivary glands appears to completely break down; and then two opposite conditions may be seen — xerostomia, or ' dry mouth,' in INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLAtiDS 46! which no saliva at all is secreted, and chronic ptyalism, or hydro- stomia, where, in the absence of any discoverable cause, the amount of secretion is permanently increased. Both conditions are said to be more common in women than in men. The Influence of Nerves on the Gastric Glands. — Like saliva, gastric juice is not secreted continuously, except in animals such as the rabbit, whose stomachs are never empty. The normal and most efficient stimulus is the eating of food and its presence in the stomach. Mechanical stimulation of the gastric mucous membrane with a non-digestible substance, such as a feather or a glass rod, causes secretion of mucus, but not of gastric juice. But the observations mentioned above on the difference of response of the salivary glands to different substances suggest that the local mechan- ical stimulation of the food on the gastric glands may be more effective. There is also at first thought much to indicate that the gastric glands are stimulated chemically in a more direct manner than the salivary glands by the local action of food substances reaching the cells by a short-cut from the cavity of the stomach, or in a more roundabout way by the blood. And it might be very plausibly argued that the gastric glands are favourably situated for direct stimulation, while the large salivary glands are not ; and that the great function of saliva being to aid deglutition, an almost momentary, and at the same time a perilous act, it is necessary to provide by a nervous mechanism for an immediate rush of secre- tion at any instant, while it is not important whether the gastric juice is poured out a little sooner or a little later, and therefore it is left to be called forth by the more tardy and haphazard method of local action. Nevertheless, on looking a little closer, we find that this does not exhaust the subject, and that the gastric secretion can be influenced by events taking place in distant parts of the body, just as the salivary secretion can. In a boy whose oesophagus was completely closed by a cicatrix, the result of swallowing a strong alkali, and who had to be fed by a gastric fistula, it was found that the presence of food in the mouth, and even the sight or smell of food, caused secretion of gastric juice (Richet). Here there must have been some nervous mechanism at work. The secretion cannot have been excited by the direct action of absorbed food-products circulating in the blood — an explanation which might be given, though an insufficient one, of the secretion seen in an isolated portion of the cardiac end of the stomach during the digestion of food in the rest. The efferent nervous channels through which these effects are produced have been defined by Pawlow's experiments on dogs. He first made a gastric fistula, then a few days afterwards divided the oesophagus through a wound in the neck, and stitched the two cut ends to the edges of the wound. After the animals had recovered, it was observed that when meat was given to them by the mouth, a copious secretion of 402 DIGESTION Py/oru. Plexus gast anterior gastric juice followed in five or six minutes, notwithstanding the fact that in this ' sham feeding ' the food immediately escaped from the opening in the upper portion of the divided oesophagus. Much the same result was seen when the food was simply shown to the animal. Indeed, when a hungry animal is tempted with the sight of meat, the flow of gastric juice, always occurring after a latent period of five or six minutes, may be even greater than with sham feeding. Division of the splanchnic nerves had no effect on this reflex secretion, while it could not be obtained after division of both vagi below the origin of their cardiac and pulmonary branches, by which disturbance of the heart and respiration are avoided. Further, stimulation of the peripheral end of the vagus in the neck* caused secretion. These experiments show that secretory fibres for the gastric glands run in the vagi. It is probable that the vagi also contain efferent fibres which inhibit ophagus , the gastric Plexus gastn'eus T,, ' posterior vagi, secretion. 1 ne excitation of the secretory fibres is not produced re- flexly by the processes of mastication and deglutition as such. Di- lute acid is the most powerful chemical stim- ulus for the buccal mucous membrane, and when it is introduced into the mouth of a dog with a double cesophageal and gastric fistula, an abundant secretion of saliva at once ensues. But no matter how long the animal continues to swallow the mixture of saliva and acid, no gastric juice is formed. The same is the case in sham feeding with salt, pepper, mustard, smooth stones, and even extract of meat. It is the desire for food —the appetite, as we call it — and the feeling of satisfaction associa- ated with eating food that the animal relishes, which is the efficient cause of the gastric secretion in sham feeding. The more eagerly the dog eats, the greater is the flow of gastric juice. * The nerve was not stimulated till a few days after the section, so as to allow the cardio-inhibitory fibres to degenerate. Otherwise the heart would have been stopped by the stimulation. Fig. 161. — Pawlow's Stomach Pouch. AB, line of incision; C, flap for forming the stomach pouch. At the base of 'the flap the serous and muscular coats are preserved, and only the mucous membrane divided, so that the branches of the vagus going to the pouch are not severed. INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 403 rosa cularis Pawlow also performed the converse experiment. In dogs in which a pouch had been isolated from the stomach and made to open to the exterior by the surgical procedure illustrated in Figs. 161 and 162, he introduced into the large stomach, without the animal's knowledge, food of various kinds. This is best done in a sleeping dog. The secretion of gastric juice, both in the main stomach and in the pouch or miniature stomach", which is known in a great variety of conditions to present an exact picture of the process of secretion in the large, is markedly delayed and scanty when it does appear. Bread and coagulated egg-white did not yield a single drop during the first hour or more. Raw flesh excited a secretion, but after an inter- val of fifteen to forty - five minutes, in- stead of five or six to ten, as in sham feeding. It was very scanty during the first hour (only o n e - third the nor- mal amount), and possessed a very low di- gestive power. The impor- tance of the psychical ele- ment is shown by the fact that in one dog, which, after a weighed amount of meat had been introduced into its stomach (without its knowledge), received a sham meal of meat, the amount of protein digested after one and a half hours was five times greater than in another animal treated exactly in the same way, except that the sham meal was omitted. But even after division of the vagi, gastric secre- tion is still caused by the introduction of various substances into the stomach, especially water and meat extract. The active substances in the meat extract are, for the most part, insoluble in alcohol. Kreatin is inactive. It is in virtue of these substances that raw meat placed directly in the stomach causes some secretion after a Fig. 162. — Pawlow's Stomach Pouch. S, the completed pouch; V, cavity of stomach. 4°4 DIGESTION time. Milk and gelatin solution are also direct excitants of gastric secretion apart from the water in them. Starch, fat, and egg-white are totally inert. After section of both vagi in dogs, no marked qualitative or quantitative changes have been observed in the gastric juice. The secretion caused by the presence of food in the stomach is still obtained when, in addition to the vagi, all other nerves which can possibly connect the central nervous system with the organ have been severed and the sympathetic abdominal plexuses have been destroyed (Popielski). We must therefore sug- pose that the gastric glands, while normally under the control of a nervous mechanism in the upper portion of the cerebro- spinal axis whose efferent fibres run in the vagi, are also capable of being locally stimulated throug!i the peripheral ganglia in the stomach walls or the chemical action of the products of digestion absorbed into the blood. Edkins showed that the injection of food substances or the products of their digestion (broth, dextrin, peptone) or of acid into the blood caused no secretion of gastric juice, while the injection of an extract of the pyloric mucous membrane, made by boiling it with water, acid, or peptone, excited a certain amount of secretion. He therefore concluded that the secondary secretion of gastric juice is determined, not by local stimulation of a reflex mechanism in the gastric wall, but by the production in the mucous membrane of the pyloric end of a chemical substance, the gastric secretin or gastric hormone,* which is absorbed by the blood, and acts as an excitant to all the gastric glands. The cardiac mucosa was found incapable of forming this substance. It is not to be imagined that the ' psychical ' secretion and the secretion called forth by the direct action of the food or food- products in the stomach perform independent offices. They can, in various instances, be shown to supplement each other. For example, not more than one-half or one-third of the gastric juice secreted during the digestion of bread or boiled egg-albumin can be ascribed to the psychic effect. Yet these substances, when introduced directly into the stomach, cause practically no secretion. We must suppose that during the digestion of the bread and albu- min by the psychically secreted juice certain products analogous to those in the meat extract are formed, which act as chemics.1 excitants of the local secretory apparatus. The psychic juice is indispensable in this case to start the process, ' to set the stove ablaze,' as Pawlow puts it. In the case of meat it is not indispensable, since the meat can chemically excite the gastric glands; but it greatly hastens the process of digestion. These facts emphasize the importance of * ' Hormone ' (from 6p/taw, I arouse or excite) is the name given to a sub stance which, carried by the blood from the place where it is formed, acts as a chemical messenger in exciting the activity of some more or less distant organ. The classical example is the pancreatic secretin which, manufactured in the intestinal mucosa, excites the secretion of the pancreatic juice. INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 405 appetite in digestion, a truism in treatment which thus receives for the first time a rational explanation. The influence of good- humour upon nutrition, which experience has crystallized into the proverb ' Laugh and grow fat,' has also been shown to depend — in great part, at least — upon a beneficial action on the digestive functions, both motor and chemical. The movements of a cat's stomach and intestines have been observed to cease when the animal became angry or excited by unpleasant emotions; and in a dog whose gastric glands were pouring out a copious psychical secretion in response to a sham meal, secretion stopped abruptly when the animal's wrath was awakened by what is probably to the normal dog the most specifically ' adequate ' stimulus for the emotion of anger — the sight of a cat which he was restrained from chasing. By means of experiments with the miniature stomach it has been further shown. that each kind of food has its own characteristic curve of gastric secretion. With flesh diet the maximum rate of secretion occurs during the first or second hour, and in each of the first two hours the quantity of juice furnished is approximately the same. With bread diet we have always a sharply-indicated maximum in the first hour, and with milk a similar one during the second or the third hour (Fig. 163). The juice secreted on different diets also differs in digestive power — i.e., in the amount of protein which a given quantity of it will digest in a given time. ' Bread juice ' is much stronger in ferment than ' meat juice/ and ' meat juice ' somewhat stronger than ' milk juice ' (Fig. 164). But ' meat juice ' has a higher acidity than ' bread juice,' ' milk juice ' being intermediate. These differences do not necessarily indicate that the gastric mucous membrane responds in a specific way to each kind of food substance, as suggested by Pawlow. They may depend on several circumstances, and particularly on this — that the quantity, though not the quality, of the psychical or ' appetite ' juice is related to the relish with which the animal eats the food. The products formed in the digestion of the different foods by the psychical juice may therefore be different in nature and amount, and thus the quantity of the gastric hormone which determines the secondary secretion may vary with the food. The young mammal, like the adult, secretes gastric juice before the food reaches the stomach. In puppies from one to eighteen days old sham feeding (sucking the teats of the mother after'an oesophageal fistula has been made in the younger animals and a double oesophageal and gastric fistula in the older) causes a liquid with the properties of gastric juice to gather in the stomach. This power, then, is a congenital one. The individual does not gain it by experience; it comes into the world with him (Cohnheim). The Influence of Nerves on the Pancreas. — Like the stomach, the pancreas receives secretory fibres through the vagus. These are 406 DIGESTION 45678 » 23456 789101 23456 probably connected with a reflex centre in the medulla oblongata. It has long been known that when the medulla is stimulated a flow of pancreatic juice is occasionally set up, or is increased if already going on. The same is true when the vagus is stimulated in the ordinary way in the neck. But the experiment often failed, for the pancreas is peculiarly susceptible to circulatory disturbances, and stimula- tion of the bulb or the vagus may interfere with the blood- flow through the gland by exciting its vaso-con- strictor fibres or causing inhibition of the heart. These disturbing influences may be avoided, as Pawlow has shown, by stimulating the vagus, three or four days after dividing it, with slowly-recurring stimuli (induction shocks or light blows from a small hammer worked by an Flesh, 200 grm. Bread, 200 grm. Milk, 600 c.c. Fig. 163. — Rate of Secretion of Gastric Juice with Diets of Meat, Bread, and Milk (Pawlow). electro - mag- H°urs 10.0 net at the rate of about t go one in the «j second). The £| e,o secret or y ^3 fibres are still ° « *.° susceptible of J excitation, while the car- c dio-inhibitory fibres, which degenerate Fig. 16- more rapidly. digest i *_. protei are almost or nour> altogether in- 2345678234 5678923456 / ^ i i \ \ ^ — _ — ' / \ f S \ __ ^ \ / \ / \ / \ ^-- j ^ j ^ Flesh, 200 grm. Bread, 200 grm. Milk, 600 c.c. ).. — Digestive Power of Gastric Juice (Pawlow). The ive power of the juice, as measured by the length of the n column digested in Mett's tubes, is represented hour by with diets of flesh, bread, and milk. excitable, and the vaso-constrictors are but little affected by these slow rhythmical stimuli, which excite the secretory nerves (p. 175). A pancreatic fistula has previously been established by excising a small portion of the duodenal wall containing the open- ing of the pancreatic duct, closing the intestine by sutures, and I NFL UENCE OF NER VO US S YSTEM ON DIGEST I VE GLA NDS 407 stitching the orifice of the duct into the abdominal wound. On stimulation of the vagus the juice will begin in two to three minutes to drop from a cannula in the duct, and will continue to flow for several minutes after cessation of the stimulus. The sympathetic also contains secretory fibres for the pancreas. Efferent fibres which inhibit the secretion have been also discovered in the vagus. Their presence may be most clearly demonstrated when that nerve is stimulated during the flow of pancreatic juice excited by the introduction of dilute acid into the duodenum. Stimulation of the central end of the vagus and oi the other nerves is capable of reflexly inhibiting the pancreatic secre- tion. Painful im- pressions have a strong inhibitory in- fluence. This is one of the reasons why many observers failed to detect the secretory nerves. The inhibition caused by vomiting is probably due to impulses ascending the vagus. It is pos- sible that through these nervous chan- nels the pancreatic secretion is affected by the psychical con- ditions connected with eating and the desire for food, just as in the case of the gastric secretion ; but our information on this subject is scantier and less precise. A flow of juice may un- doubtedly take place within three or four minutes after food is taken, but it is not quite certain whether this is not determined by the passage of some of the acid gastric contents into the duodenum. Secretin. — We have already referred to the fact that pancreatic secretion is excited by the presence of acid in the duodenum. The mechanism of this action is of great interest. Two or three minutes after the introduction of 0-4 per cent, hydrochloric acid into the duodenum, pancreatic juice begins to flow. A similar effect is seen when the acid is placed in the jejunum, but not when it is injected Fig. 165. — Secretion of Pepsin. C shows the quantity of pepsin(ogen) in the mucous membrane of the cardiac end of the stomach at different times during digestion; P, the quantity of pepsin(ogen) in the mucous membrane of the pyloric end ; S, the quantity of pepsin in the secretion of the cardiac glands. The numbers marked along the horizontal axis are hours since the last meal. About five hours after the meal, S reaches its maximum. From the very be- ginning of the meal C falls steadily down to the tenth hour, and then begins to rise — i.e., the gland- cells of the cardiac end of the stomach become poorer in pepsin(ogen) as secretion proceeds. 403 DIGESTION into the lower part of the ileum. It is obtained as strongly and as promptly from an isolated loop of intestine when all the nerves passing to it have been cut, and the solar plexus extirpated, and also after the administration of atropine, which paralyzes the endings of secretory nerves elsewhere. The secretion accor lingly does not depend upon a local reflex mechanism, with its afferent endings in the intestinal mucous membrane, but upon some substance which is carried to the pancreas by the blood, and acts directly upon its cells. This substance is not the acid, for the injection of 0-4 per cent, hydrochloric acid into the blood produces no effect upon the pan- creas. It has been shown by Bayliss and Starling that the exciting substance is a diffusible body of low molecular weight, probably of organic nature, but not a protein, which they call secret in. It is soluble in alcohol or alcohol and ether, and is not destroyed by boiling. It is produced in the mucous membrane of the jejunum or duodenum on exposure to dilute hydrochloric acid. Extracts of mucous membrane so treated cause a copious pancreatic secretion, and a smaller secretion of bile, when injected in small quantities into the blood of animals in which no such secretion is taking place, but have no influence on any other gland. At the same time the arterial blood-pressure falls somewhat. The substance which produces the fall of blood-pressure is different from secretin, since acid extracts of the lower end of the ileum, which have no effect on the flow of pan- creatic juice, diminish the blood-pressure. A precursor of secretin, called pro-secretin, exists in the intestinal mucous membrane, and can be extracted from it by physiological salt solution. ' It does not affect the pancreatic secretion. By boiling or by the action of acid secretin is split off from it. Pro-secretin is most abundant in the duodenum, and diminishes as we pass down the .intestine. Secretin is very widespread in the animal kingdom. In the monkey, dog, cat, rabbit, man, ox, sheep, pig, squirrel, goose, tortoise, salmon, dog-fish, and skate evidence of its presence has been obtained. The secretin of one animal will excite a flow of pan- creatic juice in an animal of a different kind as well as in one of the same kind. In normal digestion secretin is formed under the influence of the acid chyme, not in the stomach, but after it has passed into the duodenum. The passage of the chyme through the pylorus, as previously mentioned (p. 337), is regulated by the re- action of the duodenal contents, as well as by the consistence of the gastric contents. So long as the liquid in the duodenum is acid, the pylorus remains closed. As soon as the first small portion of acid chyme ejected from the stomach has been neutralized by the in- creased secretion of the pancreatic juice and the outpouring of bile from the gall-bladder in response to the stimulus of the acid, the pylorus opens again. According to Pawlow, certain food substances, notably fat, and I NFL UENCE OF NER VO US S YSTEM ON DIGESTI VE GLANDS 409 water stimulate the pancreatic secretion, and with great promptness, even before any acid has been produced in the stomach, and there- fore before any can have passed into the duodenum. Possibly this effect is elicited through the long reflex paths already described as running in the vagi or through a local nervous mechanism, which, although it does not take part in the excitation of the pancreatic secretion by acid, may yet exist for the performance of other offices. It is more probable, however, that it is due to the passage of some of the gastric contents through the pylorus; for when oil is introduced into the small intestine, it causes the production of secretin, although, unlike dilute acid, it is quite ineffective in forming secretin when rubbed up with the scraped-off mucous membrane. That secretin acts on the pancreatic cells in a different way from the secretory nerve fibres contained in the vagus is indicated by the difference in the characters of the juice secreted under the influence of the two mechanisms. The nervous secre- tion is thick and opalescent, rich in enzymes, including trypsin in the active form, and proteins, but its alkali content is low. Like other secretions excited through nerves, it is inhibited by atropine. The chemical secretion due to secretin, is thin and watery, rich Fi«- 166.— Rate of Secretion of Pan- n v • j creatic Juice. S shows the variation in alkalies, poor in proteins and in the rjate o{ secretion of the pan. in enzymes, and among the latter, creatic juice in a dog; P, the varia- trypsin occurs only in the inactive tion in the percentage of solids in /c ., •, x the juice. It will be seen that the n ^awitSCnj. < maxima of S fall at the same time The pancreatic, like the gastric, as the maxima of P. The numbers juice is said to vary as regards alonS the horizontal axis are hours j-. _,• /• •,T~ ,-u since the last meal, its digestive properties with the nature of the food. On a diet of bread the juice is very poor in fat-splitting ferment, while on a diet of flesh it is richer, and on a diet of milk richest of all. With bread the juice is relatively rich in amylolytic ferment. When we take the quantity of the juice as well as its strength in ferments into consideration, it is stated that bread occasions the secretion of a juice with a greater quantity of proteolytic ferment than either milk or meat, although it is relatively dilute (Fig. 169). The vegetable proteins require more ferment to digest them than proteins of animal origin. * There is no more evidence that the adaptation of the pancreatic juice to the nature of the food is due to a specific sensibility of the duodenal mucosa to the various food-stuffs than there is in the case of the adaptation of the gastric juice. If the volume of the chyme and its acidity are related to the nature of the food, then the amount of secretin formed, and therefore the intensity of secretion in the 4io DIGESTION pancreag, will be similarly related. The one apparently proved example of specific adaptation of the pancreatic juice has not stood the test of a critical examination. It was asserted that in dogs fed for some days with food containing lactose (milk) the ferment, lactose, is present in that secretion, while the pancreatic juice of dogs whose food is free from lactose does not contain lactase. The adaptation of the pancreas to lactose was supposed to be achieved through some substance produced by the action of lactose on the i n in iv v i M in iv vviviivmi ii in iv v vi intestinal mucous membrane, which plays the part of a specific chemical stimulus to the pancreatic cells or their secretory nervous mechan- ism, causing them to form lactase. But it has been conclusively shown that when dogs are fed with lac- tose for weeks no lactase appears in the pancreatic juice (Plimmer). The natural se- cretion of pan- creatic juice is by no means so inter- mittent as that of saliva. In the rab- bit the pancreatic, like the gastric, juice flows con- tinuously. In the dog it begins al- most as soon as food is taken, rises in two or three hours to a maximum, then faUs till the fifth or sixth hour, after which it may mount 'again somewhat, and then, gradually diminishing, ultimately stops (Figs 166, 167) . D uring normal activity the bloodvessels of the gland are dilated. But under experimental conditions the increased secretion caused by secretin is accompanied sometimes by an increase and sometimes by a diminution in the blood-flow, and secretion may continue for some time after complete cessation of the circulation Flesh, 100 grm. Bread, 250 grm. Milk, 600 grm. Fig. 167. — Secretion of Pancreatic Juice with Different Diets (Pawlow). The hours are in roman numerals. I NFL UENCE OF NER VO US S YSTEM ON DIGESTI VE GLA NDS 411 while the increased consumption of oxygen which goes hand in hand with the increased secretion is also independent of the blood-supply (May, Barcroft and Starling). This shows how far the secretory pro- cess is from a mere mechanical nitration, although it does not follow that, under normal conditions, a decreased blood-flow ever does accompany an increased secretion. There is one difference between the normal secretion of pancreatic juice and of saliva which may still be mentioned : the pressure of the latter in the submaxillary duct may, as we have seen, greatly exceed the arterial blood-pressure, without reabsorption and consequent oedema of the gland occurring ; but the secretory pressure of the pancreatic cells is very low, not more than a tenth of that of the salivary glands. (Edema begins before a manometer in the duct shows a pressure of 20 mm. of mercury, the secreted fluid passing very easily into the lymph spaces. The mutual relations of the spleen and pancreas have formed the subject of numerous inquiries. Some authors maintain that the spleen plays an im- portant role in the elaboration of the proteolytic ferment of the pancreas, forming a substance which we may call pro-trypsinogen, since it is supposed to be carried in the blood to the pancreatic cells, and changed by them into trypsin- ogen. There is some evidence that extracts of the spleen prepared from it when congested during digestion exert a favourable influence on the proteolytic power of the pancreas (Mendel). And there is no doubt that the spleen, like other organs, contains an intracellular enzyme which can aid in the digestion of protein. The products of the action in an acid medium of this enzyme are the same as those formed by trypsin in an alkaline medium (Leathes). But this is not enough to prove that the spleen has any special relation to pancreatic digestion. The Influence of Nerves on the Secretion of Bile. — Although bile is secreted constantly, it only passes at intervals into the intestine. For the liver in many animals, unlike every other gland except the kidney, has in connection with it a reservoir, the gall-bladder, in which its secretion accumulates, and from which it is only expelled occasionally. We have therefore to distinguish the bile- secretion from the bile-expelling mechanism. To study the rate of secreting of bile (Fig. 168), a fistula of the gall-bladder can be established. Fig. 168.— Rate of Secretion of Bile. S shows how the rate of secretion of bile falls in a dog when a biliary fistula is first made, and the bile thus pre- vented from entering the intes- tine ; P shows the fall of the per- centage of solids. The numbers along the horizontal axis are quarters of an hour since bile began to escape through the fistula. The numbers along the vertical axis refer only to curve S, and represent the rate of secretion in arbitrary units. 412 DIGESTION But to learn the function of bile in digestion it is more important to know when and at what rate it enters the intestine. For this purpose a fistula is made by cutting the natural orifice of the common bile-duct with a piece of the surrounding mucous membrane out of the intestine and transplanting it upon the serous coat, where it is sutured. The loop of intestine, with the orifice of the duct facing outwards, is then stitched into the abdominal wound, where it is allowed to heal. Of course, since a circulation of the bile-acids takes place — i.e., an absorption from and r^ -excretion into the intestine — the formation of that juice cannot proceed upon abso- lutely normal lines when the bile no longer enters the duodenum. The only condition under which fistula bile could have the same composition as normal bile would be that in which as great an amount of bile-acids is introduced into the gut as escapes through the fistula. A circulation of a smaller proportion of the bile-pig- ments is also probable. A circulation of the biliary cholesterin is denied by some observers (Stadelmann) but affirmed by others. It is certain that cholesterin is of importance in the body, and if the supply of sterins (p. 570) in the food is insufficient it is to be sup- posed that some of the biliary cholesterin would be used over again. Of the direct influence of nerves, either on the secretion of bile or on its expulsion, we have scarcely any knowledge, scarcely even any guess which is worth mentioning here. It is true the secretion of bile may be distinctly affected by the section and stimulation of nerves which control the blood-supply of the stomach, intestines, and spleen, for the quantity of blood passing by the portal vein to the liver depends upon the quantity passing through these organs, and the rate of secretion is diminished when the blood-supply is greatly lessened. In this way stimulation of the medulla oblongata, the spinal cord, or the splanchnic nerves stops or slows the secretion of bile by constricting the abdominal vessels ; and the same effect can be reflexly produced by the excitation of afferent nerves. The right splanchnic nerve contains inhibitory with some motor fibres, and the vagi (especially the left) contain motor fibres for the gall-bladder. Probably its contraction takes place naturally in response to reflex impulses from the mucous membrane of the duo- denum, for the application of dilute acid to the mouth of the bile- duct causes a sudden flow of bile,* and the acid contents of the stomach, when projected through the pylorus into the intestine, have a similar effect. But, in addition, as we have seen, the secretin formed will cause an increase in the rate of secretion of the bile. In studying the effect of secretin it is necessary to obtain it free from bile- salts, since these cause of themselves an increased secretion of bile. When this is done by dissolving out with alcohol any bile-salts which may be present in the extract of intestinal mucous membrane, * This result seems to be difficult to realize experimentally. Bainbridge and Dale could not elicit reflex contraction of the gall-bladder (in anaesthetized animals) in this way. / NFL UENCE OF NER VO US S YSTEM ON DIGESTI VE GLA NDS 413 a solution of the residue containing the secretin still evokes a rapid secretion of bile. The fact that the same hormone excites the formation both of pancreatic juice and bile is obviously related to that common action of the two juices in digestion on which we have already dwelt. When food passes into the stomach, there is at once a sharp rise in the rate of secretion of bile. A maximum is reached from the fourth to the eighth hour — that is, while the food is in the intestine. There is then a fall, succeeded by a second smaller rise about the fifteenth or sixteenth hour, from which the secretion gradually declines to its minimum. Upon the whole, the curves of secretion of pancreatic juice and bile show a fairly close correspondence, except that the latter is more nearly continuous. But when we compare the curves representing the rate at which the bile actually enters the intestine with the curve of pancreatic secre- tion (Fig. 169), we are struck by their almost absolute parallelism. This lends additional support to the conclusion deduced from their chemical and physical properties, that in digestion they are partners in a common work. While the rate at which bile passes into the intestine seems to be influenced by digestion much in the Same way as the rate Of pancreatic sent the hourly rate of pancreatic Secretion, the details are as yet less secretion, and the lower the rate at pvartlv krmwn Tn thp fasting which the bile enters the intestine ; ,tiy Known. „„& *. «'. milk diet; b, b', meat; c, c', animal no bile enters the gut. When bread. Only the general form of food is taken, the flow begins after the curves is to be compared. The a definite interval, which varies for scale of the ordi nates of the various ,, , , curves was not the same. the different kinds of food. As long as digestion lasts bile continues to escape, but both the quantity and quality depend upon the nature of the food. Water, raw egg-white, and starch paste, whether given by the mouth or introduced directly into the stomach of a dog, cause no flow of bile. But fat, the extractives of meat, and the products of digestion of egg-white produce a copious discharge. This discharge may be determined by the relatively large amount of acid chyme passed through the pylorus when proteins are digested in the stomach and the stimulus to the formation of secretin occasioned by the presence of this chyme or of fatty material in the duodenum. In the case of fat a further favourable influence on the secretion of bile is the absorption of bile-salts which accompanies the absorption of the fatty acids and soaps produced in fat digestion. Bile-salts stimulate the secretion of bile, including bile-salts themselves. An increased 4*4 DIGESTION flow of bile-salts into the intestine accelerates the splitting of fats by the pancreatic juice, and therefore the absorption of bile- salts acting as solvents for, or chemically united to, the fatty acids and soaps. A circle analogous to the ' vicious circle ' of the logicians, but con- stituting a physiological adaptation of most potent virtue in the digestion of fats, is thus established. Not only is the quantity of bile poured into the intestine increased on a diet rich in fat, but it is said that a given amount of it aids the fat-splitting action of the pancreatic juice more powerfully than if the diet were poor in fat. This may depend upon an increase in the concentration of the bile- salts in bile secreted when a large amount of fat is ingested. But it is well to recognize that we do not at present know with any great exactness the mechanism by which the rate of secretion and ex- pulsion of bile and the properties of that juice are influenced by digestion. It has been conjectured that the first abrupt rise may be started by reflex nervous action, and that later on secretin and, in the case of fat digestion, bile-salts may directly excite the hepatic cells. The pressure under which the bile is secreted is higher than the pressure of the portal blood, and therefore the liver ranges itself with the high-pressure salivary glands rather than with the low-pressure pancreas. But although the biliary pressure is high relatively to that of the blood with which the secreting cells are supplied, it is absolutely low, the maximum being no more than 25 mm. of mer- cury.* This is a point of practical importance, for a comparatively slight obstruction to the outflow, even such as is offered by a con- gested or inflamed condition of the duodenal wall about the mouth of the duct, may be sufficient to cause reabsorption of the bile through the lymphatics, and consequent jaundice. Of course, complete plugging of the duct by a biliary calculus is a much more formidable barrier, and inevitably leads to jaundice, just as ligature of a salivary duct, in spite of the great secretory pressure, inevitably causes oedema of the gland. The Influence of Nerves on the Secretion of Intestinal Juice. — As to the influence of nerves on the secretion of the succus entericus, our knowledge is almost limited to a single experiment, and that an in- conclusive one. Moreau placed four ligatures on a portion of the small intestine, so as to form three compartments separated from each other and from the rest of the gut. The mesenteric nerves going to the middle loop were divided, and the intestine returned to the abdomen. After some time a watery secretion was found in the middle compartment, little or none in the others. This is a true ' paralytic ' secretion, and not a mere transudation depending simply on the vascular dilatation caused by section of the vaso- * In the dog, cat, and monkey the average maximum pressure at which as much bile is secreted as is taken up from the bile-paths by the portal lymphatics is about 300 mm. of bile. The highest pressure recorded was 373 mm. of bile in a cat (Herring and Simpson). INFLUENCE OF NERVOUS SYSTEM ON DIGESTIVE GLANDS 415 constrictor nerves, for it has the same composition and digestive action as normal succus entericus obtained from a fistula. The secretion begins about four hours after section of the nerves, goes on increasing for about twelve hours, and then rapidly diminishes, so that after about two days the middle loop, as well as the other two, will be found empty. The interpretation usually put upon the experiment is that nerves which normally inhibit the local secre- tory mechanism have been divided. But there is no real proof of the existence of such nerves. The same adaptation is seen in the secretion of the succus entericus as in the secretion of the other digestive juices, and the adaptation is naturally most striking in regard to those points in which the intestinal juice is peculiar. While mechanical stimulation of the stomach is ineffective as regards the secretion of gastric juice, mechanical stimulation of the intestine, as by the contact of a cannula, produces a free flow of succus entericus. The reaction is a localized one, the secretion only taking place from the portion of the mucous membrane stimulated. This fact acquires significance when we reflect that the food moves very slowly in the intestine, and a secretion could be of use only at the points where the food happened to be. The juice secreted in response to mechanical stimulation is poor in enterokinase. But if a little pancreatic juice be put into the intestine, and left there for some time, the juice afterwards secreted is rich in enterokinase. Summary. — Here let us sum up the most important points relat- ing to the secretion of the digestive juices. They are all formed' by the activity of gland-cells originally derived from the epithelial lining of the alimentary canal. The organic constituents or their precursors (including the mother-substances of the ferments) are prepared in the intervals of rest — absolute in some glands, relative in others — and stored up in the form of granules, which during activity are moved towards the lumen of the gland tubules, and there discharged. The nerves of the salivary glands are, as regards their origin, (a) cerebral, (b) sympathetic ; the former group is vaso-dilator, the latter (usually] vaso-constrictor ; both are secretory. Secretion of saliva depends strictly on the nervous system. That nerves influence the gastric and pancreatic secretions is also made out. The psychical secre- tion is of greater importance for the saliva and gastric juice than for the pancreatic juice. The direct action of secretin (produced in the intestinal mucous membrane by the influence of the chyme) is the most characteristic factor in pancreatic secretion. As regards the intestinal glands and the liver, it has not been proved that their secretive activity is under the control of the nervous system, except in so far as the latter may indirectly govern it through the blood-supply, although various circumstances suggest the probability of a more direct action. All the digestive juices show a certain adaptation to the nature of the food, 4i 6 DIGESTION although it has not been demonstrated that this is due to a specific sensi- bility of the mucous membranes for each kind of food-stuff. The action of one juice on the secretion of another is also of great significance. Thus, the water of the saliva directly excites a flow of gastric juice when it reaches the stomach ; the acid of the gastric juice excites a fiow of Pancreatic juice when it reaches the duodenum; and the pancreatic juice excites the intestinal mucous membrane to the production of enter okinase, the most characteristic constituent of the succus entericus. In all the glands the blood-flow is increased during activity ; in some (salivary glands) this is known to be caused through vaso-motor nerves. In the salivary glands electro-motive changes accompany the active state, and more heat is produced. Both in the salivary glands and the Pancreas it has been shown that much more carbon dioxide is given off, and much more oxygen used up, during secretion than during rest. In the other glands we may assume that the same occurs. This is one proof that work is done in the separation or manufacture of the con- stituents of the various secretions. SECTION VI. — SURVEY OF DIGESTION AS A WHOLE. Having discussed in detail the separate action of the digestive secretions, it is now time to consider the act of digestion as a whole, the various stages in which are co-ordinated for a common end. The solid food is more or less broken up in the mouth and mixed with the saliva, which its presence causes to be secreted in consider- able quantity. Liquids and small solid morsels are shot down the open gullet without contraction of the constrictors of the pharynx, and reach the lower portion of the oesophagus in a comparatively short time (fa second); while a good-sized bolus is grasped by the constrictors, then by the oesophageal walls, and passed along by a more deliberate peristaltic contraction. Chemical digestion in man begins already in the mouth, a part of the starch being there converted into dextrins and sugar (maltose), as has been shown by examining a mass of food containing starch just as it is ready for swallowing (p. 456). This process is no doubt continued during the passage of the food along the oesophagus. The first morsels of a meal which reach the stomach find it free from gastric juice, or nearly so. They are alkaline from the ad- mixture of saliva ; and the juice which is now beginning to be secreted, in response to the psychical excitement, and reflexly through the presence of the food and the water of the saliva in the stomach, is for a time neutralized, and amylolytic digestion still permitted to go on. For 20 to 40 minutes after digestion has begun there is no free hydrochloric acid in the stomach, although some is combined with proteins, and during this period the ptyalin of the swallowed saliva will be able to act even better than in the mouth, being favoured by SURVEY OF DIGESTION AS A WHOLE 417 a weakly acid reaction. Indeed, for a time, as the meal goes on, the successive portions of food which arrive in the stomach will find the conditions more and more favourable for amylolytic digestion. But as the acidity continues to increase, the activity of the ptyalin will first be lessened, and ultimately abolished; and, upon the whole, a considerable proportion of the starches must usually escape com- plete conversion into sugar until they are acted upon by the pan- creatic juice. This is particularly the case with unboiled starch, as contained in vegetables which are eaten raw ; and, indeed, we know that sometimes a certain amount of starch may escape even pan- creatic digestion, and appear in the faeces. Meanwhile, pepsin and hydrochloric acid are being poured forth; the latter is entering into combination with the proteins of the food ; and before the end of an ordinary meal peptic digestion is in full swing. The movements of the pyloric end of the stomach increase, and eddies are set up in its contents, which carry the morsels of food with them, and throw them against its walls. In this way not only are the contents thoroughly mixed, and fresh portions of food constantly brought into contact with the gastric juice secreted mainly in the more passive cardiac end, but a certain amount of mechanical disintegration is brought about. This is aided by the digestion of the gelatin-yielding con- nective tissue which holds together the fibres of muscle and the cells of fat, and the digestible structures in vegetable tissue which enclose starch granules. Such nucleo-proteins as come into contact with the gastric juice will be split up and the proteins digested to peptone. The globin of the blood pigment will undergo the same change, while the haematin is not much affected. If milk has formed a portion of the meal, the caseinogen will have been curdled soon after its entrance into the stomach, by the action of the rennet ferment alone (see p. 353) when the milk has been taken at the beginning of digestion before the gastric contents have become distinctly acid, by the acid and ferment together when it has been taken later. The caseinogen and other proteins of milk, like the myosinogen and other proteins of meat, and the globulins, albumins, and other proteins of bread and of vegetable food in general, are acted upon by the pepsin and hydrochloric acid, yielding ultimately peptones; while variable quantities of these proteins and of the acid-albumin and proteoses derived from them may escape this final change, and pass on as such into the duodenum. In the dog, indeed, a very large proportion of a meal of flesh has been found to be digested to the peptone stage while still in the stomach, leaving for the juices that act on it in the intestine only its further hydrolysis to amino-acids, etc. But we may safely assume that, in the case of a man living on an ordinary mixed diet, a good deal of the food proteins passes through the pylorus chemically unchanged, or having undergone only the first steps of hydration. For, even a few minutes after food has been •27 418 DIGESTION swallowed, especially liquid food or water, the pyloric sphincter may relax and allow the stomach to propel a portion of its contents into the intestine; and such relaxations occur at intervals as diges- tion goes on, although it is not for several hours (three to five) that the greater portion of the food reaches the duodenum. During this period the acidity has at first been constantly increasing, although for a time the hydrochloric acid has combined, as. it is formed, with the proteins of the food. Then comes a stage where the hydrochloric acid has so much increased that, after combining with all the proteins, some of it remains over as free acid. After a time the total acidity begins to fall, the partially digested proteins continually passing on through the pylorus, while a considerable proportion is so fully digested as to be absorbed by the gastric mucous membrane itself. Thus, in one experiment on the digestion of meat in a dog, it was found that 30 per cent, was absorbed in the stomach, while 40 per cent, passed through the pylorus as peptone, over 20 per cent, as undissolved or soluble protein (acid-albumin), and a little more than 8 per cent, as proteose (Tobler). The large proportion of peptone is noteworthy, as indicating some kind of selective passage of the different digestive products from the stomach into the duodenum. For the gastric contents contain plenty of proteose, although only traces of peptone. The total ' titratable acidity ' goes on diminishing till the third or fourth hour, the proportion of free to combined acid continuing, nevertheless, to rise, since nearly all that is now secreted remains free. In addition to a certain amount of protein, small quantities of soluble and easily diffusible substances, like sugars and some of the organic crystalline constituents of meat — e.g., kreatin — may also be absorbed into the blood by the gastric mucous membrane. The substances which reach the duodenum are — (i) The greater part of the fats. The partial digestion in the stomach of the enve- lopes and protoplasm of the cells of adipose tissue, and of the protein which keeps the fat of milk in emulsion, prepares the fats which are not split up by the gastric juice for what is to follow in the intestine. (2) All the proteins which have not been carried to the stage of peptone, and much peptone. (3) All the starch and dextrins — and glycogen, if any be present — which have not been converted into sugars, and probably a portion of the sugars. (4) Nucleins, haematin, cellulose, and other substances not digestible, or digestible only with difficulty, in gastric juice. (5) The constituents of the gastric juice itself, including pepsin. Most of the pepsin is soon destroyed in the unfavourable environment of the intestinal contents. But it has been shown that a certain amount of active pepsin may be present for a rime in the intestine, even in the free condition, and still more when enclosed in the interior of masses of protein which protect it, and which still continue to be digested by it. This is particularly true of certain materials, like elastin and connective tissue, which are SURVEY OF DIGESTION AS A WHOLE 419 more readily hydrolysed by pepsin than by trypsin. The ptyalin of the saliva has been already destroyed in the stomach. It must be remembered that all this time, even from the beginning of digestion, a certain amount of pancreatic juice has been finding its way into the duodenum in response first perhaps to the psychical excitation, and later to that action of the acid chyme on the in- testinal mucous membrane which has been described. In the duodenum its trypsinogen is becoming activated to trypsin by the enterokinase of the intestinal juice. The secretion of bile, too, has quickened its pace, the gall-bladder is getting more and more full as the meal proceeds and gastric digestion begins, and some of the bile may very soon escape into the intestine. The pylorus opens occa- sionally for a moment whenever the small portions of chyme which at this stage are beginning to pass through have been sufficiently neutralized by the pancreatic juice and bile, although it is not necessary that the reaction should become actually neutral. When the acid chyme, a greyish liquid, turbid with the debris of animal and vegetable tissues— with muscular fibres, fat globules, starch granules, and dotted ducts — gushes through the pylorus and strikes the duodenal wall, the muscular fibres of the gall-bladder contract, and sudden rushes of bile take place from the common duct. By-and-by as bile and pancreatic juice continue to be poured out, the reaction in the duodenum becomes less acid and even weakly alkaline. The observations purporting to show changes in reaction of the intestinal contents at different levels made with indicators like litmus, phenolphthalein, methyl orange, etc., have lost much of their value since the introduction of physico-chemical methods for measuring the hydrogen-ion concentration. However, properly chosen colour indicators can still be employed for estimating the acidity of the gastric contents, at least with sufficient accuracy for most clinical purposes. It must be remembered that the differ- ences in true reaction at different stages of intestinal digestion and at different levels of the gut below the duodenum are slight. There is never a great preponderance either of hydroxyl or of hydrogen ions between the point at which the pancreatic juice and bile are mingled with the gastric chyme and the lower part of the ileum. In the duodenal contents of adult human beings a hydrogen-ion con- centration of 0-00000002 (or ax icrs) normal has been found by the gas chain method (McClendon) and about the same in the intestinal contents of dogs (Auerbach and Pick). This is a slightly alkaline reaction, the hydrogen-ion concentration of pure water being about five times as great (O-ooooooi, or i x 10-7). In the stomach, of course, a very different state of affairs is found. The hydrogen-ion concen- tration rises in the course of i to 3 or 4 hours after a meal to a maxi- mum which is very considerable. In a series of patients, including a number suffering from gastric disorders, the hydrogen-ion concentra- tion ranged from 0*03 to 0-00000007 (or jx io-s). The lowest concen- 420 DIGESTION tration is practically the same as that of water. In other words, in this patient the gastric contents after the test meal were neutral, and it would be impossible for peptic digestion to. proceed. It must be distinctly noted that the acidity of the gastric contents during the digestion of test meals is not the same thing as the acidity of the pure gastric juice. Observations on juice obtained from a case of gastric fistula without admixture with saliva showed that in ' hunger ' juice, where continuous secretion was going on, the hydrogen-ion concentra- tion varied from 0-056 to o-ioo normal in several samples collected on different days (Menten). This question of reaction has significance in two ways : in the first place the reaction determines whether a given ferment shall be destroyed or not by another ferment or by the alkalinity or acidity of the medium. Thus pepsin can be destroyed by the alkali of the pancreatic juice, enterokinase and trypsin by the hydrochloric acid of the gastric juice, trypsinogen by the pepsin and hydrochloric acid. Trypsin has no destructive effect on enterokinase or tryp- sinogen (Mellanby). Secondly, the reaction affects the activity of this or that ferment on the food substances. The optimum hydro- gen-ion concentration for peptic digestion is relatively high (0-02 to 0-03 normal). When it is decreased to o«oooi normal the rate of digestion is only one-half to one-fifth as rapid. The high con- centration of hydrogen-ions in the gastric contents of healthy persons is clearly advantageous, and the low concentration in the gastric contents in cases of hypoacidity clearly disadvantageous. Trypsin acts best in a medium which contains more hydroxyl than hydrogen ions. When the alkalinity is diminished it becomes less active, although it is not entirely inhibited even by an acid reaction until the hydrogen-ion concentration reaches about o-oooi (or IXIQ-*) normal. (See footnote, p. 1139.) In the intestine it is possible that trypsin may perform its work in a medium which is sometimes acid; and although the cause of the acidity and the character of the medium are far from being the same as in the gastric juice, it is obviously an advantage that the chief proteolytic ferment should be able to act upon the proteins in all parts of the intestine and at every stage of intestinal digestion whether the reaction is acid or alkaline. The proteins of the chyme are all carried by the trypsin to the stage of peptone, and the pep- tone, even in perfectly normal digestion, is further split up into amino- and diamino-acids by the trypsin and by the erepsin of the succus entericus. In the lower portions of the small intestine bacteria of various kinds are present and active; and it is not unlikely that even throughout its whole length a certain range of action is permitted to them, checked by the acidity of the chyme, though scarcely by the feeble antiseptic properties of the bile. The lower end of the small intestine is not cut off by any bacteria- proof barrier from the large intestine, in which putrefaction is con- SURVEY OF DIGESTION AS A WHOLE 421 stantly going on. It has been actually shown that small particles, such as lycopodium spores, suspended in water, soon reach the stomach when injected into the rectum. So that micro-organisms, aided by the antiperistalsis of the colon, may be able to work their way above the ileo-colic sphincter and valve, even against the downward peristaltic movement of the small intestine. But even if this were not the case, a few bacteria or their spores, passing through the stomach with the food, would be enough to set up extensive changes as soon as they reached a part of the alimentary canal where the conditions were favourable to their development. In- deed, from the time when the first micro-organism enters the diges- tive tube soon after birth, it is never free from bacteria; and their multiplication in one part of it rather than another depends not so much on the number originally present to start the process, as on the conditions which encourage or restrain their increase. A certain amount of already emulsified fats is broken up into their fatty acids and glycerin in the stomach, unemulsified fats entirely by the fat-splitting ferment of the pancreatic juice. The acids will form soaps with alkalies wherever they meet them in the intestinal contents, or even in the mucous membrane. A portion of those soluble soaps may be immediately absorbed; the rest will aid in the emulsification of the fats not yet chemically decomposed, and thus greatly hasten the fat -splitting action of the pancreatic juice. The phosphatides are in all probability acted upon in the alimentary canal much in the same way as the fats. Lecithin is decomposed by pancreatic and intestinal juice into fatty acids and glyceryl-phosphoric acid, and cholin is liberated. As regards the behaviour of the sterins of the food little is known, but it is not unlikely that their esters are split up, and the sterins thus set free as well as those originally free in the food may then be absorbed, in part at least, without further change. The starch and dextrin which have escaped the action of the saliva are changed into maltose by the amylase of the pancreatic juice, and the maltose into dextrose by the maltase of the same secretion and of the succus entericus. The succus entericus, in addition to its important functions already mentioned, aids as an alkaline liquid in lessening the acidity of the chyme and establishing the reaction favourable to intestinal digestion. It will convert into monosaccharides any cane-sugar, maltose, or lactose, which may reach the intestine; but it cannot be doubted that some cane-sugar may be absorbed by the stomach, after being inverted by the hydrochloric acid of the gastric juice or by inverting ferments taken in with the food, or on its way through the gastric walls. Upon the whole no great amount oi water is absorbed in the small intestine, or at least the loss is balanced by the gain, for the intestinal 422 DIGESTION contents are as concentrated in the doudenum as in the ileum. But as soon as they pass beyond the ileo-caecal valve water is rapidly absorbed, and the contents thicken into normal faeces, to which the chief contribution of the large intestine is mucin, secreted by the vast number of goblet cells in its Lieberkiihn's crypts. Bacterial Digestion. — So far we have paid no special attention to other than the soluble ferments of the digestive tract, although we have incidentally mentioned the action of the lactic acid bacilli on carbo-hydrates and of the fat-splitting bacteria on fats. It is now necessary to recognize that the presence of bacteria is an absolutely constant feature of digestion; and although their action must in part be looked upon as a necessary evil which the organism has to endure, against the consequences of which it has to struggle, and to which in all probability it has to a great extent adapted itself, it is not unlikely that in part it may be ancillary to the pro- cesses of aseptic digestion. But bacteria are not essential (in mam- mals, at any rate, living on milk), as some have supposed. For it has been shown that a young guinea-pig, taken by Caesarean section from its mother's uterus with elaborate aseptic precautions, and fed in an aseptic space on sterile milk, grew apparently as fast as one of its sisters brought up in the orthodox microbic way. The ali- mentary canal remained free from bacteria (Nuttall and Thierfelder). On the other hand, chickens hatched from sterile eggs and kept in a sterile enclosure lived, indeed, for a time, but did not thrive in comparison with the control animals, and died at latest after eighteen days (Schottelius) . It is probable that the difference in the results is to be attributed to the difference in the food, purely vegetable food requiring the aid of bacteria for its proper digestion, especially for the decomposition of the cellulose, while an easily-digestible food like milk does not. Among the more important actions of bacteria on the protein food-products in the intestines may be mentioned the formation of indol, phenol, and skatol, the first having tyrosin for its precursor, and being itself after absorption the precursor of the indican in the urine; the second being to a small extent thrown out with the faeces, but chiefly absorbed and eliminated by the kidneys as an aromatic compound of sulphuric acid; the third passing out mainly in the faeces. The view put forward by Metchmkoff, that in the putrefactive bacteria of the intestine the body carries within itself the seeds of premature decay, owing to the harmful effects of absorbed products of decomposed protein, cannot be looked upon as established, although certainly the prophylaxis suggested by him (the increase of the lactic acid content of the intestine by the addition of sour milk, butter-milk, etc., to the diet) might well be a useful modifica- tion of the dietetic habits of many persons, especially if associated SURVEY OF DIGESTION AS A WHOLE 423 with a reduction in the total amount of protein consumed. That the intestinal contents may include substances capable of inducing severe toxic symptoms if absorbed unchanged scarcely needs proof. Filtered extracts of faeces from normal persons made with salt solution cause, when injected in small amounts into the circulation of dogs, a fall of blood-pressure which may be speedily recovered from or may be quickly fatal according to the specimen (Fig. 170). The large intestine is the chosen haunt of the bacteria of the alimentary canal; they swarm in the faeces, and by their influence, especially in the caecum of herbivora, but also to some extent in man, even cellulose is broken up, the final products comprising certain fatty acids, such as butyric, acetic and valerianic acids, carbon dioxide and marsh gas. A cellulose-dissolving enzyme of great activity is present in the hepatic secretion of the snail, which rapidly produces sugar from that polysaccharide. Dextrose is also formed when it is hydrolysed by dilute acid. Apart from the im- portance of solution of the cellulose in facilitating the action of the digestive juices on the starch and other nutrient materials enclosed by it, it can be assumed that some of the intermediate products of its hydrolysis by the bacteria — e.g., bodies analogous to the dex- trins which appear in the hydrolysis of starch — can be acted on by the ferments of the succus entericus and the pancreatic juice, so as to form dextrose, which on absorption then takes its place in the carbo-hydrate metabolism just as if it had been derived from starch. In the herbivora the contribution thus made to the nutri- tion of the animal may be of considerable importance; in omnivora it is not negligible. In man as much as 40 per cent, of the cellulose of young vegetables is said to be capable of assimilation. In car- nivorous animals, however, it appears that cellulose when taken in the food is quantitatively excreted in the faeces. In addition to the action of the intestinal flora on cellulose, certain of the bacteria of the alimentary canal affect some of the other carbo-hydrates in a not unimportant way. Dextrose, for instance, can be decomposed into two molecules of lactic acid, according to the equation C6H12O6 = 2C3H6O3. This is called the lactic acid fermentation, and is due to a special bacillus. Another micro-organism splits up dextrose into butyric acid, carbon dioxide and hydrogen, the so-called butyric acid fermenta- tion, according to the equation C6H12O6 - C4H8O2 + 2CO2 + 2H2. The contents of the large bowel are generally acid from the products of bacterial action, although the wall itself is alkaline. Faeces. — In. addition to mucin, secreted mainly by the large 424 DIGESTION intestine, the faeces consist of indigestible remnants of the food, such as elastic fibres, spiral vessels of plants, and in general all vegetable structures chiefly composed of cellulose. They are coloured wich a pigment, stercobilin, derived from the bile -pigments. Stercobilin is identical with urobilin, which forms a common, though not an Fig. 170.— Effect of Extract of Faeces on Blood-Pressure. The extract was injected at 2. Time-trace, seconds. invariable, constituent of bile itself. A portion of it is absorbed by the intestine and then excreted in the urine, the urobilin in which is often much increased in fever (' febrile ' urobilin). No bilirubin or biliverdin occurs in normal faeces, although pathologically they may be present. The dark colour of the faeces with a meat diet is due to haematin and sulphide of iron, the latter being formed by the action of the sulphuretted hydrogen which is constantly present in the large intestine on the organic compounds of iron contained in the food or in the secretions of the alimentary canal. A small amount of altered bile-acids and their products is also found; and in respect to these, and to the altered pigments, bile is an excretion. And although its entrance into the upper instead of the lower end of the intestine, the ascertained importance of its function in diges- tion, and the fact that the greater part of the bile-salts is reabsorbed, show that in the adult it is very far from being solely a waste product, the equally cogent fact, that the intestine of the new-born child is filled with what is practically concentrated bile (meconium), proves that it is just as far from being purely a digestive juice. Skatol and other bodies, formed by putrefactive changes in the proteins of the food, are also present in the faeces, and are responsible for the faecal odour. Masses of bacteria are invariably present, and often make up a very considerable proportion of the total faecal SURVEY OF DIGESTION AS A WHOLE 425 solids. Of the inorganic substances in faeces the numerous crystals ot triple phosphate are the most characteristic. When the diet is too large, or contains too much of a particular kind of food, a con- siderable quantity of digestible material may be found in the fasces — e.g., muscular fibres and fat. But it should be remembered that under all circumstances the composition of the faeces differs from that of the food. The intestinal contribution is always an important one, although relatively more important with a flesh than with a vegetable diet. The purin bases normally found in human faeces come both from the food directly and from the metabolism of the tissues. They are increased in amount on a diet rich in purin bodies (such as meat extract or thymus), but are also formed on a diet like milk, from which purin bases cannot be obtained. An interesting constituent of faeces on which light has recently been thrown, especially by the researches of Gardner, is the so-called copro- sterin (dihydrocholesterin), which appears to be produced from cholesterin by reduction, probably under the influence of bacteria, and perhaps also from the phytosterins of vegetable food. CHAPTER VII ABSORPTION SECTION I. — PRELIMINARY PHYSICO-CHEMICAL DATA. Imbibition, or molecular imbibition, is the term applied to the en- trance of liquid into a colloid, without the loss of its properties as a solid, when no preformed capillary spaces are present. The entrance of water into a piece of gelatin, or an epidermic scale, is an example of molecular imbibition. Most animal and vegetable tissues possess this property, which is believed to be of importance in such physiological processes as absorption, secretion, and the excretion of water from the lungs and skin. The process by which liquid passes into a solid with preformed capillary spaces — e.g., a sponge — is sometimes spoken of as capillary imbibition. Diffusion. — When a solution of a substance is placed in a vessel, and a layer of water carefully run in on the top of it, it is found after a time that the dissolved substance has spread itself through the water, and that the composition of the mixture is uniform throughout. The result is the same when two solutions containing different proportions of the same substance, or containing different substances, are placed in contact. The phenomenon is called diffusion. The time required for complete diffusion is comparatively short in the case of a substance like hydrochloric acid or sodium chloride, exceedingly long in the case of albumin or gum. In both it is more rapid at a high temperature than at a low. Osmosis. — If the solution be separated from water by a membrane absolutely or relatively impermeable to the dissolved substance, but permeable to water, water passes through the membrane into the solu- tion. This phenomenon is called osmosis. E.g., a membrane of ferro- cyanide of copper, nearly impermeable to cane-sugar, can be formed in the pores of an unglazed porcelain pot by allowing potassium ferro- cyanide and cupric sulphate to come in contact there. If the pot is filled with, say, a i per cent, solution of cane-sugar, closed by a suitable stopper, connected to a manometer, and then placed in a vessel of water, water passes into it till the pressure indicated by the manometer rises to a certain height. With a 2 per cent, solution it reaches twice this height, and in general the osmotic pressure, as it is called, is in an}' solution proportional to the molecular concentration* of the solution, * The molecular concentration is strictly denned as the number of grammes of the dissolved substance in a litre of the solution divided by the gramme-molecular weight. The gramme-molecular weight, or gramme molecule, is the number of grammes corresponding to the molecular weight. Thus, the gramme-molecular weight of sodium chloride (NaCl) is 58*36 grammes, and of cane-sugar (CjsH^Ojj), 342 grammes. 426 PRELIMINARY PHYSICO-CHEMICAL DATA 427 JO. or, in other words, to the number of molecules of the dissolved substance in a given volume of the solution. If in this sentence we substitute ' gaseous pressure ' for ' osmotic pressure,' and ' gas ' for ' solution,' we have a statement of Boyle's law, which asserts that the pressure of a gas is proportional to its density. Indeed, it has been shown that the osmotic pressure of the dissolved substance is the same as the pressure that would be exerted by a gas, say hydrogen, if all the water were removed, and a molecule of hydrogen substituted for each molecule of the substance, or as would be exerted by the substance itself if, after removal of the solvent, it could be left as a gas filling the same volume. And the osmotic pressure of a solution of one substance is the same as that of a solution of any other substance which contains in a given volume the samo number of molecules of the dissolved substance. In other words, the osmotic pressure is not dependent on the nature, but on the molecular concentration, of the substance. The analogy of the laws of osmotic to those of gaseous pressure becomes still more obvious when it is added that the osmotic pressure of a substance with any given molecular concentration is proportional to the absolute temperature ; and that when a solution contains more than one dissolved substance the total osmotic pres- sure is the sum of the partial osmotic pressures which each substance would exert if it were present alone in the same volume of the solution. The osmotic pressure of a solution may reach an enormous amount. Thus, a i per cent, solu- tion of cane-sugar has a pressure at o° C. of 493 mm. of mercury. A 10 per cent, solution of cane-sugar would have an osmotic pressure of more than six atmospheres, and a 17 per cent, solution of ammonia a pressure of no less than 224 atmospheres. The manner in which the phenomenon known as osmotic pressure is de- veloped is not definitely known. One -Cheory attributes it to the attraction between the dissolved molecules and the molecules of the solvent on the other side of the membrane. The most commonly accepted view is that the osmotic pressure is due to the kinetic energy of the moving molecules. Where the mole- cules are hindered from passing a bounding membrane, the pressure excited by their im- pacts on the boundary is greater than where the membrane is easily permeable, because in the latter case many of the molecules pass through, carrying with them their kinetic energy. The pressure must be still less when a dissolved substance diffuses freely into water ; but however small it may become, it is in the same force which gives rise to the osmotic pressure of the molecules of the dissolved substance that the cause of diffusion must be sought. Recently interest in the nature of the membrane itself as an important factor in osmosis has been revived (Kahlenberg, Armstrong, etc.). There are many facts which indicate that in physiological processes the affinity of the dissolved sub- stances for, or their solubility in, the cell envelopes or the cytoplasm plays an important role. Fig. 171. — Beckmann's Apparatus. For de- scription, see p. sac. 428 ABSORPTION It is as yet impossible or at least very difficult to directly measure the osmotic pressure with accuracy by means of a semi-permeable mem- brane. Recourse is therefore had to indirect methods, especially one which depends on the fact that the freezing-point of a solution is lower than that of the solvent, salt water, e.g., freezing at a lower temperature than fresh water. The amount by which the freezing-point is lowered depends on the molecular concentration of the dissolved substance, to which, as we have seen, the osmotic pressure is also proportional. When a gramme-molecule of a substance is dissolved in water, and the volume made up to a litre, the freezing-point is lowered by r86° C. ; the osmotic pressure is 22-35 atmospheres (16,986 mm. of mercury). It is therefore easy to calculate the osmotic pressure of any solution if we know the amount by which its freezing-point is lowered. A i per cent, solution of cane-sugar, for example, would freeze at about — 0-054° C. Its osmotic pressure = —,^x 16,986=493 mm. of mercury. A convenient apparatus for making freezing-point measurements is shown in Fig. 171. The details of the method are given in the Practical Exercises, p. 529. The osmotic pressure of different solutions may also be compared by observing the effect produced on certain vegetable and animal cells. When a solution with a greater osmotic pressure than the cell-sap (a hyperisotonic solution) is left for a time in contact with certain cells in the leaf of Tradescantia discolor, plasmolysis occurs — that is, the proto- plasm loses water and shrinks away from the cell-wall. If the osmotic pressure of the solution is lower than that of the coloured cell-sap (hypoisotonic solution), no shrinking of the protoplasm takes place. By using a number of solutions of the same substance but of different strength, two can be found, the stronger of which causes plasmolysis, and the weaker not. Between these lies the solution which is isotonic with the cell-sap — that is, has the same molecular concentration and osmotic pressure. The strength of an isotonic solution of some other substance can then be determined in the same way with sections from the same leaf. Animal cells (red blood-corpuscles) may also be employed, the libera- tion of haemoglobin or the swelling of the corpuscles, as measured by the haematocrite (p. 27), being taken as evidence that the solution in contact with them is hypoisotonic to the contents of the corpuscles. Here we may suppose that the impacts of the molecules of the salts of the corpuscle on the inside of its envelope, not being balanced by similar impacts on the outside, tend to distend it, and thus to create a potential vacuum for the surrounding water, which accordingly enters. If the corpuscles shrink, the solution is hyperisotonic to their contents. But since the cells are much more permeable to certain substances than to others, this method does not always yield trustworthy results. Electrolytes. — We have said that the osmotic pressure is proportional to the concentration of the solution, but this statement must now be qualified. For certain compounds, including all inorganic salts and many organic substances, the osmotic pressure decreases less rapidly than the theoretical molecular concentration as the solution is diluted. The explanation is that in solution some of the molecules of these bodies are broken up into simpler groups or single atoms, called ions. Each ion exerts the same osmotic pressure as the molecule did before. The proportion between the average number of these dissociated molecules and of ordinary molecules is constant for a given concentration of the solution and a given temperature. But as the solution is diluted, the proportion of dissociated molecules becomes greater. The bodies which PRELIMINARY PHYSICO-CHEMICAL DATA 429 behave in this way are electrolytes — that is, their solutions conduct a current of electricity; bodies which do not exhibit this behaviour do not conduct in solution. And there are many reasons for believing that the dissociation of the electrolytes is the essential thing in elec- trolytic conduction. We may suppose that in a solution of an electro- lyte— sodium chloride, for instance — a certain number of the molecules fall asunder into a kation (Na+),* carrying a charge of positive elec- tricity, and an anion (Cl— ), carrying an equal negative charge. These electrical charges, it must be remembered, are not created by the passage of a current through the solution. We do not know how they arise, but the ions must be supposed to be electrically charged at the moment when the molecule is broken up. And the ions of different sub- stances must each be supposed to carry the same quantity of electricity. But since they are all wandering freely in the solution, no excess of negative or of positive electricity can accumulate at any part of it — in other words, no difference of potential can exist. When electrodes connected with a voltaic battery are dipped into a solution of an elec- trolyte, a difference of potential, an electrical slope, is established in the liquid, and the positively charged kations are compelled to wander towards the negative pole, the negatively charged anions towards the positive pole. In this way that movement of electricity which is called a current is maintained in the solution. It is clear that the greater the number of ions, and the faster they move in the solution, the greater will be the quantity of electricity carried to the electrodes in a given time, when the difference of potential between the electrodes, or the steepness of the electric slope, remains constant. In other words, the specific conductivity of a solution of an electrolyte varies as the number of dissociated molecules in a given volume and the speed of the ions. It increases up to a certain point with the concentration, because the absolute number of dissociated molecules in a given volume increases. The molecular conductivity — that is, the conductivity per molecule, or, strictly, the ratio of the specific conductivity to the molecular concen- tration— increases with the dilution, because the relative number of dissociated molecules, as compared with undissociated, increases. At a certain degree of dilution the molecular conductivity reaches its maximum, for all the molecules are dissociated. The ratio of the molecular conductivity of any given solution to this maximum or limiting value is therefore a measure of the proportion between the number of dissociated, and the total number of molecules. The molec- ular conductivity of the salts dissolved in the liquids of the animal body, for the degree of dilution in which they exist there, is such that we must assume them to be for the most part dissociated. Surface Tension. — This is a property of surfaces which is typically illustrated in such instances as a globule of mercury, a drop of water on a greasy slide, or a drop of oil suspended in a liquid with which it does not mix. The tendency of such drops to assume the spherical form when not large enough to be distorted by gravity is due to the fact that the surface layer is under a certain tension in virtue of which it strives to contract and to render the surface of the drop as small as * It has been shown that the chemical atoms themselves are not homo- geneous, but are all built up of simpler particles and possess a certain struc- ture. All atoms, e.g., contain electrons, minute particles charged with negative electricity. The number of electrons in an atom appears to be not far from half its atomic weight. Thus in the carbon atom there are 6 electrons, in the oxygen atom 8, and in the hydrogen atom probably only i. There is evidence that the electrons in the atom are divided into groups or rings one within another (Thomson) . 43o ABSORPTION possible, just as if it were a stretched elastic membrane. The cause of this tension is to be sought in the mutual attraction exerted by mole- cules which are very close to each other. This molecular pull is enormously strong. It has been calculated, for instance, that the so-called internal pressure which is due to it is in the case of water not less than 23,000 atmospheres. In the interior of the drop each mole- cule, being surrounded by other molecules, is pulled by this attractive force equally in all directions — that is to say, on the whole it is not pulled at all, since the pulls of all the surrounding molecules balance each other. At the free surface, on the contrary, the molecules are pulled towards the surface, but not away from it, and the pull of the molecules below the surface layer is not balanced by the pull of mole- cules above it. The resultant tension on this layer is the surface tension. Changes in the amount of this surface tension in the case of a given liquid can be produced by bringing gases, solids, or other liquids into contact with the surface layer — that is, by bringing molecules of other substances so near the surface molecules of the liquid that they can attract them, and so to a greater or less degree, depending upon the nature of the substances, balance the attraction of the molecules beneath the surface of the drop. Another way in which the surface tension can be altered is by changing the temperature. The higher the temperature, the greater is the average velocity with which the molecules of the liquid are moving, and the greater the average distance between the molecules (expressed as the expansion of the liquid) . Increase of temperature therefore causes the molecules, through the kinetic energy of heat imparted to them, say, from an external source, to repel each other, and to that extent counteracts their mutual attrac- tion. Accordingly, at the surface the tension, which, as stated, depends upon the excess of this attraction acting towards the interior of the drop, will be diminished. When the temperature is diminished the surface tension will increase. The surface tension can also be altered by altering the electrical charge on the surface. An instance of this is described on another page in connection with the capillary electrom- eter (p. 702). In such ways, then, the surface tension at the inter- face where the cells lining the intestine come into contact with the contents of the gut or with the tissue lymph, or at the interfaces within the cells where solid and liquid ' phases ' come into contact with each other or where different liquids touch, may undergo alterations in either direction. If the tension of the surface is alte-red. the surface energy or power of doing work inherent in the existence of this tension will, of course, be altered too. In this way the energy, or a portion of it, which is unquestionably expended in absorption may be supplied ultimately at the expense of the chemical energy of cell constituents or of food substances on their way through the cells, by means of which the original surface tensions are restored. It has been surmised that changes of surface tension may also be concerned in the secretion of glands, in muscular contraction (p. 744), and other functions (Macallum, Bernstein, etc.). Adsorption. — Connected with the peculiar properties of surfaces referred to in the last paragraph are certain phenomena spoken of as adsorption phenomena. Adsorption is typically seen when a solid in such a form that the surface is greatly increased (e.g., a fine powder or a colloidal solution in which the substance is suspended in the form of exceedingly small particles) is placed in contact with a gas or a solution. There occurs a diminution in the concentration of the gas or the dissolved substance, and a corresponding accumulation of it on the suvl'ace of the solid. Equilibrium is rapidly established, and the MECHANISM OF ABSORPTION 431 characteristic thing about adsorption is that at the equilibrium-point the concentration of the dissolved substance (or the gas) on the surface is immensely greater than in the general mass of the solution. The con- centration on the surface can, indeed, be increased by increasing the concentration of the solution, but in a far smaller proportion. Accord- ing to the thermodynamic law enunciated by Willard Gibbs, sub- stances which diminish the surface tension must tend to accumulate at the surface, and substances which increase the surface tension must tend to diminish in concentration at the surface. If a small quantity of a substance diminishes the surface tension at a given surface more in proportion than a larger quantity, not only will there be an accumula- tion of the substance at the surface, but this will be proportionally greater for small than for larger concentrations of the substance in the solution. This characteristic feature of adsorption may thus depend entirely on surface forces due to the conditions under which the attrac- tion of the molecules for each other acts at the surface. It has not been shown, however, that chemical forces due to the interaction of the electrically charged ions are not also concerned. What is especially important to point out is that in the tissues of the body there is a great development of surfaces. The cell walls or cell envelopes come into contact with the tissue lymph or the contents of the digestive tube or the secretions in the alveoli of glands on the one side, and the cell contents on the other, and constitute in the aggregate an immense surface. The surfaces separating the nuclei from the cytoplasm, and, above all, the surfaces of the particles of the contents of cytoplasm and nuclei suspended in colloid solution, offer prodigious opportunities for such surface phenomena as adsorption. SECTION II. — MECHANISM OF ABSORPTION. In the preceding chapter we have traced the food in its progress along the alimentary canal, and sketched the changes wrought in it by digestion. We have next to consider the manner in which it is absorbed. Then, for a reason which has already been explained, instead of following its fate within the tissues, until it is once more cast out of the body in the form of waste products, it will be best to drop the logical order and pick up the other end of the clue — in other words, to pass from absorption to excretion, from the first step in metabolism to the closing act, and afterwards to return and fill in the interval as best we can. Comparative. — And here, first of all, it should be remembered that the epithelial surfaces, through which the substances needed by the organism enter it, and waste products leave it, are, physiologically con- sidered, outside the body . The mucous membranes of the alimentary, respiratory, and urinary tracts are in a sense as much external .as the fourth great division of the physiological surface, the skin. The two latter surfaces are in the mammal purely excretory. Absorption is the dominant function of the alimentary mucous membrane, but a certain amount of excretion also goes on through it. The pulmonary surface both excretes and absorbs, and that in an equal measure. But it is by no means necessary that the surface through which oxygen is taken in and gaseous waste products given off should be buried deep in the body, and communicate only by a narrow channel with the exterior. 432 ABSORPTION In the frog the skin is largely an absorbing as well as an excreting surface ; oxygen passes freely in through it, just as carbon dioxide passes freely out. In most fishes, and many other gill-bearing animals, the whole gaseous interchange takes place through surfaces immersed in the surrounding water, and therefore distinctly external. In certain forms it has even been shown that the alimentary canal may serve con- spicuously for absorption and excretion of gaseous, as well as liquid and solid substances. Still lower down in the animal scale, the surface of a single tube may perform all the functions of digestion, absorption and excretion. Lower still, and even this tube is wanting, and every- thing passes in and out through an external surface pierced by no per- manent opsnings. Indeed, even in man the functions of the various anatomical divisions of the physiological surface are not quite sharply marked off from each other. Though gaseous exchange goes on far more readily through the pulmonary membrane than anywhere else, swallowed oxygen is easily enough absorbed from the alimentary canal and carbon dioxide given off into it ; and to a small extent these gases can also pass through the skin. Though water is excreted chiefly by the skin, the pulmonary and the urinary surfaces, and on the whole absorbed chiefly from the digestive tract, there is no surface which in the twenty-four hours pours out so much water as the mucous membrane of the stomach. Under normal conditions, it is true, by far the greater part of this is reabsorbed in the intestine, yet in diarrhoea, whether natural or caused by purga- tives, the intestines themselves may, instead of absorbing, contribute largely to the excretion of water. Again, although the solids of the excreta are normally given off in far the greatest quantity in the urine and faeces (only part of the latter is truly an excretion, since much of the faeces of a mixed diet has never been physiologically inside the body at all), yet salts and traces of urea are constantly found in the sweat, and salts and mucin in the excretions of the respiratory tract. Further, although the solids and liquids of the food are usually taken in by the alimentary mucous surface, it is possible to cause substances of both kinds to pass in through the skin; and a certain amount of absorption may also take place through the urinary bladder. So that really it may be considered, from a physiological point of view, as more or less an accident that a man should absorb his food by dipping the villi of his intestine into a digested mass, rather than by dipping his fingers into properly prepared solutions, as a plant dips its roots among the liquids and solids of the soil ; or that he should draw air into organs lying well in the interior of his thorax, instead of letting it play over special thin and highly vascular portions of his skin ; or that the surface by which he excretes urea should be buried in his loins, instead of lying free upon his back. It has been already explained that, although digestion is a necessary preliminary to the absorption of most of the solids of the food, we are not to suppose that all the food must be digested before any of it begins to be absorbed. On the contrary, the two processes go on together. As soon as any peptone, or, at least, any amino-acids, have been formed from the proteins, or any dextrose from the starch, they begin to pass out of the ali- mentary canal ; and by the time digestion is over, absorption is well advanced. Even in the mouth it has already begun, although the amount of MECHANISM OF ABSORPTION* 433 absorption here is quite insignificant, and it is continued with greater rapidity in the stomach. Here a not inconsiderable part of the proteins — at least, in the easily digested form of animal food — a certain amount of the sugar representing the carbo-hydrates and diffusible substances like alcohol, and the extractives of meat, which form an important part of most thin soups and of beef-tea, are undoubtedly absorbed. Water is very sparingly taken up by the stomach. It is in the small intestine that absorption reaches its height. The mucous membrane of this tube offers an immense surface, multiplied as it is by the valvulae conniventes, and studded with innumerable villi. Here the whole of the fat, much sugar, proteose and peptone, or rather the products of the further action of the ferments of the intestine on these derivatives of the native proteins, and certain constituents of the bile are taken in. In the large intestine, as has been already said, water and soluble salts are chiefly absorbed. What now is the mechanism by which these various products are taken up from the digestive tube, and what paths do they follow on their way to the tissues ? Theories of Absorption. — Not so very long ago it was supposed by many that the processes of diffusion, osmosis and nitration offered a tolerably complete explanation. of physiological absorption. At that time the dominant note of physiology was an eager appeal to chemistry and physics to ' come over and help it ' ; and as new facts were dis- covered in these sciences they were applied, with a confidence that was almost na'ive, to the problems of the animal organism. The phenomena of the passage of liquids and dissolved solids through animal membranes, upon which the work of Graham had cast so much light, seemed to find their parallel in the absorptive processes of the alimentary canal. And when digestion was more deeply studied, facts appeared which seemed to show that its whole drift was to increase the solubility and diffusibility of the constituents of the food. But as time went on, and more was learnt of the phenomena of absorption and the powers of cells, these crude physical theories broke down, and discarded ' vital- istic ' hypotheses began once more to arouse attention. Then came the investigations of De Vries, Van 'T Hoff, and others in the domain of molecular physics, which gave to our notions of osmosis the precision that was wanted before its relation to many physiological processes could be profitably discussed. At the present time it must be admitted that we possess no full explanation of absorption, none which is much more than a confession of ignorance, and does not itself need to be explained. Yet some progress has been made at least in defining the boundary between what is clearly known and what is still dark, and in showing that familiar physical processes are not without influence. Some physiologists, impressed with the vast progress of physics and chemistry, believe that it will eventually become possible to explain on mechanical and chemical principles all the peculiar phenomena which we observe in the passage of substances through the walls of the ali- mentary canal. As an aid to the framing of practical working hypo- theses this attitude has everything in its favour. Others, taking account of the number and nature of these peculiarities, oppressed with the perennial paradox of vital action, incline to the less sanguine view, 28 434 ABSORPTION that after all physical explanations have been exhausted, the real secret of the cell will still lurk in some ultimate ' vital ' property of structure or of function, and still elude our search. The only sense in which this attitude can be said to be a useful one is that it presents a standing protest against the acceptance of superficial ' physical ' explanations merely because they are physical. Both the .optimists and the pessimists, the adherents of the physical and the adherents of the vitalistic hypothesis, have, unfortunately, as a rule taken up an extreme position, as if their theories were mutually exclusive. They agree, however, in admitting that the phenomena of absorption are essentially connected with the cells that line the alimentary canal, and not with any more or less inert ' cement ' substance between them. But the one school must confess what the other proclaims, that while the processes carried on in these cells are definite, well ordered, and evidently guided by laws, these laws have as yet for the most part denied themselves to the modern physiologist, with chemistry in one hand and physics in the other, as they denied them- selves to his predecessor, equipped only with his scalpel, his sharp eyes, and his mother- wit. So that in the present state of our knowledge all we can really say is that, while absorption is certainly aided by physical processes, like osmosis and diffusion, possibly by physical processes like imbibition, and is very likely not unrelated to the molecular proper- ties of surfaces (surface tension, adsorption), it is at bottom the work of cells with a selective permeability which we do not fully understand, or at least which we cannot as yet explain in terms of known physical processes acting through a membrane of known physico-chemical structure. Thus, dissolved substances pass with equal ease in either direc- tion through an ordinary diffusion membrane, but in general they pass, with the water in which they are dissolved, more readily out of the intestine than into it. This normal direction of the stream is still maintained for a considerable time after stoppage of the circulation, provided that the intestine is kept in good condition — for example, by being suspended in well-oxygenated blood. Water or solutions of sodium chloride or sugar disappear from the lumen. And this is not due to mere imbibition by the intestinal wall, but the liquid is actually transported across it. The theory that liquids might be taken up from the gut by imbibition, and the water then mechanically removed by the blood flowing on the other side of the imbibing cells, is incompatible with this experiment (Cohnheim). Nor is it necessary that differences of concentration of the dissolved substances on the two sides of the absorbing intestinal membrane, which would permit osmosis and diffusion to go on, should exist. When the excised intestine of a holothurian was filled with sea- water and suspended in the same sea-water, its contents continued to diminish in bulk for hours or entirely disappeared. Here a liquid identical in composition and concentration with the external liquid was moved in a definite direction across the wall of the intestine from the lumen to the exterior surface. In like manner, when a piece of intestine from a newly-killed rabbit is stretched across a vessel of salt solution so as to divide if into two separate compart- ments, the solution continues for a while to leave the compartment MECHANISM OF ABSORPTION 435 to which the mucosa is turned, and to accumulate in the other. In the cases mentioned the transportation of the liquid depends upon the survival of those properties of the mucous membrane which characterize its elements as living cells. For when the cells that line the intestine are injured or destroyed, or subjected to the action of certain poisons, absorption from it is diminished or abolished. In the dead intestine the characteristic set of the tide out of the lumen across the mucosa can no longer be observed. It must be remarked further, in this connection, that in its normal state the mucous membrane does not take up indiscriminately all kinds of diffusible substances, or absorb those which it does take up in the direct ratio of their diffusibility. Nor does it reject everything which does not diffuse. Albumin, for example, which does not pass through dead animal membranes, is to a certain extent taken up from a loop of intestine without change. Cane-sugar (after inversion) and dextrose are absorbed more rapidly than their velocity of diffusion would indicate, when compared with inorganic salts. Glauber's salt diffuses in water fifteen times as fast as cane- sugar, but cane-sugar is absorbed from the intestines ten times faster than Glauber's salt. The velocity of absorption is different even for simple stereoisomeric sugars — i.e., sugars whose molecule, with the same number of atoms combined in the same way, has a dif- ferent form (Nagano). Nor is there any clear relation between the rate of absorption of the various sugars and their osmotic pressure. Dextrose and cane-sugar are always absorbed in greater amount than lactose from solutions of the same osmotic pressure. Indeed, as we shall see, lactose is practically not taken up at all as such (p. 445), and in concentrated solutions may even cause a reversal of the normal movement of water, and act as a purgative. Even the water, organic and inorganic solids of the serum of an animal, are absorbed from a loop of its intestine when the blood-pressure in the capillaries of the intestinal wall is considerably greater than the pressure in the cavity of the gut. Since the serum in the intestine and the plasma in the capillaries must be isotonic, and practically identical in chemical composition, the absorption cannot be due to ordinary osmosis or diffusion. Nor can it be due to filtration, since the slope of pressure is from the capillaries to the lumen of the gut (Reid). It is therefore extremely difficult to reconcile this experiment with any purely physical theory of absorption. The same investigator, summing up the result of careful experiments on the absorption of weak solutions of glucose, concludes that, ' with the intestinal membrane as normal as the experimental procedure will permit, phenomena present themselves which are as distinctly opposed to a simple physical explanation as those previously studied in the absorption of serum.' There is also evidence that even during the absorption of liquids 436 ABSORPTION which undergo no chemical changes in the gut — e.g., salt solutions of different kinds and different concentrations— chemical energy must be transformed, and on no mean scale, in the intestinal mucosa, for the consumption of oxygen and the production of carbon dioxide by the intestine is markedly increased (Brodie and Vogt). It may be taken, then, as quite certain that in absorption from the alimentary canal an essential factor is the activity of the living cells of the mucosa, which, in some way at present unknown, main- tain the ' set of the tide ' from the lumen to the bloodvessels (or lymph spaces), whether the slope of concentration of the dissolved substances favours or opposes it, or when the concentration is the same on both sides of the membrane. It is even probable that this action of the cells is much the most important of all the factors involved. It would be highly misleading, however, to assume that, because this is so, other factors — osmosis, diffusion, possibly even nitration due to differences of pressure caused by the intestinal movements or the contractions of the muscular fibres of the villi (p. 443) — are of necessity negligible. On the contrary, these other factors cannot be adequately taken account of, nor can there be any possibility of assigning to them their proper value until it is recognized that their influence in absorption from the digestive tract is never under ordinary conditions expressed as a simple and uncomplicated effect, such as may be observed in experiments with dead membranes, but, on the contrary, is constantly overlaid, thwarted, or totally reversed, by the special action of the cells. For this reason the discussion of the mechanism of absorption under the time-honoured captions of ' mechanical or physical theory ' versus ' physiological or vital theory/ as if the process must of necessity be purely ' physical ' or purely ' vital,' has lost interest. It may be confidently assumed, indeed, that just because the physiological factor is so dominant, the familiar physical forces must often appear to exert a smaller influence than is really the case, and that, could we disentangle the currents which they create and sustain from that steady drift of material out of the lumen of the gut maintained at the expense of its chemical energy by the still unknown machinery of the cells, we should be impressed with the magnitude rather than the insignificance of their total effect. The following attempt by Hober to analyze on these lines an old experiment of Heidenhain, which the latter observer had interpreted as showing that diffusion and osmosis play no essential part in absorp- tion from the alimentary canal, is of interest in this connection. A loop of small intestine was tied off at both ends, but its circulation was not otherwise interfered with. Solutions of sodium chloride of dif- ferent strengths were introduced into the loop, and in each observation after fifteen minutes the contents of the loop were recovered and analyzed for the chloride. MECHANISM OF ABSORPTION 437 Introduced. Recovered, c.c. Parcentage of NaCl. Total NaCl in Grammes. C.C. Percentage of NaCl. Total NaCl in Grammes. I2O 0-30 0-36 18 O-6O O'loS 120 0-50 0-60 35 0-66 0-230 II7 i-oo 1-17 75 0-90 0-670 120 1-46 i'75 109 1-20 I-3IO Here it is seen that, from the markedly hypotonic solutions of the first two observations, sodium chloride was absorbed, of course, along with much water. But from the strongly hypertonic solution of the last observation some water was also taken up, instead of water passing into it from the blood. The suggested explanation of these and other data yielded by the experiment is that an osmotic stream of water out of the loop into the blood is established on introduction of the hypo- tonic solutions, which raises their percentage of salt, while in the case of the hypertonic solution a diffusion stream of sodium chloride is estab- lished in the same direction, and its salt content falls. The volume of the hypotonic solutions in the loop rapidly diminishes because the osmotic current conspires with the normal ' physiological ' drift from the lumen outwards. In this drift both salt and water are involved, as if the cells were filters which maintained through expenditure of their own energy a slope of hydrostatic pressure from free surface to depth. The hypertonic solution diminishes only slowly in bulk, be- cause the ' physiological ' current out of the lumen is opposed by the osmotic stream of water into the lumen. Nevertheless, even the hyper- tonic solution is gradually absorbed, because the pull of the cells — the suction, if it may be so expressed, of the cellular pump — is powerful enough to overcome the osmotic current and to force water up the slope into the blood or into the tissue liquid, whose osmotic pressure is not much more than half that of the solution. Permeability of the Intestinal Epithelium and Lipoid-Solubility of Absorbed Substances. — If the. cells of the intestinal mucosa are to move materials out of the lumen of the gut, it is obvious that these materials must first be able to enter the cells. The ease or difficulty with which different substances are absorbed may there- fore depend upon the degree in which the cells are permeable for them. The question of the factors on which the permeability of cells depends has already been discussed to some extent in the case of the coloured blood-corpuscles. A famous theory attributes the degree of permeability of erythrocytes and many other animal and plant cells to given substances to the degree of the solubility of these substances in the lipoids which are supposed to form the essential constituents in the outer layer or envelope of cells (Overton) Substances which are readily soluble in lipoids are supposed to gain an easy entrance by going into solution in the envelope ; those which are insoluble in lipoids are checked at the boundary. Attempts have been made to apply this theory to the explanation of selective 438 ABSORPTION absorption by the intestine, but without much success. The very fact that the theory is held to apply to practically any cell greatly circumscribes at the outset its power of dealing with cells like those lining the intestine, which are adapted to absorb nutrient materials for all the body, and must necessarily differ in their permeability from cells adapted for some limited function involving only a limited and specialized nutritive exchange. Thus sodium chloride practically does not penetrate the red blood-corpuscles, the muscle fibres, and many other tissue elements. It is a lipoid-insoluble sub- stance, and the lipoid theory says that this is the reason why it penetrates these cells with such difficulty. But sodium chloride must and does penetrate the intestinal mucosa, and with consider- able ease, in order that the body, especially its extracellular liquids, may obtain a sufficient supply of this indispensable material. It is still a lipoid-insoluble substance, and pays no heed to the lipoid theory at all. It is perfectly true that some substances — e.g., ethyl alcohol — which are much more soluble in lipoids than sodium chloride, are also even more readily absorbed from the intestine. It has been stated also that, as regards the velocity of their absorp- tion, the three alcohols, glycerin, erythrite, and mannite, are related to each other in the same way as in regard to their lipoid-solubility. There is, of course, some reason for this, and also some reason why ethyl alcohol is taken up more easily than salt, but we do not know that it has anything to do with lipoid-solubility. If there is a lipoid layer at the free ends of the cells covering the villi, it is very possible that a substance soluble in lipoids may be able to enter cells which would otherwise have denied it entrance. It may even inflict temporary or permanent injury on the cells in doing so, and may thus be taken up in greater amount than by normal cells, and this possibility has to be reckoned with in giving a physiological value to experiments with materials essentially foreign to the intestine, and to which it cannot have developed any adequate adaptation. For the essential food materials it is quite certain that, apart from any general relations of cell envelope and environing liquids which are common to the intestinal and to other cells, special relations of an adaptive nature have been developed between the intestinal cells and the very special liquids, elsewhere unknown in the body, with which they come in contact in the lumen of the gut. It is unlikely that the mucosa has developed a special adaptation for lipoid- soluble food materials; it must have developed an adaptation for such food materials, lipoid-soluble or not, as have been offered to it through countless ages, and as are necessary for the nutrition of the organism. But if it be true that the action of the columnar epithelium of the intestinal mucous membrane in the absorption of the food is in the main a process of selective secretion such as is found in glandular MECHANISM OF ABSORPTION 439 organs, an action which we may perhaps describe as making use of purely physical processes, but not mastered by them, the possi- bility must be admitted that in the cells of endothelial type which line the serous cavities, the lymphatics, the bloodvessels, the alveoli of the lungs, and the Bowman's capsules of the kidney (p. 489), the element of secretion may be less marked, and more overshadowed by the physical factors. And it may very plausibly be urged that changes of considerable physiological complexity can only be wrought on substances that have to pass through a cell of con- siderable depth, while a mere film of protoplasm suffices for, and indeed favours, mechanical filtration and diffusion. We have already seen (p. 264), in the case of the lungs, that whatever the complete explanation may be of the gaseous exchange which takes place through the alveolar membrane, physical diffusion undoubtedly plays an important part. We shall see, too (p. 500), that in the case of the kidney the endothelium of the Bowman's capsule, although by no means devoid of selective power, does seem to have allotted to it a simpler task than falls to the share of the ' rodded ' epithelium. Absorption from the Peritoneal Cavity. — Further, it has been stated that interchange between blood-serum, circulated artificially in the vessels of dogs and rabbits which have been dead for hours, and liquids introduced into the peritoneal cavity, is essentially the same as in the living animal, and can be explained on purely physical principles (Hamburger). But there is one experiment, at any rate, which is certainly difficult so to explain — viz., the absorption from the peritoneal cavity of sodium chloride solution isotonic with the blood-serum, an absorption which goes on with considerable rapidity. Starling has supposed that this is due to the circum- stance that the proteins of the serum exert osmotic pressure, the peritoneal membrane being almost or altogether impermeable for them in comparison to its permeability for the salt solutions. In consequence, water passes into the bloodvessels from the peritoneal cavity. The solution thus becomes more concentrated as regards sodium chloride, some of which accordingly enters the blood by diffusion, and so on. But even isotonic serum is absorbed from the peritoneal cavity, and it seems to savour of special pleading to suggest, as has been done, that this takes place through the lymphatics, and not at all through the bloodvessels. Up to a certain point an increase in the intraperitoneal pressure favours absorption, but beyond this it hinders it by interfering with the circulation. The removal of a portion of the fluid in this con- dition facilitates the absorption of the rest — a fact which has long been applied in the operation of tapping. Ligation of the thoracic duct has little effect on the fate of liquids injected into serous cavities, since the bloodvessels play the chief part in their absorp- tion, just as strychnine, when injected under the skin — i e., into 440 ABSORPTION the lymph spaces of areolar tissue — is taken up by the blood and does not appear in the lymph. But even if we admit that substances can pass, by physical pro- cesses alone, from serous cavities into the blood, and from the blood into serous cavities, this has little bearing upon the question of intestinal absorption. For we can hardly put anything into the peritoneal cavity which is not foreign to it. It was never intended to come into contact with the hundred and one solutions, extracts, suspensions, and what not, which the industrious experimenter has offered to its unsophisticated endothelium. It cannot possibly have developed any high degree of ' selective ' power. In the intestine everything is different. The mucosa is adapted to come into contact with an immense variety of materials, all kinds of food- substances mingled with many kinds of refuse, the products of the action of numerous digestive ferments, and of a vigorous and varied bacterial flora. All these it has to sift and try. It cannot fail to have properties which suggest a severe and searching selection. The difference between a serous cavity and the intestine is well illustrated by the following experiment, in which the changes in the composition of a hypotonic (3 per cent.) solution of dextrose introduced into the peritoneal sac and into a loop of intestine respectively were compared (Cohnheim). Introduced. After Recovered. Peritoneum - 50 c.c. 90 mins. - 19-5 c.c., containing i per cent, dextrose and 0-55 per cent. NaCl. Intestine - - 44 » 25 .. 19 c.c., containing 3-8 per cent, dextrose and 0-04 per cent. NaCl. Here the water and sugar are both taken up from the intestine and the peritoneal cavity; but while the sugar concentration in the serous sac falls markedly, as ought to be the case if the sugar is diffusing into the blood along the slope of concentration, the percentage of sugar in the intestine actually increases. Still more striking is the fact that sodium chloride accumulates in the peritoneal liquid in a concentration obviously tending to equality with that of the blood, as would happen if the peritoneal lining were a dead diffusion membrane. On the other hand, practically no sodium chloride passes into the lumen of the gut. Closely connected with the question of absorption from and secretion (or transudation) into the serous cavities is the question of the factors concerned in the formation of the lymph (which will be considered in the next chapter), even although recent researches throw grave doubt on the common view that these sacs are merely expanded lymph spaces, and indicate that the liquid found in them has a different origin from lymph. ABSORPTION OF THE VARIOUS FOOD SUBSTANCES 44* •>. ••?", i SECTION III. — ABSORPTION OF THE VARIOUS FOOD SUBSTANCES. Absorption of Fat — How the Fat gets into the Intestinal Epithelium. — It has been already mentioned that fat is split up in the intestine into the corresponding alcohols (mostly glycerin) and fatty acids, but it has been a subject of discussion whether it all undergoes this change or only a portion of it. The common view has long been that the greater part of the fat escapes decomposition, and, after emulsification by the soaps formed from the liberated fatty acids, is absorbed as neutral fat by the epithelial cells covering the villi. If an animal is killed during digestion of a fatty meal, these cells are found to contain globules of different sizes, which stain black with osmic acid, are dissolved out by ether, leaving vacuoles in the cell substance, and are therefore fat (Fig. 172). It has always been difficult to explain how droplets of emulsified fat could get into the interior of the epithelial cells, al- though, perhaps, no more difficult than to explain the passage of living tubercle bacilli from the contents of the intestine into the chyle of the thoracic duct — a fact which has been clearly demonstrated (Ravenel) . The fat is certainly contained within the cells, and not between them. When fat is found in the cement sub- stance between the cells, it has been mechanically squeezed out of them by the shrinking of the villi in preparation. This difficulty is obviated if we suppose that the whole of the fat is split up in the intestine, the products being absorbed in solution, the glycerin as such, and the fatty acids either as soaps or in the free state, or partly free and partly saponified. If this is the true theory — and the evidence of its truth has of late years been continually growing — neutral fat must again be built up in the epithelial cells from the absorbed glycerin and the fatty acids or soaps. Now, it has been shown that when an animal is fed with fatty acids they are not only absorbed, but appear as neutral fats in the chyle of the thoracic duct, having combined with glycerin in the intestinal wall; and the epithelial cells contain globules of fat, just as they do when the animal is fed with neutral fat. Further, it is known that fat-splitting goes on in the alimentary canal to a much greater extent than would be necessary merely for the forma- tion of a quantity of soap sufficient to emulsify the whole of the fat in the food. Indeed, at certain stages of digestion most of the Fig. 172. — Mucous Membrane of Frog's Intestine during Absorp- tion of Fat (Schafer). ep, epithe- lial cells; sir, striated border; C, lymph corpuscles; /, lacteal. 442 ABSORPTION fatty material, both in the small and large intestine, has been found to consist of fatty acids. The reversibility of the reaction under the influence of lipase, which has already been alluded to, does not enter into the question so far as fat-splitting in the intestine is con- cerned, for the products of the reaction can be absorbed as quickly as they are formed. To clinch the matter, it has been proved that when mixtures of paraffin and fat, which can be emulsified in a watery solution of sodium carbonate, are eaten, the paraffin is com- pletely excreted with the faeces, while the greater part of the fat is absorbed. And fatty substances which are not easily split up and saponified (for example, lanolin, the fat of sheep's wool, a mixture of compounds of fatty acids with isocholesterin, a substance closely related to cholesterin and allied bodies) are not absorbed even when they are easily emulsified. Even fats with a melting-point far above the temperature of the body can be absorbed after being split up. The palmitate of cetyl alcohol, the chief constituent of sper- maceti, melting at 53° C., was absorbed to the extent of 15 per cent., 85 per cent, being excreted in the faeces. It appeared as palmitin in the chyle of a human being flowing from a fistula, the palmitic acid having been absorbed as such, or as a sodium soap, and having then united with glycerin to form the neutral fat, palmitin. Some observers have endeavoured to show that the fat is absorbed without change by introducing into the intestine fat stained with dyes, such as alkanna red or Sudan III., which are insoluble in water. The stained fat was found in the epithelial cells of the villi, in the lacteals, and, in the case of a patient suffering from chyluria, in the urine. But this evidence is not conclusive, for it has been shown that the pigments might easily have been absorbed after decomposition of the fat, since, although insoluble in water, they are soluble in fatty acids, and therefore to some extent in the intestinal contents, and readily pass into the lymph. As already pointed out, the bile plays an important part in the solution of the fatty acids, which may form loose compounds with the amide group of the bile-acids. In these loose combinations, soluble in water, the fatty acids can be absorbed from the intestinal contents (Pfliiger). In whatever way the fat which can be seen in the epithelial cells during absorption of fat gets into them, it must be carefully noted that there is no quantitative proof that it represents all or even the greater part of the absorbed fat. So far as microscopic observations go, much of the fat may pass through the mucosa in the form of soluble decomposition products without appearing in particulate form in the epithelium. How the Fat gets out of the Intestinal Epithelium. — Leucocytes have been asserted to be active agents in the absorption of fat. They have been described as pushing their way between the epithelial cells, fishing, as it were, for fatty particles in the juices ABSORPTION OF THE VARIOUS FOOD SUBSTANCES 443 of the intestine, and then travelling back to discharge their cargo into the lymph. This view, however, is erroneous. But, although the leucocytes do not aid in the absorption of fat from the intestine, they may take up a certain amount of it from the epithelial cells, and convey it through the spaces of the network of adenoid tissue that occupies the interior of the villus, to discharge it into the central lacteal, where it mingles with the lymph and forms the so- called molecular basis of the chyle. It has been supposed that a part of the fat may reach the lacteal in another way. The con- traction of the smooth muscular fibres of the villus and the peristaltic movements, of the intestinal walls, which alter the shape of the villi, alter as well the capacity of the lacteal chamber, and so alternately fill it from the lymph of the adenoid reticulum, and empty it into the lymphatic vessel with which it is connected. By this kind of pumping action the passage of fat and other substances into the lymphatics may be aided. There is, however, no proof that all the fat accumulated in the intestinal epithelium leaves it without further change. It is quite as probable that the lipase which is known to be contained in the cells again hydrolyses the fat, or a portion of it, and that the constituents then pass out into the lymph, or even in part into the blood. In the dog most of the fat goes into the lacteals, and thence by the general lymph-stream through the thoracic duct into the blood. And in man the chyle collected from a lymphatic fistula contained a large proportion of the fat given in the food (Munk). But this bare statement would be misleading if we did not add that the fat taken in can never be entirely recovered in the chyle collected from the thoracic duct. A small fraction of the deficit might be accounted for as fat directly used up for the nutrition of the intestinal wall itself. But even after ligation of the thoracic and right lymphatic ducts a large proportion of a meal of fat (32 to 48 per cent.) is absorbed from the intestine, obviously by the channel of the bloodvessels, since the fat-content of the blood increases up to, it may be, six times the highest amount present in the blood of fasting animals. The statement that only fatty acids can be absorbed under these conditions is erroneous (Munk and Friedenthal). A dog normally absorbs 9 to 21 per cent, of the fat in a meal in three to four hours; 21 to 46 per cent, in seven hours; and 86 per cent, in eighteen hours (Harley). After excision of the pancreas the absorption of fat is hindered, though not abolished. More fat, indeed, can be recovered from the intestine than is given in the food. This at first sight paradoxical result is explained by the well-estab- lished fact that a certain amount of fat is normally excreted into the intestine. Mechanism of Fat Synthesis in the Intestinal Mucosa. — As to the manner in which the synthesis of the fat in the intestinal epithelium 444 ABSORPTION is accomplished, the most fascinating theory is that which attributes it to the reversed action of lipase, possibly the very same lipase as originally split it up in the intestine. The reversibility of the action of various enzymes under changed conditions, especial] 3^ changes in the relative concentration of the bodies concerned in the reaction, has been previously mentioned. It has been shown, e.g., that the pancreas, intestinal mucous membrane, lymph glands, etc., and even cell-free extracts of these organs have the power of synthesizing the ester ethyl butyrate from butyric acid and ethyl alcohol (p. 338), as well as the power of decomposing the ester into the fatty acid and the alcohol. Moore, however, states that in the case of ordinary fats the synthesis takes place in the intestinal wall only in situ, and while the circulation is going on. In the intestinal mucosa the greater part of the fatty acid is already combined with glycerin as neutral fat, although considerable quantities of free fatty acid are also present. In the lymph coming directly from the mesenteric glands practically the whole of the fatty acids are in the form of neutral fat. An additional, and in some respects even more remarkable, illus- tration of the synthesizing powers of the intestinal wall is the dis- covery of Munk, already referred to (p. 441), that fatty acids given by the mouth appear in the lymph of the thoracic duct as neutral fats, having somewhere or other, in all probability on their way through the epithelium of the gut, been combined with glycerin. Since, however, the amount of neutral fat recovered from the thoracic duct is not equivalent to more than one-third of the fatty acids given, it has been suggested that this synthesis of fat is only apparent, and that the whole of the fat which appears in the chyle after a meal of fatty acids comes from the fat excreted into the intestine (Frank), which is increased when fatty acids are given by the mouth. But the suggestion is more ingenious than the evidence advanced in its support is convincing. And, as we have seen (p. 443), a part of the deficit may be accounted for by absorption directly into the bloodvessels. In concluding our review of the absorption of fat, certain general considerations which have a close relation to the question may be alluded to. There is some reason to think that the lipises are enzymes less finely adjusted to minute differences in the structure of the fats on which they act than other digestive ferments — eg., maltase or lactase, to details in the chemical structure of their substrates. If this be so, a very few lipases, or even a single one, may suffice to accomplish all the enzymatic changes which occur in the fats both in the lumen of the intestine and in all the various tissue cells. At the same time the possible variation in those decom- position products which constitute the ' building-stones ' of the fats is less than in the case of, say, the proteins. Two consequences ABSORPTION OF THE VARIOUS FOOD SUBSTANCES 445 follow as regards the absorption of fat: (i) Each cell may be capable of dealing with the original neutral fats of the food, and of adapting them to its needs by decomposing and resynthesizing them so far as is necessary in its own substance. In this case it would not be necessary for assimilation by the cells that the fats should be com- pletely split or even split at all in the alimentary canal, however important this might be for their absorption from its lumen. (2) If it were necessary for absorption that decomposition of the fats should take place in the lumen of the digestive tube, the whole of the fat, or at any rate such portion of it as was not at once needed, might without disadvantage for the tissues be resynthesized after absorption. It is not difficult to see that it might even be advan- tageous that not only the relatively fixed reserve in the fat cells, but also what might be termed the floating or circulating reserve constituted by the emulsified fat in the blood should be in the insoluble form of neutral fat. Absorption of Carbo- Hydrates. — Carbo-hydrates are normally absorbed from the alimentary canal only in the form of mono- saccharides, such as dextrose, levulose or fructose, and galactose, but especially dextrose. These monosaccharides are readily formed from polysaccharides like starch and dextrin, and the disac- charide maltose, which they yield on digestion with amylase, as well as from disaccharides like cane-sugar and lactose, by the fer- ments already studied. That, as a matter of fact, the hydrolysis in the intestine must convert practically all the carbo-hydrate into monosaccharides before absorption, can be shown in various ways. The ferment lactase, while present in the small intestine of all young mammals, is regularly absent in some mammalian groups in the adult. In other species, including man, it is found in some adults, but not in all. In birds and other animals below the mam- mals, it has not hitherto been found at any age. It has been sur- mised that these differences depend upon the presence or absence of lactose (milk) as a regular constituent of the food. (But see p. 410.) If, now, lactose is introduced into a loop of intestine in an animal which does not possess lactase — e.g., an adult rabbit — it is not absorbed, but remains in the lumen till it is at last decomposed by bacterial action. In animals in which lactase is present the lactose is rapidly absorbed. Maltose is easily taken up from the intestine because of the action of the ferment maltase, which is the most widely spread of all the inverting ferments. The dextrose formed by the maltase is so rapidly absorbed that none, or only traces, of it can be detected in the contents of the intestinal loop. But if absorp- tion be interfered with by injuring the intestine, maltose disappears, and the dextrose produced from it accumulates in the lumen. The reason for the discrimination exercised by the intestinal mucosa in favour of the monosaccharides becomes apparent when an attempt is 446 ABSORPTION made to circumvent it by injecting the sugars parenterally — i.e., into subcutaneous or intramuscular connective tissue, into a serous sac, or directly into the blood. Cane-sugar and lactose so introduced are excreted unchanged in the urine. Dextrose, levulose, and galactose are used up in the body, and some maltose likewise, thanks to the presence of maltase in the blood and tissues. The cells of the body in general wall burn only monosaccharides, and not di- or poly-saccharides. Galactose and fructose are probably first con- verted into dextrose before being utilized by the tissues, a change which can also be readily induced in the test-tube. Therefore the intestine admits the simple, but rejects the more complex sugars. It is only in the presence of abnormally great quantities or ab- normally great concentrations of the sugars which are not directly utilizable that they are to a certain extent taken up unaltered, to be for the most part quickly excreted as such (p. 540). In like manner we have seen that the rtative proteins can, so to speak, force their way by storm through the intestinal mucosa when offered to it in exceptionally large amount. The sugar absorbed from the intestine passes normally into the rootlets of the portal vein, not into the chyle, for no increase in the quantity of that substance in the contents of the thoracic duct takes place during digestion, while the sugar in the portal blood is increased after a starchy meal. The blood of the portal vein of a dog in the fasting condition con- tained 0-2 per cent, of dextrose. During absorption of a meal rich in carbo-hydrates it contained as much as 0-4 per cent. In the lymph issuing from the thoracic duct the amount was the same in both conditions — viz., 0-16 per cent. In a case of lymph (chyle) fistula in a human being, where almost all the lymph from the digestive tract escaped through the fistula, out of 100 grammes of carbo-hydrate taken (50 grammes starch and 50 grammes sugar), only £ gramme, or not I per cent, of the sugar corresponding to the carbo-hydrates of the food, could be recovered in the chyle. But when a large amount of a dilute solution of sugar is introduced into the intestine, some of it is taken up by the lacteals. Absorption of Water and Salts. — The main channel for absorption of these is the bloodvessels of the intestine. As much as 3 to 5 litres of water can be absorbed in a day in the intestine of a healthy man, exceptionally even 6 to 10 litres, without the faeces altering their normal consistence. Absorption of the water and dissolved salts may theoretically take place either through the epithelial cells (intra- epithelial absorption), or between the cells (interepithelial absorp- tion). According to Hober, most metallic salts (silver, mercury, lead, bismuth, copper, manganese, etc.) are absorbed interepitheli- ally, while iron salts form an exception, and pass into the epithelial cells. The distinction between interepithelial and intra-epithelial absorption does not rest upon an absolutely sure foundation. Yet ABSORPTION OF THE VARIOUS FOOD SUBSTANCES 447 it is probable that everything which is useful in the nutrition of the body is taken up by the cells, while such substances as metallic salts which are foreign to the organism, and are denied entrance into the cells, may pass in small amount between them, their passage being perhaps associated with more or less injury to the interstitial substance. The vigilant selection exercised by the mucosa is well illustrated by the facts that, although manganese and iron are chemically so closely related, iron, which is necessary for the formation of the blood-pigment, is absorbed in immensely greater amount than manganese; and that chlorides, especially sodium chloride, are readily taken up, sulphates with difficulty. Iron is absorbed by the bloodvessels, but also to some extent by the lacteals. From the blood it is carried to various organs, especially the spleen and liver. There is reason to believe that the eosinophile leucocytes take some share in its transportation. It was supposed by Bunge that only organic compounds of iron could be absorbed, and that the undoubted benefit derived from the administration of inorganic iron compounds, such as ferric chloride, in chlorosis, was due not to their direct absorption, but to their shielding the organic compounds from the attack of the sulphuretted hydrogen in the intestine (p. 424). But this theory has been shown to be inconsistent with the facts. For instance, after the administration of salts of iron, the iron in the blood, liver, spleen, and other organs increases, but there is no accumula- tion of iron in the liver of an animal to which salts of manganese have been given, although these are equally decomposed by sul- phuretted hydrogen. Absorption of Proteins. — 'The proteins of the food or their digested products also pass directly into the blood-capillaries which feed the portal system. For it has been shown that after ligature of the thoracic duct protein substances are still absorbed from the intestine, and the urea corresponding to their nitrogen appears in the urine. The total nitrogen in the chyle flowing from a fistula of the thoracic duct in a man was not found to be increased during the digestion of protein food. The quantity of chyle escaping in a given time was also unaffected, whereas during the digestion of fats it was greatly augmented (Munk). Although a certain amount of egg-albumin, serum-albumin, alkali- albumin, and other native or slightly altered protein substances can be absorbed as such by the small, and even by the large, in- testine, there is no evidence that, under ordinary conditions, this mode of absorption is of any practical importance in nutrition, although in another relation it may possess a certain interest (p. 32), For when native proteins, with the possible exception of the serum proteins from an animal of the same species, are introduced ' parenterally,' so that they do not reach the tissues by way of 448 ABSORPTION the alimentary canal, they behave in a very different manner from the same proteins when given by the mouth. One notable differ- ence is that the parenterally administered proteins give rise in general to the formation of antibodies — -e.g., specific precipitins^ (p. 31). This is not the case when they are administered per os, unless, like raw egg-white, which, as already mentioned (p. 404), evokes no secretion of gastric juice, they remain long undigested in the alimentary canal, when an amount sufficient to cause the production of precipitins may eventually be absorbed unaltered. This has also been shown by means of the anaphylactic reaction. Secondly, they are not, as a rule, utilized in the metabolism of the body, or are utilized very incompletely. Egg-albumin, for instance, when injected into the blood, is excreted in the urine. It has been previously pointed out that the various proteins differ remarkably not so much in the kinds as in the relative quantities of the amino- and diamino-acids which can be obtained from them (p. 2). This is unquestionably one important reason why the food proteins are — for the most part, at any rate — so thoroughly hydrolysed before absorption. Another may be that it is easier for the intestine to take up the small molecules of the decomposition products than the large colloid aggregates of the original protein solutions. So far as the first reason is concerned, the degree of decomposi- tion need not be the same for all the food proteins, although all must be decomposed, for even among the proteins the products of whose hydrolysis do not exhibit qualitative differences, no two have hitherto been discovered which show the same quantitative relations among the ' building-stones. ' A new house has to be built from the materials of an old one. How far the work of demolition must be carried will depend upon the difference between the plans of the two houses. Sometimes the main part of the old building may be saved, and only the wings require reconstruction. In like manner it is conceivable that the central group or nucleus of the molecule of a given food protein may be identical with that of a given body protein, and that only the side-chains may be so different that they must be broken up and reconstructed. Or, again, the whole architectural plan of the new house may be so distinct from that of the old that the only feasible method is to completely demolish the latter, and then to use the individual bricks in the new construction; just as a protein in the food may differ so radically from a tissue protein into which it is to be transformed that it must be decomposed into the simplest products of proteo- lysis before the reconstruction of the molecule can begin. It is not known what the minimum degree of hydrolysis is which will permit of effective absorption and utilization. But it would seem that it must be very complete. Even a body so simple in comparison with the proteins as the tripeptide (p. 2) alanyl-glycyl-tyrosin, con- 449 taining only three ' building-stones,' can only be changed into the tripeptide alanyl-tyrosyl-glycin by first hydrolysing it into its three components and then synthesizing these afresh. On the other hand, in obtaining the tripeptide glycyl-tyrosyl-alanin from alanyl- glycyl-tyrosin, the dipeptide group glycyl-tyrosin can remain un- decomposed, and it is only necessary to split alanin off and link it up to the dipeptide to obtain the desired glycyl-tyrosyl-alanin (Abderhalden). There can be no doubt that by far the greater part, if not the whole, of the proteins of the food is first changed into proteoses and peptones. But proteose and peptone are absent from the blood, and, indeed, when injected into the blood they are excreted in the urine. When injected in larger amount, they pass also into the lymph, from which they gradually reach the blood again, and are eventually, as before, eliminated by the kidneys. The clear inference is that if they are absorbed as such from the alimentary canal, they must be changed in their passage through its walls. The fact that a portion at any rate of the peptones and proteoses is decomposed into amino-acids, etc., in the lumen of the intestine has been already alluded to. It is certain that this portion is a very large proportion of the whole, although the question how much, if any, of the peptone passes as such from the lumen into the mucosa must still be left undecided. It is true that along with amino-acids peptones are always found in the intestine during the digestion of protein, and the quantity of amino-acids actually present in the lumen at any moment may DC small in proportion to the quantity of peptone. But this is precisely what is to be expected if the peptone as such is incapable of absorption. For the easily absorbed amino-acids will disappear from the gut as fast as they are formed, leaving behind the peptone for further hydrolysis. The fact that all the amino-acids which the proteins are capable of yielding can be detected in the contents of the intestine, including even those which appear late and, as it were, reluctantly in artificial digestion, is a proof that the decom- position of the protein goes fast and far in the alimentary canal. If it is not complete, if some of the partially hydrolysed protein is taken up by the mucous membrane in the form of peptones or possibly even of proteoses, it would seem that this is similarly decomposed by the action of erepsin in the intestinal wall. y It has been stated that during the digestion of a protein meal the mucosa of the stomach and intestine contains proteose and pep- tone, while none is present in the muscular coat or in any other organ. They rapidly disappear from a portion of the mucous mem- brane kept at a temperature of about 40° C. outside of the body, and their disappearance is due, not to their regeneration into serum proteins, as was once supposed, but to their decomposition by the erepsin. We must suppose that the serum and organ proteins are 29 450 ABSORPTION built up from the products of this decomposition. But whether the mucosa of the alimentary tract is especially a seat of the synthesis is unknown (p. 573). On a priori grounds, it is at least equally probable that it occurs in all the cells of the body, each one building up for itself the particular kind of protein which it needs. The direct way of testing the question would be to examine the blood coming from the intestine during the absorption, of proteins, and to determine quantitatively any changes which might have occurred in the nitrogenous constituents. But the flow of blood through the intestine is so great, the absorption of the digestive products so gradual, and their removal from the blood by the tissues, in all probability, so rapid, that there is no reason for surprise that till lately the results of such determinations were ambiguous. Leathes, however, showed some time ago that when peptone, proteose, or the final products of tryptic digestion are introduced into a ligated segment of a dog's small intestine, there is always, when absorption occurs, an increase in the nitrogenous substances in the blood, in the form of compounds which are not precipitated by tannic acid, and therefore are neither native proteins nor proteose. Urea accounts for about one-half of the increase ; the rest he considered to represent probably amino-acids and similar substances. Quite recently it has been conclusively demonstrated by improved methods that the digestion of protein is associated with an increase of non-protein nitrogen in the blood, due, there is every reason to believe, to amino- bodies derived from the hydrolysed protein (Folin and Denis, Van Slyke, Abel). This proves for the mammal what had been deduced by Cohnheim for a much lower form from experiments made on the intestines of certain octopods, which, when excised and suspended in the oxygenated blood, will live for many hours. A solution of peptone was introduced into the isolated intestine, and after twenty hours the crystalline products, leucin, tyrosin, lysin, and arginin, were found in the blood. In the intact animal none of these bodies could be detected in the blood (Cohnheim). The inference was that protein in these animals is absorbed in the form of amino-acids, etc., which are then carried to the tissues and utilized there. In the mammal the same thing appears to be true. For the increase in the amino-acids during digestion of proteins occurs not only in the portal blood, but in the blood of the general circulation. So that, although a part of the absorbed amino-bodies may be removed by the liver, a portion at least is available for the tissues in general. That the tissues actually take up such decomposition products of proteins is indicated by the fact that during and after the digestion of protein in a loop of intestine the non-protein nitrogen of the tissues is increased (Folin). It may be that some of the proteose and peptone are regenerated by a shorter process, and without having been further split up, but of this, too, -there is no definite ABSORPTION OF THE VARIOUS FOOD SUBSTANCES 451 proof. The regeneration, wherever it occurs, must presumably take place in cells, and the only available cells in the digestive mucous membrane are those which line the tube, or the leucocytes which wander between them. Accordingly, both have been credited with the power of absorbing (and perhaps transforming) these substances, but the balance of evidence is in favour of the epithelial cells. We cannot, however, as in the case of the fat, single out any particular tract of epithelium as alone engaged in the absorption (and possibly in the resynthesis) of the products of the digestion of the proteins. In all likelihood the cells covering the villi are actively concerned, but there is no valid reason for denying a share to the general lining of the stomach and small intestine, even perhaps including the Lieberkiihn's crypts or intestinal glands, which morphologically form a kind of inverted villi. It is, indeed, true that the crypts do not take part in the absorption of fat, for no granules blackened by osmic acid occur in them during digestion of a fatty meal. But this is a ground for attributing to them other absorptive functions rather than for altogether denying to them a share in absorption, unless, indeed, we assume that the secretion of the succus entericus engrosses the whole activity of this extensive sheet of cells. Even within physiological limits distension of the gut causes the crypts to become shorter and broader, by a process of partial unfolding which permits a greater part of their epithelium to come into con- tact with the intestinal contents. In extreme distension they may be completely smoothed out. The extraordinary efficiency of the small intestine in digestion and absorption is shown by the fact that, after removal of even 70 to 83 per cent, of the combined jejunum and ileum in dogs, the metabolism is not necessarily much affected. On a diet poor in fat the animals absorb as much of the fat as a normal dog, although a smaller proportion when the diet is rich in fat. It has been generally stated that it is never permissible to remove more than one-third of the small intestine in man. But in one case 2f metres was resected, or quite one-half, and the patient recovered. Even the large intestine, which possesses Lieberkiihn's crypts, but no villi, is able to absorb not only peptones and sugar, especially mono- saccharides like dextrose, but also fats and native proteins. And although these are powers which can be rarely exercised to any great extent in normal digestion, they form the physiological basis of the important method of treatment by nutrient enemata. The observa- tion already mentioned (p. 331), that considerable quantities of food administered by the rectum can pass through the ileo-colic sphincter and valve into the lower part of the ileum, thanks to the antiperistaltic movements of the large intestine, indicates that an important part of the preliminary digestion and of the absorption of enemata may occur in the small intestine. But remnants of the 452 ABSORPTION proteolytic, amylolytic, fat-splitting, and inverting ferments which have done their work in the small intestine are passed on into the large, and may be demonstrated in its contents. Doubtless these are able to act upon food substances which may have escaped com- plete digestion and absorption in the higher parts of the alimentary canal, as well as upon food substances injected into the rectum. Summary. — With the proviso that in the case of the fats the statement may in the present condition of our knowledge be some- what ' diagrammatic,' we may sum up in a few words the chief points in the absorption of the food materials. All the fats must be split in order to be absorbed in soluble form from the intestine, but need not be split in the lumen of the gut in order to be utilized by the cells. For this reason they are to a great extent resynthesized to neutral fat after absorption, and find their way into the blood, mainly by way of the lymph, in particulate form. Proteins could perhaps to a small extent be absorbed as such, but must be thoroughly hydrolysed in order to be utilized by the tissues, and also in order to be freely taken up from the gut. Carbo-hydrates in certain forms (mono- saccharides) are capable without change of being both freely absorbed from the intestine and thoroughly utilized by the cells ; only the more complex carbo-hydrates need to be hydrolysed in order to be absorbed, but all above the monosaccharides must be hydrolysed to monosac- charides in order to be utilized. The substances which eventually circulate in the blood in solution reach it through the gastro-intestinal capillaries ; the substances which eventually circulate in the blood in particulate form reach it through the lymphatics. PRACTICAL EXERCISES ON CHAPTERS VI. AND VII. i. Contraction of Isolated Intestines in Ringer's Solution. — Arrange a good-sized water-bath (a water-tight garbage-can holding 20 litres will do) so that the temperature of the water is kept at 37° to 38° C. For this a gas regulator is most convenient, and also a stirring arrange- ment worked by a small motor. But if neither of these is available, a student, by a little care, can easily keep the water at the required temperature by raising and lowering the gas flame and stirring occasion- ally by hand. In the bath, support (a) a stock bottle of Ringer's solution (footnote, p. 66), (b) a wide-mouthed bottle containing Ringer for the reception of the stock of intestine, (c) a small cylinder for segments of intestine whose contractions are to be recorded, (c) is conveniently made in different sizes by cutting down glass T-pieces. One which holds 4 or 5 c.c. is convenient. The bottom is plugged with a rubber cork in which is fastened a hook. In the side-piece is fixed by a rubber cork a glass tube ending in the cylinder in a narrow orifice. This is connected with the oxygen-supply, conveniently ob- tained under constant pressure from a small gas-container which is periodically replenished from an oxygen cylinder. A separate oxygen cylinder is connected with a tube passing to the bottom of (b) . A lever with two arms is arranged on the same stand as (c), so that it can be thrown on and off a slow-moving drum by a single movement of the stand . All being ready, a rabbit is killed by being struck on the back of the PRACTICAL EXERCISES 453 head. The small intestine is immediately removed. It may be cut between double ligatures into several pieces for this purpose. The contents are rapidly washed out by a stream of warm Ringer, and the pieces placed in (b) , through which oxygen is kept bubbling. The pieces are conveniently supported in the liquid by threads fixed by the cork of the bottle. There is a hole in the cork for the escape of the oxygen. The movements of the intestines in (&) can be studied very well by inspection. Or a separate length of intestine maybe kept for this purpose, the contents not being removed, but prevented from escaping by ligatures at each end. This can be most easily observed in a shallow dish of warm Ringer. Or a separate experiment can be made in which the whole alimentary canal of a rabbit is carefully removed and examined in oxygenated Ringer's solution. A segment of intestine about 2 or 2^- cm. in length is now cut off one of the pieces. A small ring of platinum or aluminium is tied to a point on the circumference of one end of the preparation by a silk thread passed through the wall. The other end is caught by a serve- fine at a point exactly corresponding to the attachment of the ring, so that the pull of the contracting longitudinal muscle should be in the straight line joining these two points. The serve-fine has attached to it a thread with a hook on the other end. In preparing the intestinal segment it lies on a plate of glass above a vessel of warm water. The small cylinder (c) is now partially filled with warm Ringer's solution. The ring is grasped by fine forceps, and made to engage with the hook at the bottom of the cylinder, care being taken not to injure the prepara- tion in the process. The, cylinder is then fastened on its stand and lowered into the bath. The thread is connected by its attached hook to the lever, and oxygen allowed to bubble slowly and regularly through the cylinder. Very soon rhythmical contractions begin (Fig. 173), and continue for a long time. The effect on these contractions of abolish- ing, reducing, or increasing the oxygen-supply may first be studied. 2. Effect of Blood-Serum on, the Contractions of Intestinal Segments. — While a tracing is being taken as in i, fill a small bent pipette with serum already warmed in the bath, pass the point of the pipette down to the bottom of the cylinder without interfering with, the preparation, and allow the serum to flow in till the Ringer's solution is displaced. Almost at once the lever will begin to rise, indicating strong tonic contraction. The increase of tone lasts for some time, but can soon be removed on washingthe preparation with Ringer. This is most easily done, while the drum is stopped, by introducing pipetteful after pipetteful of Ringer's solution into the cylinder in the way described, allowing the liquid to overflow into the bath. The subsequent addition of serum causes a renewed increase of tone, and this may be many times repeated. Determine the greatest dilution of the serum which still produces a distinct effect upon the intestinal segment. 3. Action of Epinephrin (Adrenalin) on Intestinal Segments. — Pro- ceed as in 2, but use various dilutions of adrenalin chloride instead of serum. They must be freshly prepared. Instead of increase of tone, inhibition of the movements and decrease of tone will be obtained (Fig. 173). This experiment may be performed at another stage in the course (p. 720). 4. Quantitative Estimation of Ferment Action. — For pepsin an easy method, although not a very accurate one, of estimating the rate at which the fibrin disappears is to use fibrin stained with carmine. As solution goes on, the dye colours the liquid more and more deeply, and by comparing the depth of colour at any time with standard solutions of carmine, we get the quantity of dye set free, and therefore of fibrin 454 DIGESTION AND ABSORPTION digested. This method cannot be used for trypsin. A much better method is that of Mett (p. 343). Fluid egg-white is sucked up into fine glass tubes (of i to 2 mm. bore). The tubes are then heated in a bath at 95° C. For use short pieces (i or 2 cm. long) are cut off and placed in i or 2 c.c. of the liquid to be tested, the whole being kept at 38° to 40° C. For amylolytic ferments where rapid work is required, glass tubes filled with tinted starch paste may be used in the same way as the Mett's tubes. A more accurate method, and yet a rapid and convenient one, is based upon the time which is necessary in order that the iodine reaction with starch may just disappear when a standard starch solution is digested with a dilution of the ferment solution at 40° C. 5. Saliva — Collection and Microscopic Examina- tion of Saliva. — Chew a piece of paraffin-wax, or inhale ether or the vapour of strong acetic acid . The flow of saliva is in creased. Collect it in a porcelain capsule. Examine a drop under the microscope. It may contain a few flat epithelial scales from the mouth and a few round granular bodies, the sali- vary corpuscles, the gran- ules in which often show a lively, dancing move- ment (Brownian motion) . Filter the saliva to free it from air-bubbles, and per- form the following ex- periments : (a) Test the reaction with litmus paper. It is usually alkaline. An acid reaction may indicate that bacterial processes are abnormally active in Fig. 173. — Effect of Serum and Adrenalin on Con- tractions of a Segment of Intestine. Rabbit's intestine contracting in Ringer's solution. At 55 the Ringer's solution was replaced by dog's serum, and this at 56 by adrenalin (1:5,000,000) in serum. At 57 this weak adrenalin solution was replaced by a stronger one (i: 500,000) in serum. Time-trace, half-minutes. (Reduced to half.) the mouth. (6) Add dilute acetic acid. A precipitate indicates the presence of mucin (p. 344). Filter. (c) Add a drop or two of silver nitrate solution to the filtrate from (b). A precipitate insoluble in nitric acid, soluble in ammonia, proves that chlorides arc present. (d) Add to another portion a few drops of dilute ferric chloride slightly acidulated with dilute hydrochloric acid, and the same quantity to as much distilled water in a control test-tube. A reddish coloration is obtained, due to the presence of sulphocyanic acid, which is com- bined with potassium and other bases in the saliva. The colour is dis- charged by mercuric chloride. Or, a drop of saliva may be allowed to fall from the mouth on a test-paper (prepared by soaking filter-paper with a dilute starch solution, containing a little iodic acid, and allowing it to dry in the air). The paper is coloured blue by the union of the PRACTICAL EXERCISES 455 starch with iodine set free from the iodic acid by the action of the sulpho- cyanic acid. (e) Take some boiled starch mucilage, and test it for reducing sugar by Trommer's test (p. 10). If no sugar is found, take three test- tubes, label them A, B, C, and nearly half fill each with the boiled starch. To A add a little saliva,* to B some saliva which has been boiled, to C a little saliva which has been neutralized, and as much 0-4 per cent, hydrochloric acid as has been taken of the mucilage, so as to make the strength of the acid in the mixture 0-2 per cent., a propor- tion well below that of the gastric juice . Put the test-tubes into a water- bath at 40° C. In a few minutes test the contents for reducing sugar. Abundance will be found in A, none in B or C. In B the ferment ptyalin has been destroyed by boiling; in C its action has been inhibited by the acid. If the test-tubes have been left long enough in the bath, no blue colour will be given by A on the addition of iodine, but a strong blue colour by B and C—i.e., the starch will have completely disappeared from A. (/) Put some starch in a test-tube, add a little saliva, and hold in the hand or place in a bath at 40° C. On a porcelain slab place several separate drops of dilute iodine solution. With a glass rod add a drop of the mixture in the test-tube to one of the drops of iodine at intervals as digestion goes on. At first only the blue colour given by starch will be seen; a little later a violet colour, due to the presence of erythro- dextrin in addition to some unaltered starch. A little later the colour will be reddish, the starch having disappeared and the erythrodextrin reaction being no longer obscured. Later still no colour reaction will be obtained, the erythrodextrin having undergone further changes, and only sugar (maltose, isomaltose, and perhaps a trace of dextrose) and achroodextrin — a kind of dextrin which gives no colour with iodine — being present. (g) Put two pieces of glass tube filled with tinted starch paste (p. 454) into separate test-tubes. Cover one with 3 c.c. and the other with 6 c.c. of saliva. The saliva must all be taken from the same stock, and must not be collected separately. Put in a bath at 38° C., and when a fair amount of digestion has taken place in each, measure the length of the column digested, and determine the relation between the amount digested in the two tubes (p. 342). (h) Dilute 2 c.c. of saliva with distilled water up to 20 c.c., and filter. Take six test-tubes of the same width, and label them A, B, C, etc. Measure into A 3 c.c. of the diluted saliva, into B 2 c.c., into C 1-3 c.c., intoDo'9C.c.,into Eo'6c.c., and into F 0-4 c.c. Thus a series is obtained in which each tube contains (approximately) two-thirds as much ferment as the one it follows. Add distilled water to tubes B to F, sufficient to make up the volume in each to 3 c.c. Place the tubes in a beaker of iced water; add to each 10 c.c. of a I per cent, solution of boiled starch previously cooled in iced water, and shake so as to mix the contents. Each tube now contains starch in uniform concentration, and ferment in varying concentration. The low temperature prevents digestion till all the tubes are ready. Now put the tubes simultaneously into a water- bath at 40° C. for half an hour, and then back again into iced water to prevent further digestion. Move them about in the iced water to cool rapidly. Fill up the tubes with distilled water nearly to the top. add a drop or two of iodine solution to each, and mix uniformly. The tubes to which the smallest amounts of saliva were added will probably * As it filters slowly, unriltered saliva may be used for Experiments (e), -456 DIGESTION AND ABSORPTION still show a distinct blue colour, while those at the other end of the series will be brown or yellow, and the intermediate tubes bluish- violet. Suppose D is the last tube still showing a bluish tint, then in the next higher tube, C, all the starch has been hydrolysed at least to dextrin — that is, 1-3 c.c. of the ten-times diluted saliva, or 0-13 of the original saliva, has been sufficient to change all the starch in 10 c.c. of the i per cent, solution. With another specimen of saliva the same result might be reached in tube E, containing an amount of ferment equal to that in 0-06 c.c. of the original saliva". We could then conclude that the diastatic power of the second saliva was about twice as great as that of the first. A closer approximation can now be made by setting up two fresh tubes (C and E respectively for the two salivas) and deter- mining the time required for the blue reaction with iodine to disappear, taking out a drop from time to time and testing on a porcelain slab. («') Put a little distilled water into a porcelain capsule, and bring the water to the boil. Now put into the mouth some boiled starch paste, and move it about as in mastication. After half a minute spit the starch out into the boiling water. Divide the water into two portions. Test one for sugar, and the other for starch. Repeat the experiment, but keep the starch two minutes in the mouth. Report the result. (;') Starch solution to which saliva has been added is placed in a dialyser tube of parchment-paper for twenty-four hours. At the end of that time the dialysate (the surrounding water) should be tested for sugar and for starch. Sugar will probably be found, but no starch. If no reaction for sugar is obtained, the dialysate should be concen- trated on the water-bath, and again tested. 6. Stimulation of the Chorda Tympani. — (i) Having previously studied the anatomy of the mouth and submaxillary region in the dog by dissecting a dead animal (Fig. 174), put a good -sized dog under morphine. Set up an induction-machine- for a tetanizing current (p- 200), and connect it with fine electrodes. Fasten the dog on the holder, give ether if necessary, and insert a cannula in the trachea (p. 202). Then make an incision 3 or 4 inches long through skin and platysma muscle, along the inner border of the lower jaw, beginning about the angle of the mouth, and continuing backwards towards the angle of the jaw. Such branches of the jugular vein as come in the way may be generally pushed aside, but if necessary they may be doubly li gated and divided. Feel for the facial artery, so as to be able to avoid it. Divide the digastric muscle about its anterior third, and clear it carefully from its attachments; or, without dividing it, pull it outwards with a hook. The broad, thin mylo-hyoid muscle will now be seen with its motor nerve lying on it. Divide the muscle about its middle at right angles to its fibres, and raise it carefully. The lingual nerve will be seen emerging from under the ramus of the jaw. It runs transversely towards the middle line, and then, bending on itself, passes forwards parallel to the larger hypoglossal nerve. In its transverse course the lingual will be seen to cross the ducts of the submaxillary and sublingual glands. These structures having been identified, the lingual nerve is to be ligatured before it enters the tongue and cut peripherally to the ligature. Then a glass cannula of suitable size is to be inserted into the submaxillary duct (the larger of the two), just as if it were a bloodvessel (p. 63). A short piece of narrow rubber tubing is carefully slipped on the end of the cannula. The lingual is now to be lifted by means of the ligature, and traced back towards the jaw till its chorda tympani branch is seen coming off and running backwards along the duct. The chordo-lingual nerve (Fig. 160, p. 392) is then to be cut centrally to the origin of the chorda tympani, which can now be PRACTICAL EXERCISES 457 easily laid on electrodes by means of the ligature on the lingual. On stimulating the chorda, the flow of saliva through the cannula will be increased. The current need not be very strong. If the flow stops after a short time, it can be again caused by renewed stimulation after a brief rest. A quantity of saliva may thus be collected, and the experi- ments already made with human saliva repeated. (2) Expose the vago-sympathetic nerva in the neck on the same side ; ligature it ; divide below the ligature ; and note the effect pro- duced by stimulation of the upper end on the flow of saliva. (3) Set up another induction-machine, and connect it with electrodes. Stimulate the chorda, and note the rate of flow of the saliva. Then, while the chorda is still being excited, stimulate the vago-sympathetic, and observe the effect. If the experiment is successful, finish by stimulating the chorda for a long time. Then harden both sub- Digastric Muscle (cut). Hypoglossal Nerve. Mylo-hyoid Lingual Wharton's Muscle (cut). Nerve. Duct. Fig. 174. — Dissection for Stimulation of Chorda Tympani (after Bernard). maxillary glands in absolute alcohol, make sections, stain with carmine, and compare them. 7. Effect of Drugs on the Secretion of Saliva. — (i) Proceed as in 6 (i), but, in addition, insert a cannula into the femoral vein (p. 218). On the cannula put a short piece of rubber tubing, filled with 0-9 per cent, salt solution and closed by a small clamp, or a small piece of glass rod, or a pair of bulldog forceps. While the chorda is being stimulated inject into the vein 10 to 15 milligrammes of sulphate of atropine by pushing the needle of a hypodermic syringe through 'the rubber tube. This will stop the flow of saliva, and abolish the effect of stimulation of the chorda. See whether the sympathetic is also inactive, and report the result. (2) Now empty the cannula in the submaxillary duct by means of a feather, and fill it with a 2 per cent, solution of pilocarpine nitrate by means of a fine pipette. Fill also the short rubber tube attached to the cannula, and close it again. Compress the tube, and so force into 458 DIGESTION AND ABSORPTION the duct a small quantity of the solution. Open the tube. Secretion of saliva will again begin, and stimulation of the chorda will again cause an increase in the flow. But after a few minutes the action of the atropine will reassert itself, and the flow will stop. Renewed secretion may be caused by a fresh injection of pilocarpine. 8. Gastric Juice — (a) Preparation of Artificial Gastric Juice. — Take a portion of the pig's stomach provided, strip off the mucous membrane (except that of the pyloric end, which is relatively poor in pepsin), cut it into small pieces with scissors, and put it in a bottle with 100 times its weight of 0-4 per cent, hydrochloric acid. Label and put in a bath at 40° C. for three hours, and then in the cold for twelve hours. Then filter. (b) Take another portion of the mucous membrane, cut it into pieces, and rub up with clean sand in a mortar. Then put it in a small bottle, cover it with glycerin, label, and set aside for two or three days. The glycerin extracts the pepsin. (c) Take five test-tubes, A, B, C, D, E, and in each put a little washed and boiled fibrin or a small cube of coagulated egg-white. To A add a few drops of glycerin extract of pig's stomach, and fill up the test-tube with 0-4 per cent, hydrochloric acid. To B add glycerin extract and distilled water; to C glycerin extract and i per cent, sodium carbonate; to D 0-4 per cent, hydrochloric acid alone; to E glycerin extract which has been boiled, and 0-4 per cent, hydrochloric acid. Put up another set of five test-tubes in the same way, except that a few drops of a watery solution of a commercial pepsin are substituted for the glycerin extract. Label the test-tubes A', B7, C', D', E'. Into another test-tube put a little fibrin (or an egg-white cube), and fill up with the filtered acid extract from (a). Label it F. Place all the test-tubes in a tumbler, and set them in a water-bath at 40° C. Put a piece of a Mett's tube (p. 343) into each of two test-tubes, and add 15 c.c. of 0-4 per cent, hydrochloric acid. To one tube add 5 drops and to the other 10 drops of the same filtered glycerin extract of gastric mucous membrane. Put the tubes in the bath, and when digestion is distinct at the ends of both tubes measure the length of the column digested in each. What is the relation between the two (p. 342) ? The experiment can be repeated with the hydrochloric acid extract of the mucous membrane. After a time the fibrin (or egg-white) will have almost completely dis- appeared in A, A', and F, but not in the other test-tubes. Filter the contents of A, A', and F into one dish. (d) Test the filtrate for the products of gastric digestion : (a) Neutralize a portion carefully with dilute sodium hydrox- ide. A precipitate of acid -albumin may be thrown down. Filter. (jS) To a portion of the filtrate from (a) add excess of sodium hydroxide and a drop or two of very dilute copper sulphate. A rose colour indicates the presence of proteoses or peptones. The cupric sulphate must be very cautiously added, because an excess gives a violet colour, and thus obscures the rose reaction. If still mere cupric sulphate be added, blue cupric hydroxide is thrown down, and nothing can be inferred as to the presence or the nature of proteins in the liquid. (y) Heat another portion of the filtrate from (a) to 30° C., and add crystals of ammonium sulphate to saturation. A precipitate of proteoses (albumoses) may be ob- tained. Filter off. PRACTICAL EXERCISES 459 (6) Add to the filtrate from (y) a trace of cupric sulphate and excess of sodium hydroxide. A rose colour indicates that peptones are present. More sodium hydroxide must be added than is sufficient to break up all the ammonium sulphate, for the biuret reaction requires the presence of free fixed alkali. A strong solution of the sodium hydroxide should therefore be used, or the stick caustic soda. The addition of ammonium sul- phate will cause the red colour to disappear; so will the addition of an acid. Sodium hydroxide will bring it back. Ammonia does not affect the colour. (e) To some milk in a test-tube add a drop or two of rennet extract, and place in a bath at 40° C. In a short time the milk is curdled by the rennin. (See p. 353 ) 9. (i) To obtain Normal Chyme.— Inject subcutaneously into a dog, one and a half hours after a meal of minced meat and water, 2 mg. of apomorphine. Half of one of the ordinary tabloids is enough. Collect the vomit. (2) To obtain Pure Gastric Juice. — If the laboratory possesses a dog with Pawlow's double cesophageal and gastric fistula, the juice may be obtained in large amount by sham feeding with meat (p. 4°2)- I* not, proceed as follows: Put a fasting dog under ether, and fasten on the holder. Clip the hair and shave the skin in the middle line below the sternum. Make a longitudinal incision through the skin and subcutaneous tissue from the xiphoid cartilage downwards for 3 or 4 inches. The linea alba will now be seen as a white mesial streak. Open the abdomen by an incision through it. Pull over the stomach towards the right, and stitch it to the abdominal wall, open it, and insert a stomach cannula (Fig. 175). Make an incision through the serosa and muscularis. Doubly ligate and divide any vessels exposed in the submucosa. Then make an opening in the mucosa of sufficient size to just admit the gastric cannula. This will go into a smaller opening if it is provided with a nick in the flange which enters the stomach. Be careful to prevent blood from getting into the stomach. Immediately stitch the wound in the stomach over the flange of the cannula, but do not pass the stitches through to the internal surface of the mucosa. Suture the muscles and skin separately. Then stitch up the wound in the abdomen. Wash out any stomach contents with warm water. Put a cork in the cannula, and cover the animal with a cloth. The follow- ing experiments may now be performed: Expose both vagi in the neck. Connect two pairs of electrodes with the secondary coil of an induc- torium arranged for single shocks. By means of a key in the primary stimulate the nerves with slow rhythmical induction shocks at the rate of about one a second. Continue the stimulation for fifteen minutes, collect any juice that may have been secreted, and apply the tests in (3). If secretion is slow, a little distilled water may be put into the stomach, and the vagus stimulation repeated. Mechanical stimulation of the mucous membrane with a feather causes no secretion of acid gastric juice, but may cause a secretion of alkaline mucus. (3) (a) Test the reaction to litmus of the chyme obtained in (i), and of the pure juice obtained in (2). (b) Test their proteolytic powers by putting in a bath at 40° C. for two hours two test-tubes containing respectively filtered chyme and Fig. 175. — Stomach Cannula. 46o DIGESTION AND ABSORPTION fibrin, and gastric juice and fibrin. The fibrin will be digested in both. Estimate the proteolytic power quantitatively by Mett's tubes (p. 454). (c) Add a few drops of the chyme and gastric juice to milk in two test -tubes, and place them in a bath at 40° C. Repeat (c) after neutral- izing the liquids. (d) Examine a drop of the unfiltered chyme under the microscope. Partially digested fragments of the food will be seen — muscular fibres or fat cells. Filter, and proceed as in 8 (d) (p. 458). (4) Test the filtrate from the chyme and the gastric juice for lactic acid by Uffelmann's test or Hopkins 's test (p. 821), and for hydrochloric acid by Gunzburg's reagent. Uffelmann's Test for Lactic Acid. — The reagent is a dilute solution of carbolic acid to which dilute ferric chloride has been added till the colour is bluish (say a drop of a I per cent, ferric chloride solution to 5 c.c. of a i per cent, carbolic acid solution). The blue colour of the mixture is turned yellow by lactic acid, but not by dilute hydrochloric acid. Since Uffelmann's test is given also by phosphates, alcohol, and sugar, which may sometimes be present in the stomach contents, it is best to shake the gastric contents with ether, dissolve the ethereal extract in water, and then make the test on the watery solution. Gunzburg's Reagent for Free Hydrochloric Acid in Gastric Juice is made by dissolving 2 parts of phloroglucinol and I part of vanillin in 30 parts by weight of absolute alcohol. A few drops of the reagent are added to a few drops of the filtered gastric juice in a small porcelain capsule, and the whole evaporated to dryness over a small bunsen flame. If free hydrochloric acid is present, a carmine-red residue is left. If all the hydrochloric acid is united to proteins in the stomach contents, the reaction does not succeed. It is also hindered by the presence of leucin. 10. Pancreatic Juice. — (a) Take a piece of the pancreas of an ox or dog which has been kept twenty-four hours at the temperature of the laboratory, and make a glycerin extract in the same way as in the case of the pig's stomach in 8 (6). Put in a small bottle, and set aside for a day or two. (6) Put a little boiled fibrin into each of six test-tubes, A, B, C, D, E. F. To A add a few drops of glycerin extract of pancreas, and fill up with a i per cent, sodium carbonate solution ; to B add glycerin extract and distilled water; to C glycerin extract and excess of 0^05 per cent, hydrochloric acid ; to D i per cent, sodium carbonate alone ; to E i per cent, sodium carbonate in which a few drops of glycerin extract of pancreas have been previously boiled ; to F glycerin extract and excess of 0-2 per cent, hydrochloric acid.* Set up six test-tubes, A', B', C', D', E', F', in the same way, but substitute a few drops of a solution of commercial pancreatin for the glycerin extract. Set up two test-tubas as in expsriment 8 (p. 458) with Mett's tubes. Put all the test-tubes in a tumbler, and place in a bath at 40° C. The fibrin will be gradually eaten away in A and A1, by the action of the trypsin, but will not swell up or become clear before disappearing, as it does in dilute hydrochloric acid with glycerin * With hydrochloric acid of different strengths the rapidity of digestion of boiled fibrin by glycerin extract of dog's pancreas (i volume of extract to 25 of acid) was found about the same for 0-3 and o'iy per cent, acid; much less for o-o8 per cent., while in 0^04 per cent, acid there was practically no digestion at all. In 0-4 per cent, acid digestion took place more rapidly than in o-o8 per cent., but much less rapidly than in 0-17 per cent. In acid of all strengths digestion was markedly slower than in i per cent, sodium car- bonate. PRACTICAL EXERCISES 46i extract of stomach. Filter the contents of these test-tubes. Neutralize the filtrate with dilute acid ; a precipitate will consist of alkali-albumin. If such a precipitate is obtained, filter it off and test the filtrate for proteoses and peptones as in 8 (d) (p. 458). Some digestion, and perhaps a considerable amount, may also have taken place in F and F'; less or none at all in C and C' ; and none in the other iest-tubes (pp. 359, 420). (c) Add a few drops of the glycerin extract to a test-tube containing starch mucilage, which has been previously found free from reducing sugar. Put in a bath at 40° C. After a short time abundance of reducing sugar will be found, owing to the action of the ferment, amylopsin, or pancreatic amylase. (rf) Mince thoroughly a good-sized piece of fresh pancreas, and shake up well with three or four times its bulk of water. Put 5 c.c. of fresh cream into a test-tube, then 10 c.c. of the extract, a few drops of chloro- form to prevent the growth of bacteria, a few drops of litmus solution, and if necessary enough of very dilute sodium hydroxide to just render the colour distinctly blue. Shake up, and divide the mixture into two portions, A and B. Boil one portion (B), and place the test-tubes at 40° C. Examine from time to time. The blue colour will disappear in A, owing to the formation of fatty acids from the neutral fats, and sodium hydroxide must be added to it to restore the colour. In B the fat-splitting ferment has been destroyed by boiling, and fat-splitting will not occur. Probably a distinct result will not be obtained for several hours, and it will be best to leave the tubes in the incubator overnight. (e) If the laboratory possesses an animal with a pancreatic fistula, the following experiment may be done by a limited number of students with fresh pancreatic juice* collected from the fistula. Take five test- tubes, A, B, C, D, E. Add 5 c.c. of pancreatic juice to each tube. Boil E, and then cool it. Put into A and B small pieces of heat-coagulated egg-white, into C a little starch mucilage, and into D and E 5 c.c. of fresh cream. Add further to B a scraping of the mucous membrane of the upper part of the small intestine which has first been washed free of contents. To D and E add a drop or two of litmus solution, and, if necessary, enough of dilute sodium hydroxide to just establish a blue colour. Then put the test-tubes at 40° C., and examine after a time. No digestion will have taken place in A, because the pancreatic juice, as secreted, does not contain active trypsin. In B digestion may take place, because the enterokinase in the intestinal mucous membrane will activate the trypsinogen to trypsin. In C and D there will be evidence of the production of reducing sugar and fatty acids respec- tively, since the pancreatic juice, as secreted, contains active amylase and steapsin. E will be unchanged unless by bacterial action. (/) Leucin and Tyrosin. — As examples of amino-acids formed when pancreatic digestion of proteins (fibrin or casein, e.g.) is allowed to go on for some days.fleucin and tryosin maybe isolated. Add bromine- water by drops to 5 c.c. of the digest; a pink colour indicates trypto- phane. If the ' digest ' be neutralized, then filtered, and the filtrate concentrated and allowed to stand, a crop of tyrosin crystals will separate out, since tyrosin is only slightly soluble in watery solutions of neutral salts. These crystals having been filtered off, the proteoses (albumoses) and peptones can be precipitated together by alcohol, and * A considerable flow of pancreatic juice can be obtained from a dog with a pancreatic fistula by injecting intravenously an extract of intestinal mucous membrane containing secretin (p. 407). • | A little chloroform is added to prevent bacterial growth. 4&a DIGESTION AND ABSORPTION afterwards separated, if that is desired, by redissolving the precipitate in water and throwing down the proteoses by saturation with am- monium sulphate. The alcoholic nitrate will contain any leucin that may be present, since that body is moderately soluble in alcohol, as well as traces of tyrosin, which, however, is much less soluble in this medium. On concentration, crystals of both substances will be ob- tained. Tyrosin crystallizes characteristically from animal liquids in beautiful silky needles united into sheaves, leucin in the form of in- distinct fatty-looking balls, often marked with radial striae and coloured with pigment (Figs. 186 and 187, p. 488). Tests for Tyrosin by M timer's Test. — Put a small quantity of tyrosin into a test-tube. Add about 3 c.c. of the reagent,* and heat gradually and gently to the boiling-point. A green colour is obtained. II. Bile. — (a) Test the reaction of ox bile. It is alkaline to litmus. (b) Add dilute acetic acid. A precipitate of bile-mucin (really nucleo-albumin) falls down. Some of the bile-pigment is also pre- cipitated. Filter. (Pig's bile contains more of the mucin-like sub- stance than ox bile.) (c) Put a little of the filtrate from (&) or of the original bile into a porcelain capsule, add a drop or two of a dilute solution of cane-sugar, and mix with the bile. Then add a few drops of strong sulphuric acid, and stir; then a few drops more of the sulphuric acid, stirring all the time. A purple colour appearing at once, or after gentle heating, shows the presence of bile-acids (Pettenkofer's reaction). The bile may be diluted before the addition of the sulphuric acid. In this case a greater amount of the acid must be added. Examine the purple liquid in a test-tube with a spectroscope (p. 74). Dilute the liquid with water, adding some sulphuric acid to partially clear up the precipitate caused by the water. Two absorption bands are seen, one to the red side of D, and the other, a stronger and broader band, over and to the right of E. When only a very small amount of bile-salts is present, the reaction is made more sensitive if a solution of f urf uraldehyde (i to i.ooo) is used instead of cane-sugar. (d) Hay's Sulphur Test. — Sprinkle a little sulphur (in the form of the fine powder known as flowers of sulphur) on the surface of some bile in a small beaker or deep watch-glass. The sulphur will soon sink to the bottom. Repeat with water; the sulphur will float. The reaction is due to the diminution of the surface tension produced by the bile -acids, and succeeds also in a solution of bile -salts. The test is very sensitive. But in stomach contents, vomit, or stools, it rarely gives good results, since alcohol or acetic acid is often present in the gastric liquid, and phenol and its derivatives in intestinal contents, and all of these cause such an alteration in the surface tension that the sulphur sinks. Ether, chloroform, turpentine, benzine and its deriva- tives, anilin and soaps, also vitiate the test in the same way. (e) Add yellow nitric acid (containing nitrous acid) to a little bile on a white porcelain slab. A play of colours, beginning with green and running through blue to yellow and yellowish-brown, indicates the presence of bile-pigment (Gmelin '^reaction). The reaction may also be obtained by putting some yellow nitric acid into a test-tube, and then running a little bile from a pipette on to the surface of the acid. The play of colours is seen at the surface of contact. Where the bile- pigment is present only in traces, some of the liquid may be filtered * The reagent for this test is prepared by mixing thoroughly i volume of formalin, 45 volumes of distilled water, and 55 volumes of concentrated sulphuric acid. PRACTICAL EXERCISES 463 through white filter-paper, and the test applied by putting a drop of the nitric acid on the paper. (/) Cholesterin or Cholesterol (Fig. 176) — Preparation. — Extract a powdered gall-stone (preferably a white one) with hot alcohol and ether in a test-tube. Heat the test-tube in warm water, not in the free flame. Put a drop of the extract on a slide. Flat crystals of cholesterin, often chipped at the corners, separate out. (a) Carefully allow a drop of strong sulphuric acid and a drop of dilute iodine to fun under the cover- glass. A play of colours — violet, blue, green, red — is seen. ($) Evaporate a drop of the solution of cholesterin. in a small porce- lain capsule, add a drop of strong nitric acid, and heat gently over a flame. A yellow stain is left, which becomes red when a drop of am- monia is poured on it while it is still warm. (7) Dissolve a little cholesterin in chloroform. . Add an equal bulk of strong sulphuric acid, and shake gently. The solution turns red and the subjacent acid shows a green fluorescence. (S) Put a drop or two of water in a watch-glass, and add a drop of an ethereal solution of cholesterin. The cholesterin is precipitated, and will not dissolve in the water even on heating. Repeat the observation with bile instead of water. The cholesterin dissolves in the bile. (g) To a little of the filtrate from a peptic digest (e.g., fibrin which has been digested for twenty-four hours with pepsin and hydrochloric acid) add some bile. A precipitate is thrown down, which is redissolved in excess of the bile (p. 370). (h) Shake up a little bile with some rancid olive -oil in a test-tube. An emul- sion is formed. Repeat the experiment with the same quantities of bile and oil, but use perfectly fresh oil. Compare the stability of the two emulsions, allowing the tubes to stand together for a while. Fig. 176. — Crystals of Cholesterin (i) To some starch mucilage, shown (Frey). to be free from sugar, add a little bile, and place in a bath at 40° C. After a time test for reducing sugar. Report the result. Bile often has a slight diastatic power. (/) To demonstrate the Presence of Iron in the Liver Cells. — Steep sec- tions of liver in a solution of potassium ferrocyanide, and then in dilute hydrochloric acid. Or a 1-5 per cent, solution of potassium ferrocyanide in 0-5 per cent, hydrochloric acid may be used. (The iron may pre- viously be fixed in the tissue by hardening it in a mixture of alcohol and ammonium sulphide.) The sections become bluish from the formation of Prussian blue. A fine-pointed glass rod or a platinum lifter should be used in manipulating them. A steel needle cannot be employed. Mount in glycerin or Farrant's solution, or (after dehy- drating with alcohol and clearing in xylol) in xylol-balsam. Blue granules may be seen under the microscope in some of the hepatic cells. Sections of spleen may also be examined for this reaction. 12. Microscopical Examination of Faeces. — Examine under the micro- scope the slides provided. Draw, and as far as possible determine the nature of, the objects seen (p. 424). 13. Absorption of Fat. — (a) Feed a rat or frog with fatty food; kill the rat in three or four hours, the frog in two or three days. Imme- diately after killing the rat open the abdomen, carefully draw out a loop of intestine, and look through the thin mesentery. The white 464 DIGESTION AND ABSORPTION lacteals will probably be seen ramifying in the mesentery. They appear white on account of the presence of globules of fat in the chyle with which they are filled. Strip off tiny pieces of the mucous mem- brane of the small intestine, and steep them in \ per cent, solution of osmic acid for forty-eight hours. Then tease fragments of the mucous membrane in glycerin and examine under the microscope. To preserve the specimens take off the glycerin with blotting-paper and mount in Farrant's medium, which is a preservative glycerin mixture. Other portions of the mucous membrane may be hardened for a fortnight in a mixture of 2 parts of Miiller's fluid and i part of a i per cent, solution of osmic acid. Sections are then maae with a freezing microtome after embedding in gum. No process must be used by which the fat would be dissolved out (Schafer). (See Fig. 172, p. 441.) (6) Feed a cat with 30 grammes of butter stained a deep red with the dye Sudan III. After five hours anaesthetize the animal with ether, insert a cannula in the carotid artery, and obtain a sample of blood. Defibrinate the blood, and separate the serum by the centrifuge. If digestion and absorption of the fat have proceeded normally, the serum will contain numerous fat droplets, and will be tinged pink by the dye, which can be dissolved out of it by shaking up with ether. On opening the abdomen it will be seen that the mucous membrane of the small intestine, as far down as the fat has reached, is stained pink, and that the lacteals in the mesentery are also pink. Observe whether any of the pigment has passed into the urine. 14.* Time required for Digestion and Absorption of Various Food Substances. — Feed three dogs, A, B, and C, which have previously fasted for twenty-four hours, with a meal containing starch (proved to be free from sugar), lard, and meat. (1) After fifteen minutes inject subcutaneously into A 2 c.c. of a o-i per cent, solution of apomorphine. Note the time which elapses before the animal vomits. Collect the vomit. (a) Examine a little of it under the microscope, and make out fat globules, muscular fibres and starch granules. The latter can be recog- nized by their being coloured blue by a drop or two of iodine solution. (b) Filter the chyme, mixing it, if necessary, with a little water, and test it as in 8 (d) (p. 458) for the products of digestion of proteins. In addition, test for starch, dextrin, and reducing sugar. (2) One and a quarter hours after the meal inject apomorphine into dog B, and proceed as in (i). (3) Two and a half hours after the meal inject apomorphine into dog C, and proceed as in (i). Compare the results from the three specimens of chyme. 15.* Quantity of Cane-Sugar inverted and absorbed in a Given Time. — Take three dogs, A, B, and C, which have fasted for twenty-four hours. The animals should be about the same size. Feed A and B with 100 c.c. of a standard solution of cane-sugar (about a 20 per cent, solu- tion), or as much more as they will take. If the dogs have been kept without water for a day they will more readily take the sugar solution. Or it may be given through a tube passed into the stomach, and in this case a larger quantity of sugar can be given. A gag consisting of a piece of wood with a hole in the middle of it, through which the tube is passed, must first be secured in the dog's mouth. Feed C with * Experiments 14 and 15 are conveniently done in a class by assigning each of the three animals to a separate set of students. The contents of the stomach and intestine are divided into three portions, so that each set has a sample from each dog. PRACTICAL EXERCISES 465 50 grammes of powdered cane-sugar mixed with lard, the mixture being rolled into little balls. (1) After a quarter of an hour put A under chloroform or the A.C.E. mixture, and fasten it on a holder. Kill the animal with chloroform, open the abdomen, tie the oesophagus, place double ligatures on the pyloric end of the stomach and the lower end of the small intestine, and divide between them. Cut out the stomach and intestine ; wash their contents into separate vessels, and test the reaction with litmus paper. Add water and rub up thoroughly. Filter. Wash the residue re- peatedly with small quantities of water, and pass all the washings through the filter. Make up each of the two nitrates to 200 c.c. (a) Test the filtrates from the contents of the stomach and intes- tines qualitatively for dextrose by Trommer's (p. 10) or Fehling's (p. 526) and the phenyl-hydrazine test (p. 525). (b) If no reducing sugar is present, add to 20 c.c. of each filtrate I c.c. of hydrochloric acid, boil for half an hour, and again test for reducing sugar. If it is now found, some cane-sugar is present. (c) If reducing sugar is found, estimate its amount as dextrose by Fehling's solution (p. 527) in a measured quantity of the original filtrate of the gastric or intestinal contents before and after boiling with one-twentieth of its volume of hydrochloric acid. (d) Estimate in the same way the amount (as dextrose) of the invert sugar in the standard solution of cane-sugar after inversion, and before inversion if it gives the qualitative test for reducing sugar before it has been boiled with acid. From the data obtained (and taking 95 parts of cane-sugar as equal to 100 parts of dextrose) calculate the amount of cane-sugar absorbed, left unchanged, and inverted, though not absorbed. (2) One and a half hours after the meal anaesthetize B, and proceed as in (i). (3) Two hours after the meal proceed in the same way with C. But in addition observe the lacteals in the mesentery, by gently lifting up a loop of intestine immediately after opening the abdomen. If the absorption of the fat has begun, they will be easily visible, as a network of fine milk-white vessels. Also examine the gastric and intestinal contents with the microscope for fat globules. Compare your results on the amount of sugar obtained from the three animals. Probably much more unabsorbed sugar will be found in C than in B, as the lard hinders it from being dissolved. 16. Auto-Digestion of the Stomach. — In some of the previous experi- ments the stomach of an animal killed during digestion should be removed from the body after double ligation of oesophagus and duo- denum, and placed in a water-bath at 40° C. After several hours the contents should be washed out and the mucous membrane examined. It may be entirely eaten away in parts. FORMATION OF LYMPH Different Kinds of Lymph. — -We ought to distinguish the lymph as we collect it from the great lymphatic trunks, not only from the liquids of the serous cavities, but still more sharply from the liquid which fills the multitudinous clefts and spaces of the tissues. It is now pretty definitely established that the tissue spaces do not com- municate by actual passages with the lymphatic vessels, but that the latter form everywhere a closed system like the blood- vascular system, the lymph capillaries merely lying in the tissue spaces (Sabin, etc.). This conception entails a radical change in the current views of lymph production. If the lymphatics form a closed system, the lymph cannot be actual tissue fluid, but only tissue fluid modified by its passage through the walls of the lymph capillaries, just as tissue fluid is not actual blood-plasma, but plasma modified by its passage through the walls of the blood capillaries as well as by exchange with the tissue elements. Although it is customary to speak of the lymph obtained from the lymphatic vessels as if it were perfectly homogeneous, there is no experimental ground for supposing that the lymph from different tracts, or the tissue liquid in contact with the cells of different organs, or even the tissue liquid in contact with one and the same cell at different parts of its periphery, has a uniform composition, or even a uniform molecular concentration. There are, indeed, certain general considerations which show that this cannot be so. Still less can it be assumed that the serous cavities, although they come into relation with lymphatics and bloodvessels in their walls, are analogous to colossal tissue spaces or even to expansions of the closed lymphatic system, or that the liquids contained in them, normally in scant amount, are simply tissue, or if not tissue, then simply lymphatic lymph. The cerebrospinal fluid, which bathes the external surface of the central nervous system and fills its cavities, and the special liquids of the eyeball — the aqueous humour and the liquid of the vitreous humour — although no doubt, in addition to their other functions, they may in some degree minister to the nutrition of the tissues with which they are in contact, are 466 FACTORS CONCERNED IN LYMPH FORMATION 467 as regards their composition and mode of formation scarcely more closely allied to lymph than sweat is. They are almost free from protein, and are secreted by special structures — the choroid plexus and the uveal epithelium — quite different from any that can be concerned in the formation of ordinary lymph. It is very true these liquids are not blood, but that is scarcely a sufficient reason for calling them lymph, else we might classify sweat or even milk as lymph also. If a term is desirable to indicate that they have certain relations with lymph, they might perhaps be spoken of as lymphoid secretions. It may be that the essential difference in the chemical composition of these lymphoid secretions and lymph — the practical absence of protein — is related to the difference in the manner of their formation. The uveal and choroidal epithelial cells, interposing the depth of their columns or cubes between the blood and the free surface at which the liquid escapes, may well be suited to hinder the passage of the protein molecules which find their way with greater ease through the thin endothelium of the capillary wall into the tissue spaces, and from these into the lumen of the closed lymphatics (see p. 439)- Nevertheless, we shall recognize later on in the glomeruli of the kidney an instance of blood capillaries but little pervious to proteins, and there are several other facts which show that the capillaries may differ considerably in different organs in the readiness with which they permit the various constituents of the plasma to pass through their walls. Further, in discussing the mechanism by which lymph is formed, we shall see reason to doubt whether mechanical filtration, due to differences of hydrostatic pressure on the two surfaces of the capillary endothelium, has much, if anything, to do with the passage either of protein or of the other constituents of the lymph from the lumen of the capillaries into the tissue spaces. At first glance, indeed, such a process would seem to be admirably fitted to explain the fact that, while lymph differs but little from blood- plasma in the proportions of its other constituents, it is at most no more than half as rich in protein. For there are many filters which allow substances in ordinary solution and their solvent to pass through without alteration in their relative proportions, while substances like proteins in colloid solution are kept back to a greater or less extent. Factors concerned in Lymph Formation. — The teaching of Ludwig, that lymph is formed by the filtration, and in a minor degree by the diffusion, of the constituents of blood- plasma through the walls of the capillaries into the tissue spaces, was based en such facts as the increase in the tissue liquid of a limb or organ which occurs when the exit of blood from it by the veins is hindered, or when the quantity of the circulating liquid is increased by the injection of blood or salt solution. It was first seriously called in question by 468 FORMATION OF LYMPH Heidenhain, who advanced the theory that lymph is secreted by the endothelium of the blood capillaries. One of Heidenhain's strongest arguments in favour of his secretion theory was the existence of substances which, when injected into the blood, in- creased the flow of lymph from the thoracic duct of the dog without affecting appreciably the arterial pressure. He divided these so- called lymphagogues into two classes: (i) Substances like peptone, extracts of the head and liver of the leech, extract of crayfish muscle, egg-albumin, etc., which cause not only an increase in the rate of flow, but an increase in the specific gravity and total solids of the lymph; (2) crystalloid substances, like sugar, salt, etc., which cause an increased flow of lymph more watery than normal. Starling has shov.n that, although the lymphagogues of the second class do not raise the arterial pressure, they do, by attracting water from the tissues and thus causing hydraemic plethora (an excess of blood of low specific gravity), bring about a marked rise of venous, and therefore, what is the important thing for lymph filtration, of capillary pressure. But it can be demonstrated that vaso-dilala- tion with increase of capillary pressure is not in itself sufficient to increase the formation of lymph. We have seen, e.g. (p. 179), that when the chorda tympani nerve is stimulated in the dog the arteri- oles of the submaxillary gland are dilated, and no doubt the pres- sure in the capillaries is increased. No increased flow of lymph, however, takes place from the submaxillary lymphatics during even prolonged excitation of the chorda, nor do the lymph spaces of the gland become distended (Heidenhain). In the horse also the spon- taneous flow of lymph from the quiescent parotid is not appreciably altered by excitation of the secretory nerves of the gland or by pilocarpine (Carlson). There is every reason to believe that during active secretion of saliva tissue liquid is really formed from the blood in increased amount, and that it is from the tissue spaces that the gland-cells directly obtain the increased supply of water and other substances necessary to sustain the increased secretion. But a balance is maintained between the production of tissue liquid and its removal by the gland-cells. When the gland is quiescent, the small amount of tissue liquid normally formed from the blood capillaries for the nutrition of the cells is balanced by, upon the whole, an equal amount of lymph secreted from the tissue spaces into the lymph capillaries. We may say, indeed, that the closed lymphatic system has for its great function the regulation of the quantity and quality of the tissue liquid. In glands with an external secretion increased irriga- tion of the tissue spaces from the blood does not as a rule lead to increased flow of lymph, because the surplus fluid is required to form the secretion. In other organs, however, such as the muscles and the ductless glands, it is probable that the augmented irriga- FACTORS CONCERNED IN LYMPH FORMATION 469 tion rendered necessary by functional activity is always associated with an accelerated flow of lymph, which carries off the surplus liquid, including a portion of the waste products. It is probable that an important factor in the production of oedema may be the derangement of the mechanism, whatever it is, through which the adjustment of the rate of formation of tissue liquid to that of lymphatic lymph is achieved. But it must be remembered that in all the organs the blood capillaries not only supply materials to the tissue spaces, but take up materials from them. Indeed, there are facts which indicate that in general water and dissolved substances pass more easily and in greater amount back from the tissues into the blood than into the lymphatics. So that, while the lymphatics constitute an accessory drainage system, the bloodvessels irrigate the tissues and drain them as well. A consequence of this, as well as of the great difference in the capacity of the different tissues for storing water, is that the amount of tissue lymph formed can never be estimated from the amount of lymphatic lymph leaving an organ. Thus the flow of lymph from the limbs at rest is very small in com- parison with the flow from the abdominal viscera, which constitutes the great bulk of the lymph passing along the thoracic duct. This, however, does not prove that very little liquid passes out of the limb capillaries, for"the chief tissue of the limbs, the muscles, pos- sesses a far greater storage capacity for water than the intestines and digestive glands. In the one case we have a field whose soil takes up a great deal of water and is not easily saturated; in the other a field whose soil soon becomes water-logged and refuses to take up any more. With the same supply from the irrigating ditch, little or no water may drain off at the foot of the first field, a great deal at the foot of the second. Where the main lymphatics are themselves blocked by mechanical pressure or by inclusion in a ligature, the balance is, of course, grossly upset by the failure of the outflow of lymph to keep pace with its formation. Where an obstruction on the course of the veins is responsible for the oedema, the lymphatic outflow, to be sure, is not directly interfered with. But the nutrition and the respiration of the vascular walls themselves, including the endo- thelium of the capillaries, necessarily suffers from the insufficiency of the blood- flow, and the crippled capillaries may very well become abnormally permeable for water, salts, and the other constituents of lymph. And while ordinary mechanical filtration may not be a factor, or a very unimportant one, in the passage of liquid through the normal capillary wall, it may become far more effective when it acts on a damaged wall. The tissue cells also suffer from lack of oxygen and nutritive material, and from the accumulation of waste products, including acid substances, which cause them to take up and to hold more water than normal. Even in the absence 47° FORMATION OF LYMPH of changes in the mechanical conditions of the blood and lymph circulations, alterations in the tissues must be recognized as among the causes of oedema (Fischer). Thus it is clear that the interpre- tation of such an apparently simple experiment as the production of oedema by the ligation of a vein needs great care. Whatever it proves, it may be said with confidence that it does not prove that the increased capillary pressure is the direct cause of the oedema. A mere increase in the capillary blood-pressure does not of itself accelerate the formation of tissue liquid from the blood any more than that of lymph from the tissue liquid, as is shown by the fact that, when the chorda tympani is stimulated after injection of a dose of atropine sufficient to prevent all salivary secretion, there is neither oedema of the gland nor increase in the flow of lymph from it, although the arterioles are as widely dilated as before. When the blood-pressure is greatly increased in the anterior portion of an animal by clamping the aorta, or in the whole animal by continued stimulation of the cut medulla oblongata or the splanchnic nerves, the blood does not alter in concentration in the least, showing that no sensible increase in the passage of water into the lymph has occurred. After division or embolism of the medulla oblon- gata, and consequent paralysis of the vaso-motor centre and general vascular dilatation, it is stated that the injection of sodium chloride produces an increase in the lymph-flow as great and as durable as in the normal animal, and which can continue even after death (Pugliese). The action of the first class of lymphagogues, which cannot be explained as the consequence of an increase of capillary pressure, because the pressure in the capillaries is not consistently increased, and may even in the case of some of these lymphagogues be diminished, is attributed by Starling to an injurious effect on the capillary endothelium (and especially on the endo- thelium of the capillaries of the liver, since nearly the whole of the increased lymph-flow comes from that organ), which increases its permeability. But it is not easy to distinguish an increase of per- meability produced by lymphagogues from an increase of secretory activity of the endothelial cells. Hamburger, too, has brought forward results which it is difficult to reconcile with a theory of filtration even for the second class of lymphagogues. Further, Heidenhain has shown that some time after injection of a crystalloid substance, like sugar, into the blood, a greater percentage of the substance may be found in the lymph than in the blood. Now, when a mixture of crystalloids and col- loids is filtered through a thin membrane, the percentage of crystal- loids in the filtrate is never, at most, greater than in the original liquid. And although Cohnstein states that, if time enough be allowed, the maximum concentration of sodium chloride in the lymph, after intravenous injection, becomes approximately the FACTORS CONCERNED IN LYMPH FORMATION 47* same as the maximum in the blood, this fact loses its weight as an argument in favour of the filtration hypothesis when we re- member that, according to Asher, all the solids of the lymph are markedly increased when even small quantities of crystalloids are injected into the veins. Upon the whole, then, it may be con- cluded that up to the present it has not been shown that filtration due to the excess of pressure in the capillaries over that in the lymph spaces is an effective factor in the formation of lymph. Nor is it at all easier to explain lymph formation as a matter of pure osmosis or diffusion. Lazarus-Barlow found, for example, that in the great majority of his experiments the injection of a concen- trated solution of sodium chloride, dextrose or urea into a vein was followed, not by an initial diminution in the outflow of lymph (as might have been expected if the exchange of water between the blood and the tissue spaces, and between the tissue spaces and the lymph capillaries, was regulated solely by differences in osmotic pressure), but by an immediate increase. And Carlson has shown that the osmotic pressure of lymph coming from the active salivary glands, as measured by the freezing-point method, may, under chloroform or ether anaesthesia, be distinctly less than that of the blood-serum. Water must therefore be passing from a liquid of higher to one of lower osmotic concentration. Nevertheless, it would be erroneous to assume that because osmosis and diffusion have not been shown to satisfactorily account for all the phenomena of lymph formation, they exert no influence upon it. It is probable, indeed, that their action is fully as im- portant as in absorption from the alimentary canal, although, as in absorption, it is often overlaid and always modified by the specific permeability of the blood-capillary walls, the lymph-capillary walls, and the tissue cells in general, in virtue of which they exert an action upon the quantity and composition of the lymph analogous to the action exerted in a higher degree by the cells of the digestive glands upon the quantity and composition of the liquids passing into their ducts. It is not difficult to illustrate the fact that phenomena of osmosis and diffusion emerge, although not of course in such purity as in physical experiments, when we study the interchange between the blood and the tissue liquids. If, for example, a hypertonic solution of sodium chloride is injected into the blood, water rapidly passes from the tissues to the blood as it would through a semipermeable membrane, and the blood becomes diluted. At the same time sodium chloride leaves the blood and passes into the tissues, as it would do by diffusion from a place of higher to a place of lower concentration. But after this has gone on for some time, and the concentration of the blood in salt has sunk, it may be, to that in the lymph, salt still continues to pass out of the blood, and the excess of water also leaves the bloodvessels till the osmotic pressure of the blood has again become normal. When isotonic and even hypotonic solutions of sodium chloride are injected, salt also 472 FORMATION OF LYMPH leaves the blood and enters the lymph, although it ought not to do so by diffusion, while water, which might pass from the hypotonic blood into the lymph by osmosis, moves in the same direction from the blood to which the isotonic salt solution has been added. Regulative mechan- isms, in short, exist which tend, with, but also without, the co-opera- tion of diffusion and osmosis, and even, so to say, in their teeth, to bring back the quantitative and qualitative composition of the blood to the normal. Exactly similar phenomena are witnessed when the equilibrium is upset from the other side by the injection of salt solutions into the subcutaneous tissue or the intramuscular connective tissue. Hypotonic sodium chloride solution injected into the sub- conjunctival connective tissue quickly loses water and gains sodium chloride, as it ought to do if under the influence of osmosis and diffusion, and hypertonic salt solutions gain water. But eventually hypertonic, hypotonic, and isotonic solutions, and even serum itself, are completely absorbed, which could not occur in the presence of diffusion and osmosis alone. Sometimes in dropsy it appears that the oedema liquid is absorbed when the patient is put on a diet free as far as possible from salts. The suggestion is that the regulative mechanism which tends to keep the molecular concentration of the blood and lymph approximately constant provides that as the salt content of the body falls, which it does through continued excretion of salts in the urine, water is eliminated in corresponding amount. The Contribution of the Tissue-Cells to the Lymph. — So far we have considered the passage of the lymph constituents, on the one hand through the endothelium of the blood capillaries into the tissue spaces ; on the other, from the tissue spaces through the endo- thelium of the lymph capillaries. But it is not to be supposed that the liquid lying in clefts, partly bounded by blood capillaries, partly by lymph capillaries, partly by tissue-cells, should be affected solely by the first two. The third anatomical element must contribute something to, or withdraw something from, the tissue liquid, and may thus play a part in the formation of lymph from the latter. The recent researches of Asher and his pupils have raised the ques- tion of the relation between the physiological activity of the organs, and especially of the glands, and the formation of the lymph. They conclude that the common doctrine that lymph is simply a diluted blood-plasma is erroneous. Lymph*, they say, far from being a mere nitrate or even a secretion from the blood, is formed by the activity of the organs, and may actually be absorbed by the blood from the tissue spaces. In fact, according to their view, the in- travenous injection of lymphagogues, both crystalloid and colloid, only causes an increased flow of lymph in so far as it leads to in- creased glandular secretion. But this generalization has had only a short-lived vogue, and one by one some of the main results which seemed to support it have been disproved or shown to be capable of interpretation in a different sense. For example, it was stated that secretin causes a flow of lymph from the lymphatics of the pancreas, as well as a flow of pancreatic juice. But it has been shown that the increased production of lymph is not due to the INFLUENCE OP NERVES ON LYMPH FORMATION 473 secretin at all, but to lymphagogue substances, including proteose, extracted with the secretin from the intestinal mucous membrane. A solution of secretin can be prepared which causes a considerable increase in the secretion of pancreatic juice and bile, but no augmen- tation whatever in the flow of lymph from the thoracic duct. Again, it was asserted that peptone, a noted lymphagogue, produces a great increase in the biliary secretion. It has been demonstrated, however, that the action of the peptone is merely to cause expul- sion of the contents of the gall-bladder by the mechanical effect of the swelling of the liver, and not at all to stimulate the liver-cells to form more bile. For it produces no effect on the flow of bile if the gall-bladder be emptied or the cystic duct tied before the injection (Ellinger). That active salivary secretion is not accom- panied by increased lymph-flow from the lymphatics of the salivary glands has been mentioned above. Nevertheless, we may safely assume that the activity of the organs does make a contribution to the lymph — to its solids, if not in any important degree to its water- content, although to say that they alone are concerned in its forma- tion, to the exclusion of the capillaries, is altogether an over-state- ment. The waste-products of the tissues pass into the lymph, and possibly, as Koranyi suggests, may, by increasing its molecular concentration, cause the passage of some water into it from the blood. Or the decomposition of the large protein molecules, which in tissue metabolism are breaking down into numerous smaller molecules, may entail an increase of osmotic pressure in the cells themselves, which in turn may lead to withdrawal of water by the cells from the tissue liquid. The osmotic pressure of the liquid may thus rise, and water may pass into the tissue spaces from the blood. The molecular concentration of lymph (except in anaesthe- tized animals as mentioned above) is in general somewhat greater than that of blood- serum — e.g., in one observation A of serum was 0-605° C., and of lymph 0-610° C. For the solid tissues, the freezing-point of which, however, cannot be as satisfactorily deter- mined as that of liquids, the following values of A were obtained : Brain, 0-65°; muscle, 0-68°; kidney, 0-94°; liver, 0-97°; while for blood it was 0-57° (Sabbatani). To sum up, we may say that while the physical processes of filtra- tion, osmosis and diffusion may play a part in the passage of water and solids through the walls of the blood capillaries, as well as from the tissue-cells into the tissue spaces, and from these spaces into the lymph capillaries, there is much which they leave unexplained, and which at present, for the want of a more precise term, we must attribute to secretory activity. Influence of Nerves on Lymph Formation. — In one instance it appears to have been shown that lymph may be formed under the influence of secretory nerves. In the males of certain 474 FORMATION OF LYMPH aquatic birds erection is due to the filling of the corpus cavernosum. not with blood, but with lymph. The lymph is secreted rapidly by the so-called bodies of Tannenberg when certain sympathetic nerve- fibres are experimentally stimulated, and passes into the corpus cavernosum, which swells up. If a small incision is made in the corpus, a large quantity of clear lymph, which clots slowly on stand- ing, escapes. There is a simultaneous vasodilatation. After erection, the lymph is rapidly and completely reabsorbed (Eckhard, Muller). Although no definite lymph-secretory nerve-fibres have as yet been discovered in mammals and for ordinary tissues, it is possible that they exist (Sihler). As already pointed out, the same volume of liquid as escapes into the ducts of the active submaxillary gland must, upon the whole, pass out of the blood capillaries. On what principle shall we distinguish one only of these processes as physio- logical secretion ? They begin together when the chorda tympani is stimulated. A drug which paralyzes secretory nerve-endings abolishes both effects. The simplest explanation is that the chorda contains secretory fibres which influence the formation both of saliva and of the tissue liquid from which it is recruited; and, so far as this consideration goes, it is just as logical to consider the increase in the supply of tissue liquid as the cause of the increase in the flow of saliva as to consider the increased salivary secretion as the cause of the increased flow of liquid into the tissue spaces. The increased flow of liquid may be brought about either by an action of the nerve on the gland-cells, causing them to produce a hormone, which then effects the blood capillaries (Carlson), or by a direct action on the capillary endot helium. The advantage to cells engaged in the active secretion of saliva of being immersed in an abundant bath of tissue liquid is obvious. The post-mortem flow of lymph, which may continue in some cases long after complete cessation of the circulation — for an hour after injection of dextrose to produce hydrsemic plethora; for as much as four hours after injection of extract of the strawberry, which is a lymphagogue of Heidenhain's first group (Mendel and Hooker) — is a phenomenon whose relation to normal lymph forma- tion has not been definitely settled. It ought to be remembered in this whole discussion that the epithelium of ordinary glands derives its supplies of material from the tissue lymph. The vicissitudes of blood-pressure affect it only in a secondary and indirect manner. On the other hand, the endo- thelial cells of the capillaries are in direct contact with the blood. And it is interesting to observe that in this respect the glomeruli of the kidney and the alveoli of the lungs (if the endothelial lining of Bowman's capsule and the alveolar membrane are assumed to be complete) take a middle place between the glands proper and the quasi-glandular capillaries. CHAPTER IX EXCRETION WE have now followed the ingoing tide of gaseous, liquid, and solid substances within the physiological surface of the body. There we leave them for the present, and turn to the consideration of the channels of outflow, and the waste products which pass along them. In a body which is neither increasing nor diminishing in mass the outflow must exactly balance the inflow; all that enters the body must sooner or later, in however changed a form, escape from it again. In the expired air, the urine, the secretions of the skin, and the faeces, by far the greater part of the waste products is elimin- ated. Thus the carbon of the absorbed solids of the food is chiefly given off as carbon dioxide by the lungs; the hydrogen, as water by the kidneys, lungs and skin, along with the unchanged water of the food ; the nitrogen, as urea by the kidneys. The f seces in part represent unabsorbed portions of the food. A small and variable contribution to the total excretion is the expectorated matter, and the secretions of the nasal mucous membrane and lachrymal glands. Still smaller and still more variable is the loss in the form of dead epidermic scales, hairs, and nails. The dis- charges from the generative organs are to be considered as excre- tions with reference to the parent organism, and so is the milk, and even the foetus itself, with respect to the mother. Excretion by the lungs and in the faeces has been already dealt with. All that is necessary to be said of the expectoration and the nasal and lachrymal discharges is that the first two generally contain a good deal of mucin, and are produced in small mucous and serous glands, the cells of which are of the same general type as those of the mucous and serous salivary glands. The lachrymal glands are serous like the parotid ; and the tears secreted by them contain some albumin and salts, but little or no mucin. The sexual secretions and milk will be best considered under reproduction (Chap. XIX.), so that there remain only the urine and the secre- tions of the skin to be treated here. 475 EXCRETION SECTION I. — EXCRETION BY THE KIDNEYS — THE CHEMISTRY OF THE URINE. Normal urine is a clear yellow liquid acid to litmus and similar indicators, but nearly neutral or very weakly acid in the physico- chemical sense (p. 24). The average specific gravity is about 1020, the usual limits being 1015 and 1025, although when water is taken in large quantities, or long withheld, the specific gravity may fall to 1005, or even less, or rise to 1035, or even more. The quantity passed in twenty-four hours is very variable, and is especially dependent on the activity of the sweat-glands, being, as a rule, smaller in summer when the skin sweats much, than in winter when it sweats little. The average quantity for an adult male is 1,200 to 1,600 c.c. (say, 40 to 50 oz.).* Composition of Urine. — This is very closely related to the quantity and quality of the food. Hence it is impossible to speak of a definite normal composition of the urine. It is essentially a solu- tion of urea and inorganic salts, the proportion of the latter being generally about 1-5 per cent., or double the usual amount of physio- logical liquids. Besides urea, there are other nitrogenous bodies in much smaller quantity, such as ammonia, uric acid, and the allied purin bases, hippuric acid, and kreatinin. Some of these at least are products of the metabolism of proteins within the tissues. And besides the inorganic salts there are certain organic bodies — indoxyl, phenyl, pyrokatechin", skatoxyl — united with sulphuric acid, which are primarily derived from the products of the putre- faction of proteins within the digestive tube. Folin has published analyses of ' normal ' urines from six persons, weighing from 56-6 to 70-9 kilos (average 63-4 kilos), who were kept for seven days on one standard uniform diet. The diet consisted of 500 c.c. of milk, 300 c.c. of cream (containing 18 to 22 per cent, of fat), 450 grammes of ^ggs, 200 grammes of Horlick's malted milk, 20 grammes of sugar, 6 grammes of sodium chloride, water enough to make the whole up to two litres, and 900 c.c. of additional water. The ingredients contained 119 grammes of protein, about 148 grammes of fat, and 225 grammes of carbo-hydrates. The average results of all the deter- minations are given in the following table : * The average quantity of urine varies not only with the season, but also with the habits of the person, especially as regards the amount of liquid taken. The average for seventeen healthy (American) students, whose urine was collected for six to eight successive days in winter, was 1,166 c.c. The highest average in any one individual for the observation period was 1,487 c.c. (for seven days), and the lowest 743 c.c. (for eight days). The greatest quan- tity passed in any one period of twenty-four hours was 2,286 c.c. (by the in- dividual whose average was the highest). The smallest quantity passed in twenty-four hours was 650 c.c. (by the individual whose average was the lowest. 47 Grammes. Containing Nitrogen (Grammes). Percentage of Total Nitrogen. Urea - Ammonia Kreatinin - Uric acid - Nitrogen in other compounds 29-8 *'55 o-37 I3'9 0-70 0-58 O'I2 0-60 87-5 4'3 3.6 0-8 3'75 Total nitrogen — 16-00 — Inorganic SO3 ... Ethereal SO3 - - - - ' Neutral ' SO3 2-92 0-22 O'I7 Percentage of Total Sulphur. 87-8 6-8 5'i Total sulphur as SO3 3'3I Total phosphates as P2O5 - Chlorine 3-87 6'i Titratable acidity in c.c. of decinormal acid - 617 Indican (Fehling's solution =100*) - "77 Volume of urine ... organic, - 1430 c.c. The great influence of diet on the composition of the urine is illus- trated in the following table. Urine I. was obtained from a man weigh- ing 87 kilos on the standard protein-rich diet described above. Urine II. was obtained from the same person on a diet very poor in protein (400 grammes of starch and 300 c.c. of cream), containing only about i gramme of nitrogen, as against 19 grammes in the first diet. I. II. Volume of urine 1170 c.c. 3850 .c. Grammes. Per Cent. Grammes. Per Cent. Total nitrogen - 16-8 3-60 Urea-nitrogen - 14-70 = 87-5 2-2O = 61*7 Ammonia-nitrogen 0-49 = 3-o 0-42 = n-3 Uric acid-nitrogen O'i8 = i-i O'O9 = 2'5 Kreatinin-nitro ge n 0-58 = 3-6 0-60 = 17-2 Nitrogen in other compounds - 0-85 = 4'9 0'27 = 7'3 Total SO3 3-64 Inorganic SO3 • 3'27 = 90-0 0-46 = 60-5 Ethereal SO3 0-19 = 5'2 O-IO = 13-2 . Neutral SO3' 0-18 = 4-8 0-20 = 26-3 Total phosphates as P2O5 4'i I'O Chlorine - 6-1 1-6 mineral, 398 /mineral, 123. organic, 407 ^ ^(organic, 201 120 - o * The indican is given in arbitrary units, the indigo blue being obtained from the urine and then estimated colorimetrically, using Fehling's solu- N ( Titratable acidity in c.c. -- acid - 805-^ Indican (Fehling's solution = 100) EXCRETION The titrable acidity of urine (see p. 25) is chiefly due to the acid (mono- basic) phosphates, such as acid sodium phosphate (NaH2PO4), but in an important degree also to organic acids. According to Folin, indeed, the organic acidity may be more than half the total acidity. Normally the acidity diminishes distinctly, or even gives place to alkalinity, during digestion, when the acid of the gastric juice is being secreted. This is sometimes fancifully denominated the alkaline tide. After a fast, as before breakfast, the opposite condition, the acid tide, occurs. The acidity varies with the quantity of vegetable food in the diet. The urine of herbivora and vegetarians is alkaline, and is either turbid when passed, or on standing soon becomes turbid from precipitated carbonates and phosphates of earthy bases, while that of carnivora and of fasting herbivora, which are living on their own tissues, is strongly acid and clear. Normal human urine may deposit urates soon after discharge, as they are more soluble in warm than in cold water. They carry down some of the pigment, and therefore usually appear as a pink or brick-red sediment. When urine is allowed to stand after being voided, what is generally described as ' acid fermentation ' occurs. The acidity gradually increases; acid sodium urate is produced from the neutral urate, and comes down in the form of amorphous granules, while the liberated uric acid is deposited often in ' whetstone ' crystals, coloured yellow by the pigment (Fig. 177). Calcium oxalate may also Fig. 177. — Uric Acid Fig. 178.— Calcium Oxalate be thrown down as ' envelope/ a, b, or less frequently, ' sand-glass ' crystals, c (Fig. 178). If the urine is allowed to stand for a few days, especially in a warm place, or in a place where urine is decomposing, the reaction becomes ultimately strongly alkaline, owing to the forma- tion of ammonium carbonate from urea by the action of micro-organ- isms (Micrococcus urece, Bacterium urees, and others) which reach it from the air, and produce a soluble ferment urease, in whose presence the urea is split up with assumption of water. Thus : /NH2 /O.NH, C=0 +2H20 - C=0 \NH2 \O.NH4, Urea. Ammonium carbonate. This is a reaction of considerable interest, for the reverse reaction occurs when blood containing ammonium carbonate is circulated through the liver, the ammonium carbonate being converted into urea with loss of water. The enzyme urease is present in higher plants than bacteria and fungi. A solution containing it, conveniently prepared from the soy bean, can be used for the quantitative estimation of urea, and this is probably the easiest and most accurate method. tion as a standard. Fehling's solution is employed because it is a blue liquid of a definite depth of tint already prepared in every physiological laboratory. 479 The substances insoluble in alkaline urine are thrown down, the deposit containing ammonio-magnesic or triple phosphate, formed by the union of ammonia with the magnesium phosphate present in fresh urine, and precipitated as clear crystals of ' knife-rest ' or ' coffin-lid ' shape (Fig. 179), along with amorphous earthy phosphates, and often acid ammonium urate in the form of dark balls occasionally covered with spines (Fig. 182). Calcium phosphate (CaHPO4) is another phos- phate found in sediments deposited from alkaline or faintly acid urine. It is usually amorphous, but sometimes in the form of long prismatic crystals arranged in star fashion, and hence spoken of as stellar phos- phate (Fig. 181). It is not pigmented. It is only in pathological conditions that the alkaline fermentation takes place within the bladder. The reaction of the urine can readily o Fig. 179. — Triple Phosphate. Fig. 180. — Cystin. Fig. 181.— Stellar Phos- phate Crystals. be made alkaline by the administration of alkalies, alkaline carbonates, or the salts of vegetable acids like malic, citric, and.tartaric acid, which are broken up in the body and form alkaline carbonates with the alkalies of the blood and lymph. It is not so easy to increase the acidity of the urine, although mineral acids do so up to a certain limit. If the admin- istration of acid be pushed farther, ammonia is split off from the pro- teins, and is excreted in the urine as the ammonium salt of the acid. Determination of the Acidity. — A titration method is described in the Practical Exercises (p. 515). In speaking of the reaction of blood, it has already been mentioned (p. 25) that we can- not determine by titration the true acidity or alka- linity of a liquid in the physico-chemical sense — i.e., the concentration of the dissociated hydrogen and hydroxyl ions respectively. E.g., when we titrate equal quantities of decinormal* acetic acid and deci- normal hydrochloric acid with decinormal potassium hydroxide, using, say, phenolphthalein as the indi- cator, nearly the same volume of the potassium hydroxide solution will be needed to neutralize each acid. Yet it can be shown by physico-chemical methods that the acetic acid in the strength used is only dissociated to the extent of a little more than i per cent., while about 80 per cent, of the hydrochloric acid is dissociated. The concen- tration of the hydrogen ions is therefore eighty times as great in the hydrochloric as in the acetic acid solution. What we determine by the titration is not the true acidity , but the total amount of hydrogen which can be replaced by metal. The concentration of the hydrogen ions in normal urine is very small, on the average only about 0-003 milli- * A normal solution of a substance contains in a litre a number of grammes of the substance equal to the number which expresses its equivalent weight — a decinormal (usually written ^) solution one-tenth of this amount, a centinormal one-hundredth, etc. Thus, a normal solution of potassium hydroxide contains 56 grammes of KOFI, and a decinormal solution 5*6 grammes in 1,000 c.c. Fig. 182. — Ammo- nium Urate (after Milroy). 48« EXCRETION gramme in the litre, or about thirty times as much as is present in the purest distilled water. Urine departs about as much from neutrality in the one direction as blood does in the other. Urea, CO(NH2)2, is the form in which by far the greater part of the nitrogen is under ordinary conditions discharged from the body. Its amount is as important a measure of protein metabolism as the quantity of carbon dioxide given out by the lungs is of the oxidation of carbon- aceous material. Yet a glance at the table on p. 477 shows that, when the total protein metabolism is greatly reduced by diminishing the protein in the food, the relative as well as the absolute amount of nitrogen eliminated as urea suffers a great diminution. The relative amount of the other nitrogenous urinary constituents, especially of the kreatinin, is markedly increased. The significance of this difference is alluded to in speaking of the kreatinin content of urine, and will have to be again considered under Protein Metabolism. Urea is soluble in water and in alcohol, and crystallizes from its solutions in the form of long colourless needles, or four-sided prisms with pyramidal ends. It can be easily prepared from urine. Urea can also be obtained artificially by heating its isomer, ammonium cyanate (NH4 — O— CN), to 100° C. This reaction is of great historical interest, as it forms the final step in Wohler's famous synthesis of urea, the first example of a complex product of the activity of living matter being formed from the ordinary materials of the laboratory. Heated in watery solution in a sealed tube to 1 80° Ci, urea is entirely split up into carbon dioxide and am- monia, a change which can also be brought about, as already mentioned, by the action of micro-organisms. Nitrous acid, hypochlorous acid, and salts of hypobromous acid carry the decomposition still farther, carbon dioxide, nitrogen, and water being the products of their oxidizing action on urea. Thus: CO.2(NH2) +3NaBrO=3NaBr + 2H2q+CO2+ N2. This reaction is the basis of the hypobromite method of estimating the quantity of urea in urine (Practical Exercises, p. 319). Ammonia. — The ammonia in urine is united with acids in the form of salts. Its formation from proteins is determined, as we shall see later on, by the necessity of neutralizing certain acids produced in metabolism — e.g., those derived from the sulphur and phosphorus of the proteins, or acids administered experimentally. According to some observers, the percentage amount of the total nitrogen in the urine in the form of ammonia remains the same whether the food be rich or poor in protein (Schittenhelm, etc.), but others state that when the protein is reduced there is a relative increase in the ammonia-nitrogen (see table on p. 477) (Folin). Uric acid (QH^l^Cy exists in large amount in the urine of birds. The excrement of serpents consists almost entirely of uric acid. In both cases it is mainly in the form of acid ammonium urate. In con- trast to urea, uric acid is very insoluble, requiring 1,900 parts of hot and 15,000 parts of cold water to dissolve it. In man and mammals the quantity is comparatively small in health, but is increased after a meal containing material (e.g., thymus gland) rich in nucleins, from the nucleic acid of which purin bodies are derived, or sub- stances containing purin bases in the free state — e.g., hypoxanthin in meat. In mammals the amount of uric acid excreted depends little, if at all, upon the quantity of protein in the food, but a great deal upon the quantity of purin bodies, whether free or combined. When nitrogenous food is omitted altogether, the absolute quantity of uric acid is diminished, but the proportion of the total nitrogen of the urine eliminated rs uric acid is increased, since the ' endogenous ' uric acid (p. 596) still continues to be formed and excreted. CHEMISTRY OF URINE 481 The purin bases (sometimes called the nuclein bases, the alloxuric bases, or the xanthin bases) are a group of substances allied to uric acid, and including, besides xanthin itself, hypoxanthin, guanin, adenin, and other bodies. They exist in very small amount in urine, but, like uric acid, are increased in amount by the ingestion of nuclein-contain- ing substances. The greater part of the purin bases produced in the body is transformed into uric acid ; it is only the untransformed residue which appears in the urine. An interesting fraction of the purin bases in the urine which is not related to the nuclein metabolism is composed of the so-called heteroxanthin, derived from caffeine in the coffee and tea, /-methylxanthin, derived from theobromine in the cocoa, and para- xanthin, derived from theophyllin in the tea, consumed as beverages. Hippuric acid (C9H9NO3) occurs in considerable quantity in the urine of herbivora (Practical Exercises, p. 524); in the urine of carnivora and of man only in traces; in that of birds not at all. Its amount is much more dependent on the presence of particular substances in the food than that of the other organic constituents of urine. Anything which contains benzoic acid, or substances which can be readily changed into it (such as cinnamic and quinic acids), causes an increase of the hippuric acid in urine. In fact, one of the best ways of obtaining the latter is from the urine of a person to whom benzoic acid is given by the mouth; the sweat Fig. 183.— Creatin. Fig. 184. — Creatinin-Zinc-Chloride. may also in this case contain a trace of hippuric acid. Chemically it is a conjugated acid formed by the union of benzoic acid and glycin. Amino-Acids. — The only amino-acid hitherto detected with certainty in normal urine is glycin. Oxalic acid is always present, although in very small amount. Some of it comes from the oxalates of the food, but a portion of it arises in the metabolism of the tissues, probably from the decomposition of uric acid. It is known that outside of the body uric acid may be made to yield oxalic acid. Calcium oxalate crystals are -often seen in urinary sediments. Creatinin (C4H7N3O). — Creatinin is the anhydride of creatin (Fig. 183). Its formula differs from that of creatin only in possessing the elements of one molecule of water less; and creatinin can be obtained by boiling creatin with dilute sulphuric acid. From its alcoholic solution it crystallizes in colourless prisms. Creatinin forms crystalline compounds with various acids and salts. One of the best known of these is creatinin-zinc-chloride, formed on the addition of zinc chloride to an alcoholic or waterv solution of creatinin, often in the shape of beautiful thick-set rosettes of needles (Fig. 184). A por- 482 EXCRETION tion of the urinary creatinin is derived from the creatin of the meat taken as food. But this is not its only source, for on a meat-free diet and in starvation creatinin is still excreted. The absolute quantity in the urine on a meat-free diet is constant for one and the same individual, although different in different persons, and independent of the total amount of nitrogen eliminated. Hence on a diet poor in protein the percentage of the total nitrogen excreted as creatinin is much greater than on a protein-rich diet, as shown in the table on p. 477. So constant is the quantity that a determination of the creatinin may be used as a check upon the complete collection of the urine. Carbo-hydrates are normally present in human urine, but only in very small amounts. Three are known with certainty — dextrose, isomaltose, and the so-called animal gum or urine dextrin. Glycuronic acid (C6H10O7), a body which can be derived from dextrose, is con- stantly present in small amount as a conjugated acid, paired with phenol or indoxyl. It gives Fehling's test, and thus may easily be mistaken for sugar. Glycuronic acid becomes coupled very easily with a large variety of substances, including many drugs, and care must be taken after the administration of camphor, chloral hydrate, chloro- form, nitrobenzol, etc., not to confound the largely increased excretion of conjugate glycuronates in the urine with glycosuria. The yeast test will turn out negative if the reduction is due to glycuronic acid, and the polarimeter will show rotation to the right if it is due to dextrose. The total quantity of carbo-hydrates, including glycuronic acid, excreted in the urine of the twenty-four hours has been estimated at 2 to 3 grammes. The quantity of dextrose in normal human urine is about 0-02 per cent., or about one-fifth of the proportion in blood. Proteins, mainly serum-albumin, are also found in normal urine in minute quantities, on the average about 0-0036 per cent. (Morner). Pigments of Urine. — The pigments of urine have not hitherto been exhaustively studied ; but we already know that normal urine contains several, and pathological urines probably additional, pigmentary sub- stances. The best-known pigments in normal urine are urochrome, the yellow substance which gives the liquid its ordinary colour; uroerythrin, the pink pigment which often colours the deposits of urates that separate even from healthy urine ; and urobilin, which, as has been already stated, is identical with the faecal pigment stercobilin, and occurs not only in many febrile conditions, but also in cases with no fever, such as functional derangements of the liver, dyspepsia, chronic bronchitis, and valvular diseases of the heart. The urobilin of urine represents, mainly at least, the portion of the stercobilin which is not excreted with the faeces, but absorbed from the intestine into the blood. The urobilin in normal urine only exists in small amount in the fully-formed con- dition, most of it being present as a chromogen or mother-substance (urobi Imogen), which by oxidation, as on standing exposed to the air, is converted into urobilin. On the addition of ammonia and zinc chloride to a solution of urobilin a beautiful green fluorescence is obtained, and the solution now shows an absorption band between b and F. Urobilin and urochrome are related substances, bat the exact nature of the relation has not been settled. There is some evidence that a portion of the urobilin of urine is not derived from the intestine, but manufactured probably in the liver. In hunger urobilin is still excreted in the urine, although in greatly reduced amount. During menstruation it is markedly increased, both in fasting and in normally fed individuals. Urorosein is a red pigment which is produced from its chromogen by the action of mineral acids — e.g., strong hydrochloric acid — in the presence of an oxidizing agent, especially nitrites. CHEMISTRY OF URINE 483 The pigments of the blood and bile and some of their derivatives are of common occurrence in the urine in disease. Hcematoporphyrin has not only been found in some pathological conditions, but is constantly present in minute traces in normal urine. Certain drugs — e.g., sulphonal — cause an increase in its amount. In paroxysmal hsemoglobinuria, methcsmoglobin, mixed with some oxyhaemoglobin, is found in the urine in large amount ; and it is worthy of note that it is not formed in the urine after secretion, but is already present as such when it reaches the bladder. In the rare condition termed alkaptonuria, a body, alkapton, now known to be identical with homogentisinic acid, C6H3.(OH)2CH2.COOH, a dioxyphenylacetic acid, is present. The urine becomes dark brown on the addition of an alkali, or simply on exposure to air. It gives Fehling's test for sugar. The substance has relations to the aromatic amino-acids tyrosin and phenyl-alanin, and when either of these is given to a person suffering from alkaptonuria, the amount of alkapton excreted is increased. We may suppose, therefore, that in this con- dition the normal decomposition of these products of proteolysis is interfered with. Ferments. — The urine contains traces of proteolytic and amylolytic ferments (Fig. 185). These may be easily separated from it by putting a little fibrin, which has the power of fixing (adsorbing) enzymes, into the urine. Of the inorganic constituents of urine the most important and most easily estimated are the chlorine, phosphoric acid, and sul- phuric acid. Fig. 185. — Pepsin in Urine. Diastatic Ferment in Urine. At Different Times of the Day (Hoffmann). Chlorine. — Much the greater part of the chlorine is united with sodium, a smaller amount with potassium. The chlorides of the urine are undoubtedly to a great extent derived directly from the chlorides of the food, and have not the same metabolic significance as the organic, and even as some of the other inorganic constituents. But it is note- worthy that in certain diseased conditions the chlorine may disappear entirely from the urine, or be greatly diminished — e.g., in pneumonia, and in general in cases in which much material tends to pass out ffom the blood in the form of effusions (p. 515). Phosphoric Acid. — The phosphoric acid of the urine is chiefly derived from the phosphates of the food, but must partly come from the waste products of tissues rich in phosphorus-containing substances, such as lecithin and nuclein. The phosphoric acid is united partly with alkalies, especially as acid sodium phosphate, and partly with earthy bases, as phosphates of calcium and magnesium. The earthy phosphates are precipitated by the addition of an alkali to urine, or in the alkaline 484 EXCRETION 'lermentation. In some pathological urines they come down when the carbon dioxide is driven off by heating; a precipitate of this sort differs from heat-coagulated albumin in being readily soluble in acids (Practical Exercises, p. 524). A small amount of phosphorus may appear in the urine in a less oxidized form than phosphoric acid. Sulphuric Acid. — This is only to a slight extent derived from ready- formed sulphates in the food. The greater part of it is formed by oxidation of the sulphur of proteins. About nine-tenths of the sulphur in normal urine is present as inorganic sulphates, mainly those of potassium and sodium. Of the other tenth, a portion is represented by ethereal sulphates, and the remainder by the so-called ' neutral ' sulphur, including the sulphur associated with the pigment urochrome. A small amount of sulphur occurs in less oxidized forms than sul- phates in such compounds as the sulphocyanide, which is probably, in part but not entirely, derived from that of the saliva, and ethyl sulphide, a substance with a penetrating odour, which appears to be a constant constituent of dog's urine (Abel). Thiosulphuric acid (H2S2O3) occurs almost constantly in cat's urine, often in dog's. It is not free, but combined with bases. The ethereal sulphates are compounds in which the sulphuric acid is united with aromatic, bodies (indol, phenol, etc.). Such are potassium- phenyl-sulphate (C6H5KSO4), potassium-kresyl-sulphate (C7H7KSO4), potassium-indoxyl-sulphate (C8H6NKSO4) , potassium-skatoxyl-sulphate (CjHgNKS^), and two double sulphates of potassium and pyrocatechin. The formation of potassium indoyxl sulphate may be thus represented : Indol, CftH4v I^TT' " on absorption from the intestine is changed into v /C OH CH /OH indoxyl, CeH4\NH ' which + SO2<^OK (potassium hydrogen sulphate) yields SO2\Qj/ e (potassium indoxyl sulphate) + H8O. The ' pairing ' of these aromatic bodies with sulphuric acid renders them innocuous to the organism. Most of the compounds are present in greater amount in the urine of the horse than in the normal urine of man. But in disease the quantity of indican in the latter may be much increased ; and to a certain extent it must be looked upon as an index of the intensity of putrefactive processes in the intestine and of absorp- tion from it. Munk made the observation that in the urine of a starving dog the phenol-forming substances are absent, while in the urine of a starving man they are present in abnormally large amount. The indigo-forming substances (indican), on the other hand, are in hunger excreted in considerable quantity by the dog, and not at all by man (Practical Exercises, p. 518). According to Folin, the indoxyl potassium sulphate or indican of the urine is not to any appreciable extent related to protein metabolism, but for the most part to the putrefaction of protein in the intestine. The indoxyl-potassium sulphate taken by itself may therefore afford a rough index of the intensity of the intestinal putrefactive processes. On the other hand, the total ethereal sulphuric acid cannot be taken as an index of the extent of the putrefaction, for, although absolutely diminished, it is increased relatively to the total excretion of sulphur on a diet poor in protein, or even protein-free (see tables on p. 477). Phenol and kresol can easily be obtained from horse's urine by mixing it with strong hydrochloric acid and distilling. These aromatic bodies pass over in the distillate. Pyrocatechin remains behind, and can be extracted by ether. It gives a green colour with ferric chloride, which becomes violet on the addition of sodium carbonate. CHEMISTRY OF URINE 485 The sulphur of the inorganic sulphates is the fraction of the total sulphur which fluctuates in proportion to the total protein metabolism. In this regard it follows the variations in the urea. It represents ' exogenous ' metabolism. The neutral sulphur occupies a position analogous to that of the creatinin : the smaller the amount of protein in the food, and the smaller therefore the total protein decomposed, the larger is the fraction which the neutral sulphur forms of the total sulphur. The neutral sulphur accordingly represents endogenous metabolism. The ethereal sulphur takes an intermediate position in this regard, but upon the whole it also becomes a more prominent fraction of the total sulphur when the food contains little or no protein. The ethereal sulphates are therefore not entirely derived from the putrefaction of protein. Carbonates of sodium, ammonium, calcium, and magnesium occur in alkaline urine. Their source is the carbonates and the vegetable organic acids of the food. In acid urine a certain amount of carbon dioxide is present, although not firmly united with bases, so that most of it can be pumped out. So-called Physico-Chemical Analysis of Urine.— The freezing-point of urine has often been determined to obtain a measure of the mole- cular concentration, which with the total quantity of urine secreted in a given time was erroneously assumed to afford an index of the work done by the kidney. Clinically the method is of little use, but for certain physiological questions freezing-point determinations are of value and are sometimes combined with determinations of the electrical conductivity, by which we obtain an approximate measure of the number of dissociated ions in unit volume, mainly the inorganic salts. Normally, A has a higher value for urine than for blood — i.e., the molecular concentration of the urine is higher than that of the serum. But when large draughts of water are taken A may be lower for urine than for blood, and in general it varies within far wider limits (from 0-115° to 2-546° C., according to Koppe). The following table from Kovesi and Roth-Sch,ulz shows the changes in A under the influence of water : Time. Urine in C.C. A 10 to 2 240 I -80 2 to 6 255 I'72 6 to 10 161 i-93 IO to 2 131 2-18 2 to 6 1 60 2-23 6 to i» I2O 1-91 II to 12 1-8 litres ' Salvator ' water taken — 12 tO 12.30 500 0-12 12.30 to i 444 O'll . i to 1.30 442 O'lO 1.30 to 2 46 0-78 2 to 2.30 45 1-30 The Urine in Disease. — Although, strictly speaking, a truly patho- logical urine has no place in physiology, the line which separates the urine of health from that of disease is often narrow, sometimes invisible ; while the study of abnormal constituents is not only of great importance in practical medicine, but throws light upon the physiological processes 486 EXCRETION taking place in the kidney, and upon the general problems of metabolism. Even in health the quantity of the urine, its specific gravity, its acidity, may vary within wide limits. A hot day may increase the secretion of sweat, and correspondingly diminish the secretion of urine, and the deficiency of water may lead to a deposit of brick-red urates. A meal rich in fruit or vegetables may render the urine alkaline, and its alkalinity may determine a precipitate of earthy phosphates. But neither the scanty acid urine with its sediment of urates, nor the alkaline urine with its sediment of phosphates, somes into the category of pathological urines; the deviation from the normal does not amount to disease. The maximum deviation from the line of health is the total suppression of the urine. If this lasts long, a train of symptoms, oi which con- vulsions may be one of the most prominent, and which are grouped under the name of uraemia, appears. At length the patient becomes comatose, and death closes the scene. Suppression of urine may be the consequence of many pathological conditions, but there is one case on record in the human subject which, in effect, though not in intention, belongs to experimental physiology. A surgeon diagnosed a floating kidney in a woman. With a natural impatience of loose odds and ends of this sort, he offered to remove it, and in an evil hour the patient consented. The surgeon, a perfectly skilful man, who acted for the best, and to whom no blame whatever attached, carried the kidney to a well-known pathologist for examination. The latter, to the horror of the operator, suggested, from the appearance of the organ, that it was the only kidney the woman possessed. This turned out to be the fact. Not a drop of urine was passed. Apart from this ominous symptom, all went well for seven or eight days; but then uraemic troubles came on, and the patient died on the eleventh or thirteenth day after the operation. The necropsy showed that her only kidney had been taken away. In disease the urine may contain abnormal constituents, or ordinary constituents in abnormal amounts. Of the normal constituents which may be altered in quantity, the most important are the water, the inor- ganic salts, the urea, the uric acid, and the aromatic substances. Water. — A marked and persistent diminution in the quantity of urine — that is to say, practically in the water, with or without an increase in the specific gravity — is suggestive of disorganization of the renal epithelium. In some infective diseases the kidney is liable to be secondarily involved, its secreting cells being perhaps crippled in the attempt to eliminate the bacterial poisons. In the form of paren- chymatous or tubal nephritis which so frequently complicates scarlet fever, the quantity of urine has in some cases fallen to 50 or 60 c.c. in the twenty-four hours. In chronic interstitial nephritis (' granular kidney '), on the other hand, where the structural changes in the tubules are, for a long time at least, comparatively circumscribed, the quantity of urine is often increased and of low specific gravity. In these cases the increase in the blood-pressure, associated with hypertrophy of the heart, may be a factor in the exaggerated renal secretion. In diabetes mellitus the quantity of urine is greatly increased, perhaps in some cases because more urea is excreted than normal, and urea acts as a diuretic, perhaps also because the elimination of sugar draws with it an increased excretion of water to hold it in solution. Although a specific gravity as low as 1002 has been seen in healthy persons (after copious potations), the persistence of a density below 1010 should suggest hydruria. Watson mentions the case of a boy with diabetes insipid us, who voided in twenty-four hours 9 or 10 pints (5 to 6 litres) of urine with a specific gravity of 1002. On the other hand, while the specific gravity has been CHEMISTRY OF URINE 487 occasionally observed to mount in health to at least 1036, its persistence at 1025 or 1030 CT anything above this, especially if the urine is pale and apparently dilute, should suggest diabetes mellitus. Inorganic Salts. — The changes in the quantity of the inorganic con- stituents of the urine in disease are not, in the present state of our knowledge, of as much importance as the changes in the organic con- stituents. The chlorides are diminished in most acute febrile diseases and may even totally disappear from the urine, and their reappearance after the crisis is, so far as it goes, a favourable symptom. In most cases in which the quantity of the urine is markedly lessened, all the inorganic substances are diminished in amount. Urea. — The quantity of urea is, as a rule, increased in fever, either absolutely or in proportion to the amount of nitrogen in the food. In the interstitial varieties of kidney disease the urea is usually not diminished, but when the stress of the change falls on the tubules (parenchymatous nephritis), it is distinctly decreased — it may be even to one-twentieth of the normal. Uric acid is diminished in the urine in gout (perhaps to one-ninth of the normal), not only during the paroxysms, but in the intervals. It accumulates in the blood and tissues, and, as sodium urare, may form concretions in the joints, the cartilage of the ear, and other situations. Watson relates the case of a gentleman who used to avail himself of his resources in this respect by scoring the points at cards on the table with his chalky knuckles. In leukaemia the quantity of uric acid and purin bases in the urine is greatly increased, not only absolutely, but also in proportion to the urea. As much as 4^ grammes of free uric acid, in addition to about ij grammes of ammonium urate, has been found in a urinary sediment in a case of leukaemia. The aromatic bodies, of which indoxyl may be taken as the type, are increased when the conditions of disease favour the growth of bacteria in the intestine — e.g., in cholera, acute peritonitis, and carci- noma of the stomach. A marked increase in the amount of the indican in the urine may, as far as it goes, be taken as an indication that the bacteria are gaining the upper hand in the intestinal tract; a marked diminution is usually a sign that the battle has begun to turn in favour of the organism (Practical Exercises, p. 517). Tryptophane, a sub- stance which we have already recognized among the products of the tryptic digestion of proteins, has been shown to be a precursor of indol, which is formed from it under the influence of bacteria. When trypto- phane is injected into the caecum of rabbits, the indican in the urine is markedly increased. Putrefactive processes in other parts of the body than the intestine may also increase the indican in the urine — e.g., a collection of putrid pus in the pleural cavity. Abnormal Substances in Urine. — Sugar, proteins, the pigments of bile and blood, or their derivatives, are the most important abnormal sub- stances found in solution in the urine. Normal urine, as has been stated, contains a trace of dextrose, but so little that it cannot be detected by ordinary tests, and for practical purposes it may be .con- sidered absent. Dextrose is the sugar found in the urine in diabetes. In the urine of nursing mothers lactose may be present. Pentoses, sugars with five carbon atoms in the molecule (instead of six, as in the hexoses, of which group dextrose is a member), may also occasionally occur in urine. Pentoses give the ordinary reduction tests for sugar, and yield osazones, but do not ferment with yeast. Various plants contain pentoses, and when these are eaten the pentoses are excreted in the urine, but in cases of true pentosuria they originate in the body, possibly from nucleo-proteins. The condition has not the same sinister significance as diabetes. Specific toxic substances produced by bac- 488 EXCRETION terial action have been demonstrated in the urine in certain diseases. Red blood-corpuscles and leucocytes (pus corpuscles, white blood- corpuscles, mucous corpuscles) are the chief organized deposits; but spermatozoa may occasionally be found, as well as pathogenic bacteria — e.g., the typhoid bacillus; and in disease of the kidney casts of the renal tubules are not uncommon. These tube-casts may be composed chiefly of red blood -corpuscles, or of leucocytes, or of the epithelium of the tubules, sometimes fattily degenerated, or of . structureless protein, or of amyloid substance. Abnormal crystalline substances, stich as the amino-acids, leucin (Fig. 186), tyrosin (Fig. 187), and cystin (Fig. 1 80), may be on rare occasions found in urinary sediments; but generally the unorganized deposits of pathological urine consist of bodies actually contained in, or obtainable from, the normal secretion, but present in excess, either absolutely, or relatively to the solvent power of the urine. Cystin is of interest because of its relations to the sulphur of the protein molecule (p. 360). It is not found in the normal organism. It very occasionally forms calculi in the bladder. There arc individuals who constantly pass as much as one-fourth of all the sulphur in the form of cystin, without any other symptoms. Various amino-acids are present in solution in the urine in many pathological conditions. Of these the least soluble are leucin and tyrosin, and this is the reason why they are most easily detected. A general reaction for ammo-acids is their precipitation as sparingly soluble compounds (|3-naphthalinsulphones) by /3-naphthalinsulpho- cWpride in the presence of an alkali (sodium hydroxide). In acute yellow atrophy of the liver leucin and tyrosin have been found in large amounts in the liver itself, as well as in the urine. In phosphorus poisoning these amino-acids, as well as glycocoll, have been detected in the urine, and there is no doubt that other amino-acids, arising from the decomposition of proteins, are also present in such conditions. Sugar. — In diabetes mellitus, although the quantity of urine is usually much increased, its specific gravity is above the normal; and this is due chiefly to the presence of sugar (dextrose), which generally amounts to i to 5 per cent., but may in extreme cases reach 10 or even 15 per • cent., more than half a kilogramme being sometimes given off in twenty- four hours. The name of the tests for dextrose is legion. They are mostly founded on its reducing action in alkaline solution. Hydrated oxide of bismuth (Boettcher), salts of gold, platinum and silver, indigo (Mulder), and a host of other substances, are reducsd by dextrose, and may be used to show its presence. The reduction of cupric salts (Trommer), Fig. 186. — Leucin Crystals. Fig. 187. — Tyrosin Crystals. fermentation by yeast, and the formation of crystals of phenyl-gluco- sazone are the best established tests. (See Practical Exercises, p. 517.) Proteins. — Serum-albumin and serum-globulin are the proteins most commonly found in pathological urine. Both are coagulated by heating the urine, slightly acidulated if it is not already acid, or' by the addition CHEMISTRY OF URINE 489 of strong nitric acid in the cold. Proteoses (albumoses) are also occa- sionally present, e.g., in the disease called ' osteomalacia ' and in con- ditions associated with the formation and especially with the decom- position of pus. They may be recognized by the tests given in the Practical Exercises (p. 525). It is doubtful whether the presence of true peptone has as yet been satisfactorily made out. The presence of bile-salts may be shown by Hay's test or Petten- kofer's test (p. 46-2). The pigments of blood and bile may be detected by the characters described in treating of these substances; the spectrum of oxyhaemo- globin, ormethaemoglobin, or any of the other derivatives of haemoglobin, with the formation of haemin crystals, would afford proof of the presence of the former, and Gmelin's test of the latter. The red blood-corpuscles, seen with the microscope, are the most decisive evidence of the presence of blood, as leucocytes in abundance are of the presence of pus. It should be remembered that pus in the urine of women has sometimes no signifi- cance except as showing that there has been an admixture of leucorrheal discharge from the vagina. (See Practical Exercises, pp. 74, 531.) SECTION II.— THE SECRETION OF THE URINE. We have now to consider the mechanism by which the urine is formed in the kidney from the materials brought to it by the blood. And here the same questions arise as have already been discussed in the case of the salivary and other digestive glands: (i) Are the urinary constituents, or any of them, present as such in the blood ? (2) If they do exist in the blood, can they be shown to be separated from it by processes mainly physical or mainly ' vital ' — in other words, by ordinary nitration, diffusion and osmosis, or by the selec- tive action of living cells ? In the case of the digestive juices it has been seen that some constituents are already present in the blood, but that physical laws alone, so far as we at present under- stand them, cannot explain the proportions in which they occur in the secretions, or the conditions under which they are separated; while other constituents — and these the more specific and important — are manufactured in the gland-cells. In the kidneys the conditions seem at first sight favourable to physical separation, as opposed to physiological secretion. Urine has been described as essentially a solution of urea and salts, and both are ready formed in the blood. The arrangement of the blood- vessels, too, suggests an apparatus for filtering under pressure. Bloodvessels and Secreting Tubules of Kidney. — The renal artery splits up at the hilus into several branches, which pass in between the Mal- pighian pyramids, and form at the boundary of the cortex and medulla vascular arches, from which spring, on the one side, interlobvlar arteries running up into the cortex between the pyramids of Ferrein, and, on the other, vasa recta running down into the boundary layer of the medulla (Fig. 188). The interlobular arteries give off at intervals afferent vessels. Each of these soon breaks up into a glomerulus or tuft of vascular loops, which gather themselves up again into a single efferent vessel of somewhat smaller calibre than the afferent. The glomerulus is fitted into a cup or capsule (of Bowman), which is closely 490 EXCRETION Pig. 1 88. — Diagram of Blood- vessels of Kidney (Klein, after Ludwig). ai, interlobular ar- tery; vi, interlobular vein; g, glomerulus, to which an afferent artery is seen coming from the interlobular artery, and from which an efferent artery proceeds to break up into a capillary network sur- rounding the renal tubules; vs, vena stellata; ar, arteriae rectae ; vb, leash of venae rectae ; vp, vascular network round ducts at apex of a papilla. Fig. 189. — Diagram of Renal Tubule (Klein). A, cortex ; a, layer of cortex immediately under capsule containing no Malpighian corpuscles; a', inner layer of cortex devoid of Malpighian cor- puscles; B, boundary layer; C, papillary zone of medulla; i, Bowman's capsule; 2, neck of cap- sule; 3, proximal convoluted tubule; 4, spiral tubule; 5, descending part of Henle's loop- tubule; 6, the loop; 7, 8, and 9, ascending limb of loop-tubule; 10, irregular tubule; n, distal convoluted tubule; 12, junctional tubule; 13, collecting tubule in a medullary ray or pyra- mid of Ferrein; 14, collecting tubule in the boundary layer; 15, large collecting tubule ending in a duct of Bellini. THE SECRETION OF THE URINE 491 reflected over it, except where the afferent and efferent vessels pass through, and forms the beginning of a urinary tubule. If we suppose the tuft plashed into the blind end of the tubule so as to indent it, it will be easily understood that the single layer of flattened epithelium reflected on the glomerulus is continuous with that lining the capsule, which in its turn is continuous with the epithelial layer of the rest of the urinary tubule. This has been divided by histologists into a number of parts which it is unnecessary to enumerate here, further than to say that the urinary tubule proper begins in the cortex in Bowman's capsule and the proximal convoluted tubule (with its continuation, the spiral tubule), and ends in the cortex with the distal convoluted tubule, the connection between the two being made by a long loop (Henle's) with a descending and an ascending limb (Fig. 189). Between the ascending limb and the distal convoluted tube is interposed the zigzag tubule. The tubule throughout its length is bounded by a basement membrane lined by a single layer of epithelium, which differs in its character in different parts of the tubule The distal convoluted tubs joins by means of the short connecting tubule one of the straight tubules which form the pyramids of Ferrein or medullary rays in the cortex, and which run down into the medulla, always uniting into larger and larger tubes as they go, until at length they open as ducts of Bellini on the apex of a papilla. The two convo- luted tubules (with the spiral and zigzag tubules) are lined by similar epithelial cells with granular contents, and the tendency of the granules to be arranged in rows perpendicular to the basement membrane gives them a striated or ' rodded ' appearance (Fig. 190). The granules are eosinophile (p. 17), which is also a character of the granules of other secreting cells. Towards the lumen the cells may show a brush of pro- cesses, looking like cilia, but in mammals these are not motile. The ascending part of Henle's loop also has cells of the same general char- acter, with numerous granules, although the ' rodding ' may not be so distinct. We shall see directly that the morphological resemblance is the index of a f unction »,1 likeness. The blood-supply of the tubules, especially of the convoluted portions, is exceedingly rich, the efferent vessels of the glomeruli breaking up around them into a close-meshed network of capillaries, from which the blood is collected into inter- lobular veins running parallel to the interlobular arteries between the pyramids of Ferrein. The straight tubules of the medulla are also surrounded by capillaries given off from straight arteries (arteriae rectae) running down into it partly from the arterial arches and partly from efferent vessels of the glomeruli nearest the boundary layer, the blood passing away by straight veins (venae rectas) which join the larger veins accompanying the arterial arches. The greater part of the blood going through the kidney has to pass through two sets of capil- laries, one in the glomeruli, the other around the tubules. Even the portion of it which does not go through the glomeruli has for the most part a long route to traverse in narrow arterioles and venules to and from its capillary distribution. And the mean circulation-time through the kidney has been found to be longer than that through most other, organs (p. 137). Theories of Renal Secretion. — To come back to our problem of the nature of renal secretion, the anatomical structure of the kidney might be expected to throw light upon the question. And, indeed, it was on a purely histological foundation that Bowman established his famous ' vital ' theory of renal secretion. Impressed with the 49« EXCRETION resemblance between the renal epithelium and the epithelial cells of other glands, and with the distribution of the bloodvessels in the kidney, he came to the conclusion that the characteristic con- stituents of urine, including urea, were secreted from the blood by the tubules. To the Malpighian bodies he assigned what he doubt- less considered the humbler office of separating water from the blood for the solution of the all-important solids. To Ludwig, on the other hand, with his whole attention fastened on the mechanical factors by which the flow of urine could be influenced, the tubules Fig. 190. — From a Vertical Section of Dog's Kidney to show the Structure of Different Portions of the Renal Tubule (Klein), a, Bowman's capsule enclosing glomerulus, the capillaries of which are arranged in lobules separated by a little connective tissue. The capsule and glomerulus together constitute a Malpighian body or corpuscle; n, neck of capsule; c, c, convoluted tubules, cut in various directions; b, irregular or zigzag tubule; d, e, and/ are straight tubules, which take part in the formation of a medullary ray or pyramid of Ferrein ; d, collecting tubule ; e, e, spiral tubule; /, narrow part of ascending limb of Henle's loop-tubule; 6, c, and e are lined with rodded epithelium. seemed of secondary importance, while the glomeruli appeared a complete apparatus for filtering urine from the blood into Bow- man's capsule. He saw that the efferent vessel was smaller than the afferent ; that it was therefore easier for blood to come to the glomerulus than to get away from it, and that the pressure in the capillaries of the tuft must be higher than in ordinary capillaries, because the resistance beyond them in the comparatively narrow efferent vessel, and especially in the second plexus, is greater than the resistance beyond a single capillary network. And experi- mental investigation soon showed him that the rate at which urine THE SECRETION OF THE URTNE 493 was formed could be greatly influenced by changes in the blood- pressure. On such considerations, Ludwig founded the ' mechanical ' theory of urinary secretion, which, although in a much' modified form, still divides with the ' vital ' theory the allegiance of physiologists. It is impossible here to enter in detail into a controversy that has extended over more than half a century and produced an extensive literature. The result of the discussion has been, in our opinion, to establish in its essential principles the ' vital ' theory of Bowman, or at least to show that no purely physico-chemical theory as yet constructed will account for all the facts. Ludwig supposed that the urine, qualitatively complete in all its constituents, was simply filtered through the glomeruli, the work done in this filtration being performed entirely at the expense of the energy of the heart -beat represented as lateral pressure in the vessels of the tufts. But as the proportion of salts, and especially of urea, is very far from being the same in urine as in blood, it had further to be assumed that the liquid which passes into Bowman's capsule is exceedingly dilute, and that absorption of water, and perhaps of other constituents, takes place in its passage along the renal tubules. This process of reabsorption he pictured as a purely physical diffusion between the dilute urine in contact with the free ends of the epithelial cells lining the tubules and the much more concentrated lymph with which their deep ends are bathed. The great length of these tubules, as compared with those of most other glands, might indeed seem to indicate a long sojourn of the urine in them, and the probability of important changes being caused in its passage along them. But if we consider the immense length (60 to 70 cm.) of the seminal tubules and of their gigantic ducts (epididymis 6 metres), where, of course, absorption of water on a large scale is out of the question, it will be granted that little can be built upon the mere length of the renal tubules. On the other hand, the salivary glands, where there are no glomeruli, secrete as much water as the kidneys are supposed to filter ; and the pancreas, whose capillaries form the first of a double set, and therefore in this respect correspond to the renal glomeruli, secretes less water than the liver, whose capillaries correspond to the low-pressure plexus around the convoluted tubules of the kidney. So that deductions drawn from the anatomical relations of the bloodvessels are not in this case of much value, unless supported by physiological results. It is somewhat unfortunate that systematic writers have fallen into the habit of discussing the mechanism of urinary secretion as if the Ludwig theory and the Bowman theory presented an exact antithesis, as if the one offered a complete ' mechanical ' explana- tion of a process, which the other viewed as entirely ' vital,' and therefore withdrawn from physical explanation. We need not concern ourselves here with the historical develop- 494 EXCRETION ment of this discussion. Three main questions require our attention : 1. Is there any evidence that reabsorption actually occurs in the tubules ? If reabsorption on an important scale does take place, it follows at once that there must be a difference of function between the two parts of the renal apparatus, through which urinary con- stituents pass in opposite directions. 2. But if there is no reabsorption, or none of importance, it may still be asked whether, the direction of movement of the urinary constituents through the glomeruli and the tubular epithelium being the same, some quantitative or qualitative difference in their activity may not exist, certain constituents, e.g., passing mainly or exclusively through the one or the other. 3. When these questions have been settled, we are in a position to consider the nature of the process by which the urinary con- stituents find their way from the blood into the lumen of the capsules and the tubules, or, if there is reabsorption, out of the tubules into the lymph and blood again, and to see whether or no it can be entirely explained on mechanical and physico-chemical principles. The Question of Reabsorption from the Tubules. — That some absorption can take place from the kidney when the pressure in the ureter is abnormally raised need not be doubted, and when substances like potassium iodide or strychnine are introduced into the ureter or the pelvis of the kidney under these circumstances, they can speedily be detected in the blood. When the ureter pressure (in dogs) is only slightly increased, instead of evidence of reabsorption, we obtain evidence of increased secretion. The volume of urine, the total quantity of sulphate in the urine when sodium sulphate is injected into the blood as a diuretic, and the total amount of reducing sugar when phlorhizin is injected, are all greater on the obstructed than on the normal side. These facts are quite opposed to the idea that nitration and reabsorption are im- portant factors in the preparation of normal urine (Brodie and Cullis) . Changes in the blood-flow through the kidney have nothing to do with the results, since the small increase in pressure in the ureter was shown not to affect the rate of flow of the blood. The attempt has been made to decide whether absorption normally occurs by removing as much of the tubules as possible, and seeing whether the character of the urine is altered. In rabbits the whole or a large portion of the medulla has been excised from one kidney and the other then extirpated. From the mutilated kidney two or three times as much urine was said to flow as was secreted by a control rabbit operated on in the same way, except for the removal of the renal medulla (Ribbert). The conclusion was drawn that the greater quantity of urine escaping was due to the smaller opportunity for reabsorption of the water. But experiments men- THE SECRETION OF THE URINE 495 tioned in Chapter XL suggest a different interpretation of these ob- servations. And Boyd, who repeated Ribbert's work, obtained quite different results after partial removal of the medulla. He found it impossible to remove the whole. So that hitherto the direct method of eliminating the tubules has left the matter where it was. Some light has been thrown on this question, by taking advantage of the anatomical fact that the kidney of batrachians, and, indeed, that of fishes and ophidia as well, has a double blood-supply. The renal artery gives off afferent vessels to the glomeruli; the vena advehens, or renal portal vein, breaks up, like the portal vein in the liver, into a plexus of capillaries surrounding the tubules, and there seems to be no communication between the two vascular systems. By tying all the arteries going to the kidneys in frogs the circula- tion through the glomeruli can be completely cut off, while ligation of the renal portal vein does not affect the blood-supply of the glomeruli, though markedly interfering with that of the tubules. Gurwitsch has found that, after ligation of the renal portal vein of one kidney in (male) frogs, the flow of urine from that kidney is much diminished as compared with the other. He argues that if reabsorption of dilute urine filtered through the glomeruli takes place in the tubules, the opposite result ought to be obtained, since the glomeruli are not affected, while any absorptive power of the tubules must be crippled or abolished. Experiments on the Excretion of Pigments by the Kidney. — In connection with the second question, and also incidentally with the first, the results of experiments on the distribution of pigments in the kidney after their injection into the blood have often been appealed to. Heidenhain injected indigo - carmine into the blood of rabbits, and after a variable time killed them, cut out the kidneys, and flushed them with alcohol, in which the pigment is insoluble. His results were as follows: (i) When the Fif- I9I--Diasrfn °* Distri])U- spinal cord was cut before the injec- tion of Pigment in Kidney after ,f ., ,, , injection into Blood. The cor- tlon m order to reduce the blood- tex between a and 6 and be- pressure, the blue granules were found tween c and d was cauterized ifl the < roddecl • epithelium of the before the injection. In the 1.1.1.1 j« blank wedge-shaped portions, i, convoluted tubules and the ascending there was no pigment, in the limb of Henle's loop, and in the lumen zones shaded like 2 there was of the tubuleS, but nowhere else. ' some pigment, but no* so much -.-. i , as in the areas shaded like 3. Bowman s capsules contained no pig- ment. The renal cortex was coloured blue. (2) When the spinal cord was not cut, the pigment was found in the medulla and pelvis of the kidney, as well as in the cortex, but always in the lumen of the tubules, and not in the epithelium, except in the situations mentioned. (3) If a portion of the cortex 496 EXCRETION of the kidney had been cauterized with nitrate of silver before in- jection of the pigment, the spinal cord being left intact, a wedge of the renal substance, corresponding to this area, remained coloured only in the cortex, although the rest was blue in the medulla also. The ' rodded ' epithelium was filled with blue granules as before (Fig. 191). (i) shows that the epithelium is capable of excreting some sub- stances at least. (2) appears to show that when the blood-pressure is normal water is poured out from some part of the tubule, and washes the pigment separated by the ' rodded ' epithelium down towards the papillae. (3) suggests that it is through the glomeruli that most of the water passes. For cauterization has not destroyed the power of the epithelium to excrete pigment, and therefore, presumably, would not have destroyed its power to excrete water if it possessed this power in any great degree; and the glomeruli and their capsules are the only other part of the renal mechanism which can have been affected. It must be carefully noted that these experiments do not prove that urea is secreted by the tubular epithelium. Indeed, after section of the cord no accumulation of urea takes place in the kidney (Cushny). It would be equally erroneous to conclude from this that the cells of the tubules do not secrete urea. For the reduction of the blood-supply may have rendered them incapable of doing so. When pigments are injected into the dorsal lymph- sac of a frog without interference with the renal circulation, they are found plentifully in the lumen of the convoluted tubules, and also in the epithelial cells lining them. The suggestion has been made that the pigments have been absorbed by the cells from the lumen, and not excreted by them into it. And certainly pigments soluble in the cytoplasm or in the substances that form the envelopes of cells, and therefore capable, like methylene blue, of staining them during life, might be taken up by the renal epithelium if excreted into the tubules by the glomeruli, and might cause staining of them, par- ticularly, of course, of the free ends of the cells next the lumen. But this suggestion is inadmissible, since, on injection of the same pigments after ligation of the renal portal vein, the convoluted tubules contain little or no pigment in their lumen. And when the urinary flow is stopped on one side in mammals by temporary com- pression of the renal artery, the corresponding kidney takes up fully as much carmine as its fellow (Carter). There is no doubt that not only pigments capable of ' vital staining,' like methylene blue, but also pigments which do not stain living cells, are taken up from the blood (or lymph) by the epithelial cells, and, lying in vacuoles in their cytoplasm, are transported towards the lumen, and there extruded. It is not the solubility of the pigments in lipoids, and therefore their solubility in the supposed lipoid envelope of the cells, which determines whether they shall be excreted. The degree in THE SECRETION OF THE URINE 497 which they are capable of being presented to the cells in non-colloid solution appears to some extent to be a determining factor. The pigments not taken up are highly colloidal (Gurwitsch, Hober). Shafer has recently confirmed Heidenhain's statements as to the place of excretion of indigo-carmine. When leuco-indigo-carmine (a colourless reduction-product of indigo-carmine) was injected, the blue oxidized substance was found in the lumen of the convoluted tubules and in the collecting tubules, but not at all in the Bow- man's capsule. The cells of the convoluted tubules were colour- less, because they kept the pigment in its reduced condition, and it only became oxidized in the lumina of those parts of the tubules whose contents, according to Dreser, show an acid reaction. On oxi- dation by peroxide of hydrogen the cells of the convoluted tubules became faintly green, but the Bowman's capsule remained colourless. This can only be explained on the assumption that the leuco-product of the pigment was excreted by the cells of the convoluted tubules. But these cells are far from taking up all pigments indifferently. Some pigments are extruded mainly by one part, others mainly by another part, of the renal tubule, and some even by the glomeruli, as shown long ago for ammonium carminate. The glomeruli, how- ever, are in general far less active in this regard than the epithelial cells, and the fact that the latter pick out from the blood such sub- stances as these foreign pigments, which pass through the Mal- pighian tufts unchallenged, renders it likely that the tubules also exercise a special function in the secretion of the normal con- stituents of urine. More direct evidence of this is not wanting, for Bowman saw crystals of uric acid in the epithelium of the convoluted tubules of birds. Heidenhain found that urate of soda injected into the blood of a rabbit is excreted by the epithelium of the convoluted tubules and the ascending part of Henle's loop, just as is the case with indigo-carmine. And Nussbaum's experi- ments, although not quite conclusive, have made it probable that in the frog urea is actually separated by the epithelium of the tubules. They were founded on the anatomical peculiarity in the renal circulation of the frog already mentioned. By tying the renal arteries in that animal, he thought he could at will stop the circula- tion in the glomeruli, and he found that after this was done there was no further spontaneous secretion of urine. But when urea was injected intravenously the secretion of urine again began, urea being eliminated by the kidneys, and water along with it. Sugar; peptone, and egg-albumin, injected into the blood, no longer passed into the urine, even when the secretion was excited by simultaneous injection of urea, although they readily did so when the arteries were not tied. He concluded that the Malpighian corpuscles have the power of excreting water, sugar, peptone, and albumin, while the epithelium of the tubules excretes urea as well as water. Beddard has confirmed Nussbaum's statement that when all th cool, and weigh. From the weight of the barium sulphate, the sulphuric acid in 50 c.c. of urine is easily calculated (SO4 in I gramme of barium sulphate, 0-41187 gramme) (Folin). (3) Quantitative Estimation of the Sulphuric Acid united with Aromatic Bodies (Aromatic or Ethereal Sulphates). — Put 200 c.c. of the same urine as used in (2) into a beaker. Add 100 c.c. of 10 per cent, barium chloride solution in the cold. Let stand for twenty-four hours. Then decant off the clear supernatant liquid, and filter it. Measure 150 c.c. of the clear filtrate, corresponding to 100 c.c. of the urine, into a 400-0. c. Erlenmeyer flask. Add 10 or 15 c.c. of concentrated hydrochloric acid and 10 to 15 c.c. of 4 per cent, potassium chlorate. Heat the mixture to boiling, and proceed as in (2). From the weight of the barium sul- phate, the ethereal sulphuric acid in 100 c.c. of urine can be calculated. Deducting this from the quantity per 100 c.c. of urine obtained in (2), we get the amount of inorganic sulphuric acid per 100 «.c. (Folin). 6. Indoxyl (contained in the urine as indican, the potassium salt of indoxyl-sulphuric acid) can be oxidized into indigo, and so detected and estimated. A qualitative test is the following: Ten c.c. of horse's urine is mixed with 10 c.c. of Obermayer's reagent (pure concentrated hydrochloric acid containing 2 to 4 parts of ferric chloride in 1,000), and shaken well 51 8 EXCRETION for a minute or two ; a bluish colour appears if, as is generally the case, indoxyl is present, indigo (C16H10N2O2) being formed by the oxidizing action of the ferric chloride on the indoxyl, the compound of which with sulphuric acid has been broken up by the hydrochloric acid. The urine must be free from albumin. In performing the test in human urine, which contains a smaller quantity of the indigo-forming sub- stance, the fault blue liquid should be shaken up with a few drops of chloroform. The latter takes up the colour, which is thus rendered more evident. If there is difficulty in obtaining the reaction, the urine may first be decolorized by precipitating it with acetate of lead, avoiding excess. The precipitate is filtered off, and the test then applied to the clear filtrate. The skatoxyl of urine can also be oxidized to indigo, but it is present in far smaller amount. The average quantity of indigo obtained from a litre of horse's urine is about 150 milligrammes ; from a litre of human urine, not a twentieth of that amount. For comparative quantitative determinations the method of Folin may be used. One-hundredth of the twenty-four hours' urine is taken. In this the indigo is developed by the addition of an equal volume of Obermayer's reagent (p. 5l«7), and the indigo-blue dissolved by means of 5 c.c. of chloroform. The chloroform solution is then compared colorimetrically with Fehling's solution. This can be done by putting the indigo solution and 5 c.c. of the Fehling's solution respectively into small test-tubes of equal calibre, and comparing the depth of tint. If the Fehling's solution is stronger than the indigo solution, run water into the former from a pipette, graduated in tenths of a c.c., shaking up after each addition, till equality of tint has been reached. If the indigo solution has a stronger blue colour than the Fehling's solution, dilute a measured amount of it first of all with such a quantity of chloroform (say an equal volume) as will make its tint distinctly weaker than that of the Fehling's solution. Then dilute the Fehling's solution with water, as before, till the tint is the same. From the amount of dilution the quantity of indigo can be expressed in arbitrary units, taking Fehling's solution as 100. Thus, if I c.c. of water must be added to the 5 c.c. of Fehling's solution, the indican can be expressed as = 5— = 83. The comparison can be made more accurately by 5 a colorimeter, if one is available. To determine the absolute amount of indigo obtained, comparison must be made with a standard solution of indigo. 7. Urea — (i) Decomposition of Urea. — Heated dry in a test-tube, it gives off ammonia. The residue contains biuret, which, when dissolved in water, gives a rose colour with a trace of cupric sulphate and excess ot sodium hydroxide (or of the hydroxides of certain other metals of the alkalies and alkaline earths (p. 8). Some proteins — peptones and albumoses — in the presence of the same reagents, give a similar colour, the so-called biuret reaction. (2) Quantitative Estimation — Marshall's Urease Method. (According to Van Slyke). — One-half c.c. of urine is measured into the bottom of tube A (Fig. 194). Five c.c. exactly of a 0-6 per cent, solution of KH2PO4 is then run in from a burette, and I c.c. exactly of a 10 per cent, watery solution of the dry urease powder. The phosphate is added as a ' buffer ' to maintain the enzyme action near its maximum rate, by preventing the development of alkalinity as the production of ammonium carbonate goes on. The solutions in the tube are well PRACTICAL EXERCISES 519 mixed, 2 drops of caprylic alcohol are added to prevent foaming during the subsequent aeration (by Folin's method, p. 521), and the stopper bearing the aerating tubes is put into place. Let stand twenty minutes at room temperature of 15°, or fifteen minutes at 20° C. or above, for complete decomposition of urea. This time must not be shortened unless more enzyme is used, but no harm is done if it is longer. Mean- while measure into tube B 25 c.c. |^ hydrochloric or sulphuric acid, i drop of i per cent, sodium alizarine sulphonate indicator, and i drop caprylic alcohol, and connect the tubes as shown in Fig. 194. After the twenty minutes for decomposition of urea has elapsed, the air current is passed for half a minute to sweep into B, a small amount of ammonia which has escaped into the air space of A during the decom- position. Then A is opened, and 4 to 5 grammes potassium carbonate added to liberate the ammonia from the ammonium carbonate formed. Now let the air cur- rent pass gently through the tubes for the first minute and then rapidly until all the ammonia has been brought into the acid in B. The time necessary for this will depend upon the speed of the air current (varying from five minutes to half an hour, according to the efficiency of the pump or vacuum by which the air is moved) and should be determined once for all. The excess- acid in B is now titrated with ^ sodium hydroxide. The Fig. 194. — Apparatus for deter- mining Urea Content by means of Urease. (After Van Slyke.) weight of nitrogen contained in the urea of 100 c.c. of urine is (25 - x) x 0-056, where x is the number of c.c. of the sodium hydroxide solution necessary to neutralize the acid. Included in the urea nitrogen is the nitrogen of any ammonia originally present in the urine. This can be determined separately, if desired, by putting 5 c.c. of urine into A without urease, adding the carbonate at once, and then aerating through the acid in B. The acid neutralized in this case is multiplied by 0-0056 to give the ammonia nitrogen in 100 c.c. of urine, and this being deducted from the previous result, the net urea nitrogen is obtained. The method as above described is adapted for urine containing not more than 3 per cent, urea, which is about the maximum found in human urine. Cat's or dog's urine should first be diluted to reduce the urea below 3 per cent. Clinical Urease Method by Direct Titration of Urine. — Two 5 c.c. portions of urine are put into flasks a and b of 200 to 300 c.c. capacity, and diluted with distilled water to about 100 to 125 c.c. One c.c. of 10 per cent, urease is added to flask a, and a few drops of toluene to each. The flasks are allowed to remain, well stoppered, at room tempera- ture over night, then titrated to a distinct pink colour with ^ hydro- chloric acid, using methyl orange as indicator. From the amount of hydrochloric acid needed for a is deducted the amount needed for b; and also the amount previously determined to be necessary to neutra- lize the alkalinity of the enzyme solution. The remaining number of c.c. multiplied by 0-6 gives the urea in gvammes per litre of urine, since 520 EXCRETION i c.c. of N hydrochloric acid is equivalent to 3 milligrammes of urea. Instead of the dried urease preparation, an extract of finely powdered soy beans can be made by mixing 25 grammes of the powder with 250 c.c. of distilled water, and allowing it to stand with occasional shaking for an hour. Twenty-five c.c. of ^ hydrochloric acid are then added, and the mixture allowed to remain a few minutes longer (best at about 35° C.). The mixture is filtered, treated with a few drops of toluene, and preserved for use in a stoppered bottle. It must be made up fresh after a few days as it does not keep long. The solution is alka- line to methyl orange, and 2 c.c. (the quantity used in the above clinical determination of urea) generally re- quires about o«3 c.c. of N hydro- chloric acid for neutralization. A less exact method which is very rapid, and is therefore much used in clinical determinations, is the Hypo- bromite Method. The urea is split up by sodium hypobromite (p. 480), and the carbon dioxide being absorbed by the excess of sodium hydroxide used in preparing the hypobromite, the nitrogen is collected over water in an inverted burette. It is easy to calcu- late the weight of urea corresponding to a given volume of nitrogen measured at a given temperature and pressure. The nitrogen of urea is f §, or fa of the whole molecular weight. Now, i c.c. of N weighs, at 760 millimetres of mercury and o° C., 0-00125 gramme. Therefore, i c.c. of N corresponds to 0-00125x^=0-00268 gramme urea. Suppose, now, that i c.c. of urine was found to yield 10 c.c. of N measured at 17° C. and 750 millimetres barome- tric pressure. Since a gas expands volume at o° for every Fig. 195. — Hypobromite Method of estimating Urea. A, glass thimble; B, bottle, through the rubber cork of which pass two short glass tubes, one connected ^_ w iw .„...„„, „,„ v iw C¥C1> by the rubber tube C with a degree above o°, we must correct the burette D, and the other armed apparent volume of nitrogen by multi- plying by f |§. Since the volume of a gas is inversely proportional to the pressure, we must further multiply by Ijft. Thus we get 10 x \ J$ x fflf - **!££• =9*29 c.c. as the volume of the "nitrogen reduced to o° C. and 760 milli- metres of mercury. Multiplying this by 0-00268, we get 0-0249 gramme urea for i c.c. urine, which for a daily yield of 1,200 c.c. would corre- spond to 29-88 grammes urea. As a matter of fact, however, it has been found that there is always a deficiency of nitrogen — that is, a given quantity of urea yields less than the estimated amount of gas. A gramme of urea in urine, instead of giving off 373 c.c. of nitrogen, gives only 354 c.c. at o° C. and 760 millimetres pressure. We must therefore take i c.c. of N as correspond- with a short piece of rubber tube F. F is provided with a pinch- cock. The burette is supported on a stand, and immersed in water contained in the glass cylinder E. PRACTICAL EXERCISES 521 ing to 0-00282 gramme, instead of 0-00268 gramme urea. But it is affectation to make this correction if, as is seldom done in hospitals, the temperature is not taken into account. A convenient apparatus is shown in Fig. 195. In B place 10 c.c. of a solution made by adding bromine to ten times its volume of 40 per cent, sodium hydroxide solution. Mix 5 c.c. of urine with 5 c.c. of water. Put 5 c.c. of the mixture into the thimble A, which is then set in the small bottle B. The cork is now carefully fixed in B, and the tube F being open, the level of the water in the burette is read off. The pinchcock having been closed, the bottle B is now tilted so that the urine in the thimble is gradually mixed with the hypobromite solu- tion, and the nitrogen given off is added to the air in the burette and its connections. The level of the water in the burette is therefore depressed. When gas ceases to be given off, and a short time has been allowed for the whole to cool, the tube is raised till the level of the water is once more the same inside and out. The. level is again read off ; the difference of the two readings gives the volume of nitrogen at the temperature of the air and the barometric pressure. In order that the temperature of the water may be the same as that of the air, the cylinder should be filled a considerable time before the observations are begun. 8. Estimation of the Ammonia in Urine (Folin's Method). — Ammonia is liberated by addition of a weak alkali (sodium carbonate). Then the ammonia is driven out at ordinary temperature by a strong current of air and taken up in decinormal acid, which is then titrated with decinormal alkali. The apparatus employed consists of — (i) A cylinder of about 45 cm. diameter, with a rubber stopper through which pass two glass tubes. One of the tubes goes nearly to the bottom of the cylinder, and the other end is connected, through a U-tube filled with cotton, with a tube containing sulphuric acid. The second tube is cut off short below the rubber cork, and its other end is connected, through a U-tube con- taining cotton, with a sulphuric acid tube (or with two in series) . (2) A water-pump to draw or force air through the apparatus (600 to 700 litres in an hour) . Put into the first sulphuric acid tube 25 c.c., into the second 10 c.c. decinormal acid and some water; into the cylinder 25 c.c. of filtered urine, 8 to 10 grammes sodium chloride, 5 to 10 c.c. of petroleum or toluol to prevent foaming, and last of all i gramme dried sodium carbonate. At once close the cylinder and allow a strong stream of air to pass through the apparatus. At a temperature of 20° to 25° (room temperature), and using 600 to 700 litres of air an hour, all the ammonia is in the sulphuric acid in one to one and a half hours. The contents of the sulphuric acid tubes are put into a beaker and titrated with decinormal alkali, using lacmoid (litmoid) or rosolic acid as indicator. Deduct the number of c.c. of alkali used from the number of c.c. of the decinormal acid originally taken, and multiply the remainder by 1-7034 to get the quantity of ammonia in milligrammes. The method can be employed also for albuminous urine. 9. Estimation of the Total Nitrogen. — It is sometimes more important to determine the total nitrogen of the urine than the urea alone. This is conveniently done by Kjeldahl's method (or some modification of it), which can also be applied to the estimation of the nitrogen in the faeces, or in any of the solids or liquids of the body. It depends on the oxidation of the nitrogenous matter (or, rather, in the case of urine, mainly its hydrolysis) in such a way that the nitrogen is all represented 522 EXCRETION as ammonia. The ammonia is then distilled over, collected and esti- mated, and from its amount the nitrogen is easily calculated. In urine the method can be carried out by adding to a measured quantity of it (say 5 c.c.) four times its volume of strong sulphuric acid, and boiling in a long-necked flask (capacity 200 c.c.), after the addition of a globule of mercury (about o-i c.c.), which hastens oxidation and obviates bumping. A part of the mercuric sulphate formed remains in solution ; the rest forms a crystalline deposit. The heating should continue for half an hour, or until the liquid is decolourized. It should be kept gently boiling. This completes the process of oxidation ; and the next step is to liberate the ammonia from the substances with which it is united in the solution, and to distil it over. Dilute the liquid with water, after cooling, up to about 150 c.c., and pour into a larger long- necked flask. Add enough of a solution of sodium hydroxide (specific gravity about 1-25) to render the liquid alkaline, avoiding excess, as this favours bumping. The proper quantity can be found by deter- mining beforehand how much of the alkali is needed to neutralize the acid used for oxidation, and a little more than this amount should be added. Twenty c.c. of strong sulphuric acid needs about 75 c.c. of 40 per cent, sodium hydroxide to neutralize it. Bumping may further be prevented by the addition of a little granulated zinc. Shake the flask two or three times. Add also about 12 c.c. of a concentrated solution of potassium sulphide (i part to i£ parts water), which favours the setting free of the ammonia from the amino-compounds of mercury that have been formed during oxidation . Commercial ' liver of sulphur ' will do quite well. Immediately connect the distilling-flask with a worm or Liebig's condenser, and distil the ammonia over into 50 c.c. of standard (decinormal) sulphuric acid (see footnote, p. 473) con- tained in a flask into which a glass tube connected with the lower end of the worm dips. Heat the distilling-flask at first gently, then strongly, and boil for three-quarters of an hour, or until about two-thirds of the liquid has passed over. Then lift the tube out of the standard acid, and continue the distillation for two or thiee minutes longer. The ammonia is now all united with the standard acid, a certain amount of which is left over. By determining this amount we arrive at the quantity com- bined with ammonia, and therefore at the quantity of ammonia. Fill a burette with a decinormal solution of potassium or sodium hydroxide. Add a little methyl-orange solution to the standard sulphuric acid, to serve as indicator. Then run in the potassium or sodium hydroxide till the pink tinge gives place to a permanent but just recognizable yellow. Let x be the number of c.c. run in. Since i c.c. of any deci- normal solution is equivalent to i c.c. of any other, x represents also the number of c.c. of the standard sulphuric acid left uncombined with ammonia; and 50 — x, the quantity combined with ammonia. Then, i c.c. of decinormal sodium or potassium hydroxide being equivalent to i c.c. of decinormal ammonium hydroxide, and i c.c. of decinormal ammonium hydroxide containing 0-0014 gramme nitrogen, we get (50- x) x 0-0014 as the quantity of nitrogen in 5 c.c. of urine. Instead of mercury, potassium sulphate and copper sulphate may be added to the sulphuric acid in order to aid the decomposition in the first stage of the estimation. About 3 grammes of potassium sulphate and i gramme of copper sulphate are added to £ c.c. of urine, and then 5 c.c. of sulphuric acid. The liquid is gently boiled for an hour, or until it is quite clear. The neutralization and distillation are conducted as before, the proper quantity of sodium hydroxide being determined in advance. No potassium sulphide is added, but a small quantity of PRACTICAL EXERCISES 523 talc may be put in to prevent bumping. Instead of methyl orange, ' alizarin red,' which is bright red in the presence of the slightest trace of alkali, may be used. 10. Uric Acid — (i) Qualitative Test for Uric Acid — Murexide Test. — A small quantity of uric acid or one of its salts is heated with a little dilute nitric acid. The colour of the residue left by evaporation becomes yellow, and then red, and on the addition of ammonia changes to deep purple-red. Potassium or sodium hydroxide changes the yellow to violet. In the reaction alloxantin is formed by oxidation of the uric acid. When ammonia acts on alloxantin it is changed into purpuric acid, and this into its ammonium purpurate, the purple-red substance called murexide. Thus: C8H6N408+ NH3 =C8H5N60?+ 2HaO. Alloxantin. Purpuric Acid. The reaction is also given by theobromine (dimethylxanthin), an alkaloid in cocoa, and theine or caffeine (trimethylxanthin), an alkaloid in tea and coffee, which are also purin derivatives (p. 48 1')- (2) Quantitative Estimation — • Folin's Modification of Hopkins' s Method. — The chief reagent is a solution of 500 grammes ammonium sulphate, 5 grammes uranium acetate, and 60 c.c. 10 per cent, acetic acid, in 650 c.c. of water. One hundred and fifty c.c. of urine is measured into a tall, narrow beaker or a cylinder, and 37^ c.c. of the reagent added. If enough urine is available, 200 c.c. of urine and 50 c.c. of reagent are to be used. Allow the mixture to stand without stirring for about half an hour. The uranium precipitate has then settled, and the clear supernatant liquid is removed by siphoning or decantation. One hundred and twenty-five c.c. of this liquid is measured into another beaker, 5 c.c. of strong ammonia added, and the mixture set aside till next day. The precipitate is then filtered off, and washed with 10 per cent, ammonium sulphate solution until the filtrate is quite or nearly free from chlorides. The filter is then removed from the funnel, opened, and the precipitate rinsed back into the beaker. Enough water to make about 100 c.c. is added, and the precipitate is then dissolved by means of 15 c.c. con- centrated sulphuric acid, and at once titrated with ^6 (one -twentieth normal) potassium permanganate solution (made by dissolving 1-581 grammes of the permanganate in a litre of water), each c.c. of which corresponds to 3-75 milligrammes of uric acid. The very first pink coloration, extending through the entire liquid on the addi- tion of two drops of permanganate solution, marks the end point. A correction of 3 milligrammes, owing to the solubility of ammonium urate, is added to the result. 11. Creatinin. — Qualitatively, creatinin may be recognized in very small amounts by Weyl's test. A few drops of a dilute solution of sodium nitro-prusside are added to urine, and then dilute sodium hydroxide drop by drop. A ruby-red colour appears, which soon turns yellow. If the urine is now strongly acidified with acetic acid and heated, it becomes first greenish and then blue. Enough acid must be added to more than neutralize the alkali. Another test which has been made the basis of a quantitative method by Folin is Jaffe's test. A little urine (say 5 c.c.) is put in a test-tube, and then a solution of picric acid in water. The mixture is rendered alkaline by the addition of potassium or sodium hydroxide solution, and a reddish colour is produced, which turns yellow on the addition of 52 4 EXCRETION acid. A similar red colour is given by dextrose, but not unless the solution is heated. Quantitative Estimation of Creatinin by Folin's Method. — It depends upon the comparison of the colour which* creatinin gives with picric acid in an alkaline solution with that of a standard solution of potassium bichromate. Ten c.c. of urine is measured into a 500 c.c. measuring- flask; 15 c.c. of a saturated picric acid solution (containing about 12 grammes per litre) and 5 c.c. of a 10 per cent, solution of sodium hydroxide are added. The mixture is allowed to stand for five minutes. Then water is added up to the 500 c.c. mark, and the flask shaken to mix uniformly. Samples of the liquid are then at once compared colorimetrically with a half-normal solution of potassium bichromate containing 24-55 grammes per litre. The colour of the urine does not introduce a sensible error on account of the great dilution". For exact work the comparison must be made with a good colorimeter. It has been found experimentally that, when to milligrammes of creatinin are present in 500 c.c. of a solution made as described, a layer of the solution 8-1 millimetres in thickness has the same depth of tint as 8 milli- metres of the bichromate solution. Suppose it takes 9 millimetres ol the urine-picrate solution to equal 8 millimetres of the bichromate, O. j then the 10 c.c. of urine contains 10 x — =9-0 milligrammes of creatinin. 12. Hippuric Acid. — From horse's or cow's urine hippuric acid is prepared by evaporating to a small bulk, and adding strong hydrochloric acid. The crystalline precipitate is washed with cold water, then dissolved in hot water, and filtered hot. Hippuric acid separates out from the filtrate in the cold in the form of long four-sided prisms with pyramidal ends. Heated dry in a test-tube, the crystals melt, and benzoic acid and oily drops of benzonitrile, a substance with a smell like that of oil of bitter almonds, are formed. ABNORMAL SUBSTANCES IN URINE. 13. Proteins — (i) Qualitative Tests. — (a) Boil and add a few drops of nitric acid. A precipitate on boiling, increased or not affected by the acid, shows the presence of coagnlable proteins (serum-albumin or globulin). A precipitate of earthy phosphates sometimes forms on boiling. It is distinguished from a precipitate of proteins by dissolving on the addition of acid. (b) Heller's Test. — Put some nitric acid in a test-tube. Pour care- fully on to the surface of the acid a little urine. A white ring at the junction of the liquids indicates the presence of albumin or globulin. If much albumose is present, a white precipitate, which disappears on heating, may be formed. When this test is performed with undiluted urine, uric acid may be precipitated and cause a brown colour at the junction. A similar ring may be found in the absence of proteins when the test is made on the urine of a patient who has been taking copaiba. In very concentrated urine a white ring of nitrate of urea may be formed. A coloured ring is frequently seen, owing to the oxidation of certain chromogens of urine. (c) Filter some urine, and add to the filtrate its own volume of acetic acid. A precipitate may indicate mucin or nucleo -albumin. If any is formed, filter ft off, and add to the filtrate a few drops of potassium ferrocyanide. A white precipitate shows the presence of proteins. (d) Test for Globulin in Urine. — Serum-globulin probably never PRACTICAL EXERCISES 525 occurs in urine apart from serum-albumin. It may be detected thus: Make the urine alkaline with ammonia, let it stand for an hour, and filter. Half saturate the filtrate with ammonium sulphate — i.e., add to it an equal volume of a saturated solution of ammonium sulphate. Serum-globulin is precipitated, serum-albumin is not. (e) Test for Albumose in Urine (Albumosuria] . — Coagulable proteins are removed by boiling the urine (acidulated if necessary), and filtering off the precipitate if any. The filtrate is neutralized. If a further precipitate falls down it is filtered off, the clear filtrate is heated in a beaker placed in a boiling water-bath, and there saturated with crystals of ammonium sulphate. A precipitate indicates that albumoses (proteoses) are present. A slight precipitate might possibly be due to the formation of ammonium urate. A further test may be performed on the original urine if it is free from coagulable proteins, or on the filtrate after their removal. Add a drop or two of pure nitric acid. If albumoses are present, a precipitate is thrown down which disappears on heating, and reappears on cooling the test-tube at the cold-water tap. (2) Quantitative Estimation of Coagulable Proteins (Serum-Albumin and Globulin) — (a) Gravimetric Method. — Heat 50 to 100 c.c. of the urine to boiling, adding a dilute solution (2 per cent.) of acetic acid by drops as long as the precipitate seems to be increased. Filter through a weighed filter. Wash the precipitate on the filter with hot water, then with hot alcohol, and finally with ether. Dry in an air-bath at 110° C., and weigh between watch-glasses of known weight. (6) Esbach's Method. — Esbach's reagent is made by dissolving 10 grammes of picric acid and 20 grammes of citric acid in boiling water (800 or 900 c.c.), and then making up the volume to a litre. The so-called albuminimeter is simply a strong glass tube graduated and marked in a certain way. Fill the tube up to the mark U with the urine. Then add the reagent up to the mark R. Close the tube with the rubber cork, and invert it a dozen times without shaking. Set the tube aside for twenty-four hours, then read off the graduation on the tube which corresponds with the top of the precipitate. The figures indicate the number of grammes of dry protein in a litre of the urine. Suppose the top of the sediment is at 4, this will indicate 4 grammes per litre, or 0-4 per cent. The method is of some clinical importance, owing to its simplicity, although it is, of course, not very accurate. 14. Sugar — (i) Qualitative Tests — (a) Trommer's Test (p. 10). — It is to be remarked that some substances present in small amount in normal urine reduce cupric sulphate — e.g., uric acid (present as urates) and kreatinin — but although a normal urine may thus decolourize the copper solution, it rarely causes so much reduction that a yellow or red precipitate is formed, as is the case in diabetic urine. Glycuronic acid (p. 482) also reduces cupric salts, as does alcapton or homogentisinic acid, a substance found in rare cases in disease (p. 483). (6) Fehling's Test. — Fehling's solution (p. 526) is brought to the boil ill a test-tube, a little of the urine then added, and the change of colour noted. Benedict's modification of Fehling's solution may also be used. It has the advantage that it keeps indefinitely, and therefore is always ready for use, and is also said to be more delicate. (c) Phenyl-Hydrazine Test. — This test depends upon the fact that phenyl-hydrazine forms with sugars such as glucose (dextrose), maltose, isomaltose, etc., but not with cane-sugar, characteristic crystalline substances (phenyl-glucosazone, phenyl-maltosazone, etc.) which can be recognized under the microscope, and are distinguished from each other by melting at different temperatures. Phenyl-glucosazone 526 EXCRETION melts at 205° C. To perform the test for dextrose in the urine, proceed thus: Put 5 c.c. of urine in a test-tube, add I decigramme of hydrochlorate of phenyl-hydrazine and 2 decigrammes of sodium acetate. It is sufficiently accurate to add as much phenyl-hydrazine as will lie on a sixpence (or a dime) and twice as much sodium acetate. Heat the test-tube in a boiling water-bath for half an hour. Then cool at the tap and examine the deposit under the microscope for the yellow phenyl-glucosazone crystals (Fig. 196). Sometimes the osazone pre- cipitate is amorphous. If this should be the case, the precipitate, if no crystals can be seen, must be dissolved in hot alcohol. The solution is then diluted with water and the alcohol boiled off, when the osazone, if any be present, will crystallize out. Very minute traces of sugar can be detected in this way (as little as oa per cent, in urine). Often in normal urine yellow crystals are deposited during the first fifteen minutes' heating. They must not be mistaken for gluco- sazone. They probably consist of a compound of glycuronic acid and phenyl-hydrazine. They are changed as the heating goes on into an amor- phous brownish - yellow precipitate (Abel). (d) The Yeast Test is an importanl confirmatory test for distinguishing the fermentable sugars from other re- ducing substances, but it is not very delicate, and will with difficulty detect sugar when less than 0-5 per cent, is present. It can be performed thus: A little yeast (the tablets of com- pressed yeast do very well) is added to a test-tube half filled with urine. The test-tube is then filled up with mercury, closed with the thumb, and inverted over a dish containing mer- cury. The dish may be placed on the top of a water-bath whose temperature is about 40° C. After twenty-four hours the sugar will have been broken up into alcohol and carbon dioxide. The latter will have collected above the mercury in the test-tube, and the former will be present in the urine. The tests for sugar will either be negative or will be less distinct than before. A control test-tube containing water and yeast should also be set up, as impurities in the yeast sometimes yield a small amount of carbon dioxide. Specially- constructed tubes are also often used for performing the test. (2) Quantitative Estimation of Sugar in Urine. — (a) Volumetrically, the sugar can be estimated by titration with Fehling's solution. As this does not keep well, two solutions containing its ingredients should be kept separately and mixed when required. Solution I.: Dissolve 34-64 grammes pure cupric sulphate in distilled water, and make up the volume to 500 c.c. Solution II.: Dissolve 173 grammes Rochelle salt in 400 c.c. of water, add to this 51-6 grammes sodium hydroxide, and make up the volume with water to 500 c.c. Keep in well-stoppered Fig. 196. — Phenyl-Glucosazone and Phenyl-Maltosazone Crystals (Mac- leod). The phenyl - glucosazone crystals are in the upper part of the figure, the phenyl-maltosazone in the lower. PRACTICAL EXERCISES 527 bottles in the dark. For use, mix together equal volumes of the two solutions. Ten c.c. of this mixture is reduced by 0-05 gramme dextrose. To estimate the sugar in urine, put 10 c.c. of the mixture into a porcelain capsule or glass flask, and dilute it four or five times with distilled water. Dilute some of the urine, say ten or twenty times, according to the quantity of sugar indicated by a rough determination. Run the diluted urine from a burette into the Fehling's solution, bringing it to the boil each time urine is added, until, on allowing the precipitate to settle, the blue colour is seen to have entirely disappeared from the supernatant liquid. The observation of the colour must be made while the liquid is still hot. Benedict's modification of Fehling's solution* may also be employed. Suppose that 10 c.c. of Fehling's solution is decolourized by 20 c.c. of the ten-times diluted urine. Then 2 c.c. of the original urine contains 0-5 gramme dextrose. If the urine of the twenty-four hours (from which this sample is assumed to have been taken) amounts to 4,000 c.c., the patient will have passed 0-05x2,000=100 grammes sugar, in twenty-four hours. (6) The polarimeter affords a rapid and, with practice, a delicate means of estimating the quantity of sugar in pure and colourless solu- tions, but diabetic urine must in general be first decolourized by adding lead acetate and filtering off the precipitate. What is measured is the amount by which the plane of polarization of a ray of polarized light of given wave-length (say sodium light) is rotated when it passes through a layer of the urine or other optically active solution of known thickness. Let a be the observed angle of rotation, / the length in decimetres of the tube containing the solution, w the number of grammes of the optically active substance per c.c. of solution, and (a)D the specific rotation of the substance for light of the wave-length of the part of the spectrum corresponding to the D line (i.e., the amount of rotation expressed in degrees which is produced by a layer of the substance I decimetre thick, when the solution contains i gramme of it per c.c.). Then (a)D = ± — -.. In this equation a and / are known from direct measurement; (a)D has been determined once for^all for most of the important active substances, and therefore w is easily calculated. For dextrose (a)D may be taken as 52-6°. It varies somewhat with the concentration, but for most investigations on the urine these variations may be neglected. It is not possible to describe here the numerous forms of polarimeter that are in use. Those constructed on what is called the ' half -shadow ' system (Fig. 197) give sufficiently satisfactory results. A half -shadow polarimeter consists, like other polarimeters, of a fixed Nicol's prism (the polarizer), and a nicol capable of rotation (the analyzer). In addition, there is an arrangement which rotates by a definite angle the plane of polarization in one -half of the field, but not in the other — e.g., a small nicol occupying only half of the field. In the zero position of the analyzer, both halves of the field are equally dark. The solution to be investigated is placed in a tube of known length, the ends of which * It contains 17" 3 grammes of cupric sulphate, I73'o grammes of sodium citrate, ioo'o grammes of anhydrous sodium carbonate made up with distilled water exactly to one litre. In making the solution the citrate and carbonate are dissolved with the aid of heat in about 600 c.c. ot water, and then made up to about 800 c.c. The cupric sulphate is dissolved in about too or 150 c.c. of water and added to the other solution, the whole being then made up to a litre. 5*8 EXCRETION are closed by glass discs secured by brass screw caps. The glass discs must be slid on, so as to exclude all air. The tube having been intro- duced between the polarizer and analyzer, the sharp vertical line which indicates the division between the two half-fields is focussed with the eye-piece, and then the analyzer is rotated till the two halves are again equally shadowed. The angle of rotation, a, is read off on the graduated arc, which is provided with a vernier. Pentoses reduce Fehling's solution, but do not give the yeast test. They give the following characteristic tests, which may be performed with gum arabic, a substance containing arabinose, one of the pentoses: (i) Phloroglucin Reaction. — Warm in a test-tube some pure concen- trated hydrochloric acid to which an equal volume of distilled water has been added. Add phloroglucin until a little remains undissolved. Fig. 197. — Mitscherlich's Polarimeter. (Half-shadow instrument.) (Simple form.) Add a small quantity of gum arabic, and keep the test-tube in a water- bath at 100° C. The solution becomes cherry-red, and a precipitate gradually separates, which may be dissolved in amyl alcohol. The solution shows with the spectroscope a band between D and E. (2) Or tin Reaction. — Use orcin instead of phloroglucin in (i). The solution becomes reddish-blue on warming, and shows a band between C and D, near D. The colour quickly changes from violet to blue, red, and finally green. A bluish-green precipitate separates, which is soluble in amyl alcohol. Glycuronic acid gives all the above reactions of pentoses. Bile-Salts (Hay's Test). — Put a little finely-divided sulphur, in the form of flowers of sulphur, on the top of a glass of urine. If bile-salts are present the sulphur will sink to the bottom. If there are no bile- PRACTICAL EXERCISES 5*9 Salts il will float on the top. The difference is due to an alteration in the surface tension of the urine produced by the bile-salts. We must exclude the presence of acetic acid, alcohol, ether, chloroform, turpen- tine, benzine and its derivatives, phenol and its derivatives, anilin and soaps, all of which also cause such an alteration in the surface tension of urine that the sulphur sinks to the bottom. The urine should be fresh, and if it has to be kept it should be preserved from decomposition by cyanide of mercury, which does not alter the surface tension. The reaction has the great advantage over other tests of being easily carried out at the bedside. Acetone — (i) Legal' s Test (Rothera's modification). — To 5 to 10 c.c. of the acetone-containing urine add enough ammonium sulphate crystals to form a layer at the bottom of the test-tube, then 2 or 3 drops cf a fresh 5 per cent, solution of sodium nitro-prusside and i to 2 c.c. of strong ammonia. The development of a colour like that of perman- ganate of potassium, often in the form of a ring a little above the undissolved salt, indicates the presence of acetone. The reaction must not be declared negative till half an hour has elapsed. The colour slowly fades. (2) Where there is doubt as to the presence of acetone, it is best first to distil it over. Put 250 to 500 c.c. of the urine suspected to contain acetone into a litre flask. Add a few c.c. of phosphoric acid; connect the flask with a worm, and distil over the urine into a small flask. For qualitative tests it is best to collect only the first 20 to 30 c.c., as most of the acetone is contained in this. Test the distillate for acetone by (i) or by Lieben's Test. — To a few c.c. of the distillate in a test-tube add a few drops of solution of iodine in potassium iodide, and then sodium or potassium hydroxide. A precipitate of yellow iodoform crystals (six- sided tables) is thrown down if acetone be present. Examine them under the microscope. On heating, the odour of iodoform may be recognized. If the precipitate is amorphous it may be dissolved in ether (free from alcohol), which is allowed to evaporate on a slide, when crystals may be obtained. Determination of the Freezing-Point of Urine.* — Study Beckmann's apparatus shown in Fig. 171, p. 427. Note the large thermometer D graduated in hundredths of a degree centigrade. It is inserted through a rubber cork into the inner test-tube A. A platinum wire, F, bent at the lower end into a circle or a spiral, which passes easily up and down between the bulb of the thermometer and the tube, serves to stir the urine. The thermometer must be so supported by the rubber cork that the bulb is in the axis of the tube and a centimetre or two from the bottom of it. The side-piece E on the tube A is not absolutely necessary, but it is convenient for ' inoculating ' the urine with a crystal of ice at the proper time. A passes through a rubber cork into a shorter and wider outer glass tube B. The space between A and B serves as a badly conducting mantle, which prevents too rapid cooling of the contents of A. B passes through a hole in the metal or wooden cover of a strong glass jar, C, which contains the freezing mixture. B should fit the hole so tightly that it does not bob up out of the mixture when A is removed. In C is a stirrear, G, of strong copper wire, the end of which passes through the lid. This serves to stir up the freezing mixture from time to time. Pulverize some ice by pounding it in a strong wooden box with a heavy piece of wood. Take the inner tube with the thermometer out of the apparatus. It is convenient to take the thermometer out of the * This is not often of clinical value, but it affords an opportunity for the student to practise a method of great importance in physiology. 34 530 EXCRETION tube, and to hang it up carefully on a stand by means of a fine flexible copper wire passing through the eye. The rubber cork can be taken out with the thermometer, and the platinum wire also, the bent free end of the latter supporting it in the cork, or it may be fastened tempor- arily to the thermometer stem by a small rubber band, which is slid up over the cork when the thermometer is reinserted. Tube A can be set temporarily in a specially heavy test-tube rack. Remove the lid of C, and with it tube B. Now put ice and salt alternately into C until it is nearly full, mixing them up well. Add some cold water from the tap till the stirrer G can move freely up and down in the mixture. For very exact work the temperature of the freezing mixture must not be more than a few degrees below the freezing-point of the liquid which is being examined. Put on the lid, and immerse tube B. Into A, which must be perfectly clean, put enough pure distilled water to fully cover the bulb of the thermometer, and introduce the latter. For ordinary purposes distilled water previously boiled to expel the carbon dioxide, and then cooled in a stoppered flask, is sufficiently pure. Immerse A directly in the freezing mixture through the hole by which G comes out, or through a separate hole (not shown in the figure) till some ice has formed in the water. Take A out of the mixture, wipe it with a cloth, and hold the lower part of it in the hand till nearly the whole of the ice has melted. If there is a cake of ice at the bottom, see that it is displaced by the platinum stirrer. A trace of ice being still left floating in the water, place A in B, and allow the temperature to fall to a few tenths of a degree below the freezing-point you expect to get, as deter- mined by a previous rough experiment. The freezing mixture is stirred up occasionally. The meniscus of the thermometer is to be carefully followed, as it goes on falling, by means of a weak hand lens. Now stir the water in A briskly. Suddenly it will be seen that the mercury begins to rise. Keep stirring with the platinum wire, and read off the maximum height of the mercury, at which it is stationary for some time. The temperature can be estimated between the gradu- ations to thousandths of a degree. Take out A, and observe the fine ice crystals in the water. Heat A in the hand as before till nearly all the ice has disappeared ; then replace A in B, and make another freezing- point determination. A third one may also be made, and the mean of the three readings taken. Take out the thermometer, and dry it and the platinum wire with clean filter-paper, or dip them in some of the urine, which is then thrown away. Dry A or rinse it with urine. Then make a determination of the freezing-point of the urine in the same way as was done with the water. The freezing-point of the urine will lie much lower on the scale. Instead of freezing the liquid first and then leaving a little ice in it when A is placed in B, A may be put into B before any ice has formed. Cooling is then allowed to go on with gentle stirring to a few tenths of a degree below the anticipated freezing-point. A small crystal of clean dry ice is then introduced through the side-piece on a clean splinter of wood or the loop of a cooled platinum wire, the end of which passes through a piece of cork, by which it is held to prevent conduction of heat. The platinum stirrer can be raised to receive the crystal. The liquid is now vigorously stirred ; freezing occurs, and the observation is made as before. Instead of the above method, the liquid may first be cooled directly in the freezing mixture, but not so much that ice forms. A is then put in B, and cooling allowed to go on while it is being stirred. When it has been undercooled to a certain extent — i.e., cooled below its freezing- point — the vigour of the stirring is increased. Ice forms suddenly, as before, and the temperature rises to the freezing-point. With urine PRACTICAL EXERCISES 531 this method is sufficiently satisfactory, but it is not usually easy to get freezing of the distilled water till the undercooling is considerable, and it has been shown that this introduces some error. Suppose the freezing-point of the distilled water on the scale of the thermometer was 5-245 and that of the urine 3-625°, the value of A for the urine is 1-620°. Since for most purposes it is sufficient to fix the second decimal point, much smaller and less expensive thermometers than the ordinary Beckmann may be employed. In the same way the freezing-point of blood-serum (or blood), bile, and other physiological liquids can be determined. Systematic Examination of Urine. — In examining urine, it is con- venient to adopt a regular plan, so as to avoid the risk of overlooking anything of importance. The following simple scheme may serve as an example; but no routine should be slavishly followed, the object being to get at the important facts with the minimum of labour. More extensive information must be sought in the treatises on examination of the urine for clinical purposes. 1. Anything peculiar in colour or smell ? If the colour suggests blood, examine with spectroscope, haemin test, guaiacum test (pp. 76, 272) ; if it suggests bile, test for bile-pigments by Gmelin's test (p. 462), and for bile-salts by Pettenkofer's test (p. 462) and by Hay's test (pp. 462, 528). 2. Reaction. 3. Sediment or not ? Sediment may be procured by letting the urine stand in a conical glass, or in a few minutes by the centrifuge. .If the appearance of the sediment suggests anything more than a little mucus, examine with the microscope. The sediment may contain organized or unorganized deposits. Organized Sediments. — (a) Red blood-corpuscles (considerably altered if they have come from the upper part of the urinary tract). (6) Leucocytes. A few are present in health. A large number indicates pus. When pus is present the sediment may be white to the naked eye. (c) Epithelium from the bladder, ureters, pelvis of the kidney or the renal tubules. A few squamous epithelial cells from the urethra are always present in normal urine. (d) Tube casts. (e) Spermatozoa (occasional). (/) Bacteria. (g) Parasites (rare). (h) Portions of tumours (rare). Unorganized Sediments. IN ACID URINE. Uric Acid. — Crystals coloured brownish - yellow with urinary pigment. Various shapes, espe- cially oval ' whetstones,' rhom- bic tables, and elongated crystals, often in bundles (Fig. 177). Urates. — Usually amorphous, in the form of fine granules, often tinged with urinary pigment, sometimes brick-red. Soluble on heating. On addition of acids (including acetic acid) they dis- IN ALKALINE URINE. Triple Phosphate. — Clear, col- ourless, coffin - lid or knife - rest crystals. Also deposited in the form of feathery stars (Fig. 179). Calcium Hydrogen Phosphate (' stellar ' phosphate), CaHPO4. — Crystals often wedge-shaped and arranged in rosettes. May also occur in a dumb-bell form. (A phosphate of calcium is also occa- sionally seen in weakly acid urine.) (Fig. 1 8 1, p. 479.) 532 EXCRETION Unorganized Sediments (continued) — > IN ACID URINE. solve and uric acid cystals appear in their place. Acid urate of sodium and of ammonium occa- sionally found in the crystalline form (rosettes of needles) . Calcium Oxalate. — Octahedral, ' envelope ' crystals, not coloured. Insoluble in acetic acid. Soluble in hydrochloric acid (Fig. 178, p. 478). Cystin. — Hexagonal plates. Rare (Fig. i8o,p. 4 ;•.;). Leucin and Ty rosin (Figs. 186, 187, p. 488). — Rare. Also found in alkaline urine, but rarely. Triple Phosphate. — Sometimes found in weakly acid urine. IN ALKALINE URINE. Calcium Phosphate, Ca3(PO4)2 •— Amorphous. Magnesium Phosphate . — Long rhombic tablets, which are dis- solved at the edges by ammonium carbonate solution, unlike triple phosphate. All the above are soluble in acetic acid without effervescence. Calcium Carbonate. — Small spherical or dumb - bell - shaped bodies soluble in acetic acid with effervescence. Ammonium Urate. — Dark balls, often covered with spines. Soluble in acetic or hydrochloric acid, with formation of uric acid crys- tals (Fig. 182, p. 479). 4. Specific gravity. 5. Quantity of urine in twenty-four hours. If the quantity is abnormally large and the specific gravity high, test for sugar. 6. Inorganic constituents not generally of clinical importance, but in special diseases they should be examined — e.g., chlorides in pneu- monia. 7. Normal organic constituents. Sometimes quantitative estima- tion of urea or total nitrogen in fever, and in diabetes and Bright's disease. 8. Chemical examination for abnormal organic constituents, especi- ally albumin and sugar. Albumin. — (i) Heat to boiling some of the urine in a test-tube. A precipitate insoluble on addition of a few drops of acetic acid consists of coagulable protein. A precipitate soluble in acetic acid consists of earthy phosphates. (2) Heller's test. Put some strong nitric acid in a test-tube and run on to it some urine. A white ring indicates protein. A very rough quantitative estimation may be made by the method of Esbach (p. 525). Sugar. — (i) Trommer's test. (Fehling's solution may be used.) I* the result is indecisive — (2) Phenyl-hydrazine test (p. 525)- (3) In case of doubt confirm by yeast test. A quantitative estimation may be made with Fehling's solution or the polarimeter. CHAPTER X METABOLISM, NUTRITION AND DIETETICS WE return now to the products of digestion as they are absorbed from the alimentary canal, and, still assuming a typical diet con- taining carbo-hydrates, fats, and proteins, we have to ask, What is the fate of each of these classes of proximate principles in the body ? what does each contribute to the ensemble of vital activity ? It will be best, first of all, to give to these questions what roughly qualitative answer is possible, then to look at metabolism in its quantitative relations, and lastly to focus our information upon some of the practical problems of dietetics. SECTION I. — METABOLISM OF CARBO-HYDRATES — GLYCOGEN. The carbo-hydrates of the food, passing into the blood of the portal vein in the form of dextrose, are in part arrested in the liver, and stored up as glycogen in the hepatic cells, to be gradually given out again as sugar in the intervals of digestion. The proof of this statement is as follows: Sugar is arrested in the liver, for during digestion, especially of a meal rich in carbo-hydrates, the blood of the portal contains more sugar than that of the hepatic vein. Popielski, on the basis of experiments in which he fed with known quantities of sugar dogs whose inferior vena cava and portal vein had been united by an Eck's fistula, and determined the amount of sugar which passed into the urine, estimates the quantity of sugar kept back by the liver at from 12 to 20 per cent, of the whole. In the liver there exists a store of sugar-producing material from which sugar is gradually given off to the blood, for in the intervals of digestion the blood of the hepatic veins contains more dextrose (2 parts per 1,000) than the mixed blood of the body or than that of the portal vein (about i part per 1,000). When the circulation through the liver is cut off in the goose, the blood rapidly becomes free, or nearly free, from sugar (Minkowski). And a similar result follows such inter- ference with the hepatic circulation as is caused by the ligation of the three chief arteries of the intestine in the dog, even when the animal has been previously made diabetic by excision of the pancreas (p. 636). 533- 534 METABOLISM, NUTRITION AND DIETETICS The nature of the sugar-forming substance is made clear by the following experiments: (i) A rabbit after a large carbo-hydrate meal, of carrots for instance, is killed and its liver rapidly excised, cut into small pieces, and thrown into acidulated boiling water. After being boiled for a few minutes, the pieces of liver are rubbed up in a mortar and again boiled in the same water. The opalescent aqueous extract is filtered off from the coagulated proteins. No sugar, or only traces of it, are found in this extract; but another carbo-hydrate, glycogen, a polysaccharide giving a port-wine colour with iodine and capable of ready conversion into sugar by amylolytic ferments, is present in large amount. (See Practical Exercises, p. 715.) (2) The liver after the death of the animal is left for a time in situ, or, if excised, is kept at a temperature of 35° to 40° C, or for a longer period at a lower temperature; it is then treated exactly as before, but no glycogen, or comparatively little, can now be obtained from it, although sugar (dextrose) is abundant. The inference plainly is that after death the hepatic glycogen is con- verted into dextrose by some influence which is restrained or de- stroyed by boiling. This transformation might theoretically be due to an unformed ferment or to the direct action of the liver-cells, for both unformed ferments and living tissue elements are destroyed at the temperature of boiling water. It ha.- been clearly shown that the action is brought about by a diastatic enzyme, which some writers call glycogenase, for it readily occurs when the minced liver is mixed with chloroform water, and chloroform kills all living tissues. Although blood contains a diastase in small amount, the change does not depend essentially upon this, since the glycogen also undergoes hydrolysis (glycogenolysis) to dextrose when all the blood has been washed out of the organ. Lymph also contains a diastase, but there is evidence that the post-mortem glycogenolysis is chiefly due to an enzyme contained in the hepatic cells (an endo- enzyme) (Macleod). The diastases in the blood and lymph seem to be ' discards ' of the tissues which are on the way to destruction or elimination (Carlson). The post-mortem change is to be regarded as an index of a similar action which goes on during life: sugar in the intact body is changed into glycogen ; glycogen is constantly being changed into sugar. There is no reason to doubt that here, too, the hydrolysis is effected by the endo-enzyme. It might be supposed, indeed, that the adjustment of the two processes glyco- gcnesis and glycogenolysis is simply a matter of the alteration of the equilibrium in a reversible reaction (p. 338), according to whether the dextrose content of the blood tends to rise or fall. If the concentration of dextrose in the blood is increased, more dex- trose might be expected to ' diffuse ' into the hepatic cells, whose content of dextrose in proportion to glycogen would increase till the equilibrium was restored by the conversion of the excess of METABOLISM OF CARBO-HYDRATES— GLYCOGEN 533 sugar into glycogen. Contrariwise, a diminution in the dextrose content of the blood might be expected to lead to diffusion of dextrose out of the liver-cells, and a consequent acceleration of the hydrolysis of the glycogen. We have already learnt, however, that in physiology — above all, perhaps, in the physiology of the glands — ' simple ' explanations are usually suspect. And when we come to study those conditions in which, as a consequence of the derange- ment of the mechanisms which regulate the carbo-hydrate metabo- lism, sugar appears in the urine, it will be seen that the matter is more complicated. For one thing, the nervous system seems to take a hand in the regulation, and where the nervous system takes a hand things are generally doing which the experienced physiologist does not expect to be simple. We may be certain, as in the case of the intracellular proteolytic ferments, that the vital action of the hepatic cells is a most important factor in controlling the rate of production of the ferment, and therefore its concentration in rela- tion to that of the substrate and the rate at which it works. (3) With the microscope, glycogeri, or at least a substance which is very nearly akin to it, which very readily yields it, and which gives the characteristic port-wine colour with iodine, can be actually seen in the liver-cells. The liver of a rabbit or dog which has been fed on a diet containing much carbo-hydrate is large, soft, and very easily torn. Its large size is due to the loading of the cells with a hyaline material, which gives the iodine reaction of glycogen, and is dissolved out by water, leaving empty spaces in a network of cell- substance. If the animal, after a period of starvation, has been fed on protein alone, less glycogen is found in the shrunken liver- cells; if the diet has been wholly fatty, little or no glycogen at all may be found. Glycogen can even be formed by an excised liver when blood containing dextrose is circulated through it. A fact of great interest recently demonstrated is that in animals where the mobilization of the hepatic glycogeri is accelerated (as in rabbits after puncture of the medulla), glycogen in the form of granules and small masses can be seen in the blood capillaries (or rather sinusoids) between the liver cells, and in the sublobular veins (Huber and Macleod). Chemical evidence has also been obtained of the presence of a glycogen-like polysaccharide in the blood leaving the liver. Formation of Glycogen from Protein. — In the liver-cells of the frog in winter-time a great deal of this hyaline material — this glycogen, or perhaps loose glycogen compound — is present; in summer, much less. The difference is remarkable if we con- sider that in winter frogs have no food for months, while summer is their feeding-time; and at first it seems inconsistent with the doctrine that the hepatic glycogen is a store laid up from surplus sugar, which might otherwise be swept into the general circulation and excreted by the kidneys. It has been found, however, that 536 METABOLISM, NUTRITION AND DIETETICS the quantity of glycogen is greatest in autumn at the beginning of the winter-sleep, and slowly diminishes as the winter passes on, to fall abruptly with the renewal of the activity of the animal in the spring. The glycogen present at any moment is, therefore, believed to be a residue, which represents the excess of glycogen formed over glycogen used up; and the amount is larger in winter, not because more is manufactured than in summer, but because less is consumed. It is possible, indeed, to produce the ' summer ' condition of the hepatic cells merely by raising the temperature of the air in which a winter frog lives ; at 20° or 25° C. glycogen dis- appears from its liver. Conversely, if a summer frog is artificially cooled, a certain amount of glycogen accumulates in the liver. The meaning of this seems to be that at a low temperature, when the wheels of life are clogged and metabolism is slow, some substance, probably dextrose, is produced in the body from proteins in greater amount than can be used up, and that the surplus is stored as glycogen; just as in plants starch is put by as a reserve which can be drawn upon — which can be converted into sugar — when the need arises. That carbo-hydrates may be formed from proteins (or their constituent amino-acids) has been shown in various ways — for example, by feeding dogs with almost pure protein (casein) after the production of permanent glycosuria by removal of the pancreas (p. 636). To induce the animal to take the casein it had to be mixed with a certain amount of butter, or serum, or meat extract. The amount of sugar excreted was much more than could possibly have come from the glycogen originally present in the animal's body, computing it on the most generous scale (41 grammes per kilo- gramme of body-weight, according to Pfliiger), or from free carbo- hydrate present in traces in the food, or as prosthetic groups (p. 2) in the ingested protein. That the source of the sugar was protein and not fat was indicated by the fact that when the amount of protein food was increased, the dextrose and the nitrogen excreted increased proportionally (see also p. 538)- Glycogen-Formers. — As true glycogen-formeis in the higher animals — that is, compounds whose elements (particularly the carbon) actually enter into the composition of the glycogen mole- cule— may be mentioned such substances as proteins (including gelatin), the fermentable sugars, and glycerin. In the case of proteins it is, of course, not the entire molecule which is transformed bodily into glycogen, but amino-acids yielded by them, or dextrose derived from the amino-acids. The liver is of itself capable of dealing only with the dextrose, and not with the amino-acids. At least, when the isolated liver (of the tortoise) is perfused with blood containing amino-acids no increase in the glycogen of the liver occurs. When glycerin is added to the blood the glycogen content of the liver is very distinctly increased. Glycerin is a tri- valent alcohol (C3H8O3) whose aldehyde, obtained from it by gentle METABOLISM OF CARBO-HYDRATES— GLYCOGEN 537 oxidation, is glycerose (C3H6O3), a substance with the typical properties of a sugar. In the laboratory it has been shown that two molecules of glycerose can be combined to form one molecule of sugar of the hexose type with six carbon atoms (C6H12O6) . A similar transformation is accomplished in the liver, and then a number of the monosaccharide molecules (C6H12O6) are condensed with loss of water to form glycogen. Thus, w(C6H12O6) — «H2O = (C6H10O5)w. Since glycerin is a normal product of the hydrolysis of fats, the possibility that the fats of the food may contribute through their glycerin component to glycogen formation must be admitted. The monosaccharides dextrose, levulose, and galactose gave a similar result, while the disaccharides cane-sugar and lactose caused no increase in the glycogen of the perfused liver, since the liver contains no ferment capable of splitting them into monosaccharides. And although the first step in the linking of the monosaccharide mole- cules would seem to be the formation of a disaccharide such as maltose, the glycogen molecule must apparently be built up from single ' bricks,' the monosaccharides, and cannot be constructed from bricks which are already coupled in pairs, the disaccharides. Of course, since the disaccharides are hydrolysed in the digestive tube to simple sugars, they are to be reckoned with the true glycogen- formers, for in the intact body they are presented to the hepatic cells in the form of monosaccharides. It is probable that levulose and galactose are first changed into dextrose. By the action of alkalies such structurally related sugars can easily be transformed into each other. Thus dextrose is an aldehyde of an alcohol with six carbon atoms, and levulose the corresponding keto- hexose. By oxidizing the alcohol we get an aldehyde or a ketone, according to whether a primary alcohol group (CHa.OH) or a secondary group (CH.OH) is oxidized, with the loss of two atoms of hydrogen. The aldehyde is characterized by the presence of the group £\v> the ketone by the group CO. Both the aldehyde and the ketone are sugars, and since each contains six carbon atoms, they are both sugars of the group known as ' hexoses.' Dextrose, being not only a hexose but an aldehyde, may be called an ' aldohexose,' and levulose, being not only a hexose but a ketone, a ' ketohexose.' CHa.OH C/2 CHa.OH I lx I CH.OH CH.OH CO CH.( CH.OH CH.OH CH.OH CH.OH CH.OH CH.OH CH.OH CH.OH CH.OH CHa.OH CH2.OH CH2.OH 6-valent alcohol. Aldehyde. Ketone, 538 METABOLISM, NUTRITION AND DIETETICS That levulose can be changed into dextrose in the body is indi- cated by the observation that after extirpation of the pancreas in dogs the administration of levulose is followed by an increase in the excretion of dextrose nearly equal to the amount of levulose ingested. It is also stated that, when the surviving liver of a normal dog is perfused with a suspension of washed blood-corpuscles to which levulose has been added, dextrose accumulates in the blood and levulose disappears from it. It has not hitherto been proved that the fatty acid component of the food fats can be converted into glycogen. But a fatty acid, propionic acid, is capable of complete transformation into dextrose when given either by the mouth or subcutaneously to dogs under the influence of phlorhizin (Ringer) . Many other bodies are known to influence the formation of glycogen by ' sparing ' substances which are true glycogen-producers, but their carbon does not actu- ally take its place in the glycogen molecule. It has been shown that proteins can directly form glycogen or sugar apart from carbo- hydrate groups contained in the protein molecule. For the proteins of meat, gelatin, and casein are capable of forming 60 per cent, of their own weight of dextrose in diabetic metabolism, and even the end-products of pancreatic digestion of meat yield so much sugar that the greater part of it must have come from the amino-bodies, and not from a sugar-group in the protein. When given to dogs with total phlorhizin glycosuria (p. 551), glycin and alanin are completely, glutamic and aspartic acids in great part (corresponding to about three carbon atoms of their respective molecules), converted into dextrose (Lusk and Ringer), and there is no reason to doubt that when such substances are produced by hydrolysis of protein in the normal body, and are not all utilized in rebuilding the bio- plasm, a portion of the surplus, after deamidization, can be trans- formed into glycogen. Extra-Hepatic Glycogen. — While the liver in the adult (con- taining as it does from 2 to 10 per cent, of glycogen, or even, with a diet rich in sugar or starch, more than 18 per cent.) may be looked upon as the main storehouse of surplus carbo-hydrate, depots of glycogen are formed, both in adult and fcetal life, in other situations where the strain of function or of growth is exceptionally heavy — in the muscles of the adult (0-3 to 0-5 per cent, of the moist skeletal muscle, or on a carbo-hydrate regimen 0-7 to 3-7 per cent.), in the placenta, in many developing organs in the embryo (muscles, lungs, epithelium of the trachea, oesophagus, intestine, ureter, pelvis of kidney, and renal tubules). The foetus, however, is not, compared with the adult, especially rich in glycogen. In the adult under favourable circumstances the absolute amount of glycogen in the muscles may be several times greater than that in the liver, and METABOLISM OF CARBO-HYDRATES— GLYCOGEN 539 usually the hepatic glycogen makes up considerably less than half the total glycogen of the body. That the muscles do not derive their glycogen by the migration of hepatic glycogen, but can them- selves form it from dextrose, has been shown by injecting that sugar subcutaneously into frogs after excision of the liver. The muscle glycogen was found to be increased. Function and Fate of the Glycogen. — The glycogen store of the liver fulfils a different function from that of the muscles. This is indicated by the fact that when dogs, after being put on a given diet for two or three days, are starved for a time, and then put again on the original diet, the hepatic and the muscular glycogen behave differently at first during the period of re-alimentation. While glycogen accumulates in the liver in greater quantity than under normal conditions of nutrition, in the muscles it at first accumulates much less rapidly than normally. This is entirely in accordance with the view that the hepatic glycogen store has for its great func- tion the regulation of the sugar content of the blood in the interest of all the tissues, while the glycogen store of the muscles and other tissues is mainly in the interest of their own nutrition and a source of energy for their own work. This does not imply that, when sugar is being absorbed in quantities too great for the liver to deal with after the current needs of the tissues have been satisfied, they do not add to their glycogen reserves. There is every reason to suppose that they do so, and thus act as a subsidiary regulating mechanism, although a less elastic one than that supplied by the liver. A third way in which a portion of the surplus sugar can be stored is in the form of fat. When a fasting dog is made to do severe muscular work, the greater part of the glycogen soon disappears from its liver. When a dog is starved, but allowed to remain at rest, the glycogen still markedly diminishes, although it takes a longer time; and at a period when there is still plenty of fat in the body, there may be only a trace of hepatic glycogen left. The glycogen which is usually contained in the skeletal muscles also diminishes very rapidly in the first days of hunger, but the heart contains the normal amount of glycogen at a time when the proportion in the skeletal muscles has sunk to Jfr to -£$ of the normal. These facts have been taken to indicate that glycogen and the sugar formed from it are the readiest resources of the starving and working organism, for the transformation of chemical energy into heat and mechanical work. To borrow a financial simile, the fat of the body has sometimes been compared to a good, but rather inactive security, which can only be gradually realized; its organ-proteins to long-date bills, which will be discounted sparingly and almost with a grudge; its glycogen, its carbo-hydrate reserves, to consols, which can be turned into money at an hour's warning. Glycogen, on this view, is especiallj 54° METABOLISM, NUTRITION AND DIETETICS drawn upon for a sudden demand, fat for a steady drain, tissue-protein for a life-and-death struggle. While there may be some such difference in the tenacity with which the different kinds of reserve material are held back from consumption when the floating supplies are wearing low, modern investigation tends to the conclusion that the interchange- ability of the various groups of nutritive substances is greater than had been supposed, and that in the long-run the cells — in normal cir- cumstances at least — never work without dextrose, even after the glycogen store has been practically all consumed, but secure it from other sources. Pavy has put forward the heterodox view that the glycogen formed in the liver from the sugar of the portal blood is never reconverted into sugar under normal conditions, but is changed into some other substance or substances, and he denies that the post-mortem formation of sugar in the hepatic tissue is a true picture of what takes place during life. But in spite of the brilliant manner in which he has defended this thesis, both by argument and by experiment, it must be said that the older doctrine of Bernard, which in the main we have followed above, is attested by such a cloud of modem witnesses that it seems to be firmly and finally established. Fate of the Sugar — Glycolysis. — What, now, is the fate of the sugar which either passes right through the portal circulation from the intestine without undergoing any change in the liver, or is gradually produced from the hepatic glycogen ? When the pro- portion of sugar in the blood rises above a certain low limit (about 1-5 or 2 parts per 1,000), some of it is excreted by the kidneys (Practical Exercises, p. 716). A large meal of carbo-hydrates is frequently followed by a temporary glycosuria, but much depends upon the form in which the sugar-forming material is taken. We have seen that poly- saccharides are quite incapable of absorption as such, and that they must be very completely hydrolysed in the lumen of the alimentary canal before their constituent sugars have any chance of reaching the blood. It is therefore not to be expected that the rapid absorption of such considerable quantities of sugar as would lead to its excretion should easily occur when the carbo-hydrate is in this form. Miura for example, after an enormous meal of rice (equivalent to 6-4 grammes of ash- and water-free starch per kilo of body- weight), which, as he mentions, tasked even his Japanese powers of digestion for such food to dispose of, found not a trace of sugar in the urine. Dextrose, cane-sugar and lactose, on the other hand, when taken in large amount, were in part excreted by the kidneys, as was also the case with levulose and maltose in a dog (Practical Exercises, p. 717).* The amount of any carbo-hydrate which can be eaten * Twenty-four healthy students, whose urine had previously been shown to be free from sugar, ate quantities of cane-sugar varying from 250 grammes to 750 grammes. The urine was collected in separate portions for twelve to twenty-four hours after the meal. In only three cases was reducing sugar found in the urine (by Fehling's and the phenyl-hydrazine test), and then merely in traces. In eight cases cane-sugar was found, and estimated by the polarimeter, and, after boiling with hydrochloric acid, by Fehling's solution. METABOLISM OP CARBO-HYDRATES 54! without the appearance of sugar in the urine is sometimes called the assimilation limit for that carbohydrate. The attsmpt has been made to fix the limit of tolerance of dextrose for normal persons and persons suffering from incipient diabetes, with the object of aiding in early diagnosis of that condition. But the limit varies with so many conditions, only some of which can be controlled, that such observations are not easily interpreted. Except as an occasional phenomenon, glycosuria other than ali- mentary is inconsistent with health; and therefore in the normal body the sugar of the blood must be either destroyed or transformed into some more or less permanent constituent of the tissues. The transformation of sugar into fat we have already mentioned, and shall have again to discuss; it only takes place under certain con- ditions of diet, and no more than a small proportion of the sugar which disappears from the body in twenty-four hours can ever, in the most favourable circumstances, be converted into fat. The dextrose which is taken up from the blood by the tissues and there condensed to glycogen suffers sooner or later the converse change, in all probability under the influence of diastases or glycogenases produced in the cells, and reappears as dextrose to take its place in the cellular katabolism and begin the series of cleavages and oxidations by which its chemical energy is set free. Accordingly, it is the destruction of sugar which concerns us here, and there is every reason to believe that this takes place, not in any particular organ, but in all active tissues, especially in the muscles, and to a less extent in glands. It has been asserted that the blood which leaves even a resting muscle, or an inactive salivary gland, is poorer in sugar than that coming to it ; and the conclusion has been drawn that in the metabo- lism of resting muscle and gland sugar is oxidized, the carbon passing off as carbon dioxide in the venous blood. This is indeed extremely likely, for we know that, when the skeletal muscles of a rabbit or guinea-pig are cut off. from the central nervous system by curara, the production of carbon dioxide falls much below that of an intact animal at rest ; and the carbon given off by such an animal on its ordinary vegetable diet can be shown, by a comparison of the chemical composition of the food and the excreta, to come largely from carbo-hydrates. But, considering the relatively feeble metabolism of muscles and glands when not functionally excited, the large volume of blood which passes through them, the difficulty of determining small differences in the proportion of sugar in such a liquid, the possibility that even in the blood itself sugar may be The greatest quantity of cane-sugar recovered from the urine was 8 grammes (7.92 grammes by Fehling's method and 8-29 grammes by the polarimeter) ; the highest proportion of the quantity taken which appeared in the urine was 2-5 per cent. When dextrose was found, cane-sugar was always present as well. 54* METABOLISM, NUTRITION AND DIETETICS destroyed, or that it may pass from the blood, without being oxi- dized, into the lymph, too much weight may be easily given to the results of direct analysis of the in-coming and out-going blood. And although the results of Chauveau and Kaufmann, obtained in this way, fit in fairly well with what we have already learnt by less direct, but more trustworthy, methods, such as the study of the respiratory exchange, they cannot be accepted as yielding exact quantitative information. They found that in one of the muscles of the upper lip of the horse the quantity of dextrose used up during activity (feeding movements) was 3-5 times as much as in the same muscle at rest, and this corresponded with the deficit of oxygen in the blood entering the muscle, and with the excess. of carbon dioxide in the blood leaving it. More dextrose was also destroyed in the active than in the passive parotid gland of the horse, but the excess per unit of weight of the organ was far less than in muscle. In dogs whose abdominal viscera have been removed, so that they constitute practically preparations composed of skeletal muscles it has been found that the amount of dextrose which disappears from 100 grammes of blood per minute varies from 0-47 to 1-8 milligrammes, the irregu- larities probably depending largely upon the irregular consumption by the muscles of the glycogen stored in them (Macleod and Pearce). Intermediary Metabolism of Carbo- Hydrates. — Concerning the pro- cesses and the stages by which dextrose is destroyed in the tissues, we have no very exact information, and it cannot be definitely stated at present what share is taken by cleavage and what by oxidation, or rather through what intermediate products, formed, it may be, now by simple cleavage, now by oxidation, again by a combination of cleavage and oxidation, the dextrose molecule is finally resolved into carbon dioxide and water. It must be remembered that the synthetic powers of animal cells are now known to be very extensive. They build carbo-hydrates, fats, phosphatides, and proteins, as well as destroy them, and at any of the earlier stages at any rate the degradation products of dextrose, or some of them, may be utilized in the construction of new compounds — for example, of fat — either in the cells where they arise or elsewhere in the body. In like manner the decomposition of a molecule of dextrose begun in one cell or in one tissue may be consummated in another to which intermediate products are conveyed in the blood. In such ways it is obvious that the katabolic processes may be finely regulated both qualitatively and quantitatively in accordance with the specific wants of different organs and the intensity of their func- tional activity from time to time. It must be said, however, that at present there are few definitely ascertained facts to guide us in trying to form a scheme of the actual changes which occur in the intermediate katabolism of carbo-hydrates, and the sequence which METABOLISM OF CARBO-HYDRATES 543 they normally follow. Glycuronic acid has been previously men- tioned as a substance occurring even in normal urine in small amount. It is very closely related to dextrose, as a comparison of their constitutional formulae shows: COOH H— C— OH OH— C— H I H— C— OH OH i2OH d-dextrose. CHaOH rf-glyconic acid. COOH rf-glycuronic acid. Glycuronic acid agrees with dextrose in containing the character- istic aldehyde group CTT, but differs in that by oxidation two atoms of hydrogen in the primary alcohol group CH2OH have been replaced by one atom of oxygen. There is reason to believe that in the tissues glycuronic acid can be formed from dextrose in the same way, possibly through the mediation of an enzyme, and it may therefore represent a stage in the katabolism of sugar. But it is not known whether this is a normal transformation through which the whole or the greater part of the dextrose passes, or only a transformation involving a small part of the sugar under normal conditions. The appearance in the urine of large quantities of glycuronic acid, paired as already explained with , various com- pounds, in certain pathological states or after the administration of certain drugs (p. 482) , might be explained either as the result of an increased production of that substance through a deflection of the normal trend of carbo-hydrate degradation, or as the result of a failure on the part of the cells to further transform the glycuronic acid in the quantities normally produced. Lactic acid is the one intermediate stage in the decomposition of dextrose in the tissues whose importance seems to be definitely ascertained. The muscles and the liver have been proved to possess the power of producing lactic acid from dextrose obtained directly from the blood or from the hydrolysis of their own store of glycogen, and there is little doubt that this power is shared by many, perhaps by all, of the other organs. There is also good evidence that the lactic acid thus formed can be, and under normal conditions is, in large part oxidized so as eventually to yield carbon dioxide and water, although there is reason to believe that a portion of it may be utilized for the synthesis of more complex bodies. The chemistry of the change or series of changes by which lactic add is produced from dextrose and the end-products, carbon dioxide 544 METABOLISM, NUTRITION AND DIETETICS and water, from lactic acid has given rise to much discussion, and is not yet clearly known. The following scheme, based on the researches of Embden and others, and quoted from Abderhalden, illustrates one suggestion as to the course of the series of transformations, although it must be taken only as a diagram of the sequence of some of the possible stages. A molecule of dextrose is represented as giving rise to two molecules of glyceric aldehyde, each of which then yields a mole- cule of lactic acid. Each molecule of lactic acid, losing two atoms of hydrogen, becomes converted into a molecule of pyruvic acid, which by the loss of the elements constituting a molecule of carbon dioxide becomes acetaldehyde or acetic aldehyde, and this by oxidation acetic acid, which is then oxidized to carbon dioxide and water. Thus — SH H— C— OH HO— C— H H— C— OH H— C— OH CH2OH ({•dextrose. COOH CO -CO, — H— C— OH CH2OH \ H— C— OH CH2OH 2 molecules gjyceric aldehyde. COOH H— C— OH -H2 CH, d-lactic acid. c/-o U +O CH3 Pyruvic acid. CH3 Acetaldehyde. COOH Acetic acid. 2COa -> + 2HaO Carbon dioxide and water. It has been shown that acetaldehyde and carbon dioxide are formed from pyruvic acid by the action of a ferment contained in yeast, and there is some evidence that a similar transformation may occur in the liver. It is to be particularly remarked that according to this scheme the whole of the dextrose molecule is still represented in the lactic acid formed from it. Up to this stage no part of the molecule has been burnt. Nearly the whole of the chemical energy — i.e., all but about 3 per cent, of it — is still available. For a gramme of lactic acid yields 3,661, and a gramme of dextrose 3,762, small calories on complete combustion. The intermediate products of the decom- position may therefore be transported from the place of origin and utilized elsewhere with scarcely any loss of energy. Further, it is indicated in the scheme that the degradation process is not merely a series of cleavages and oxidations, but that these may be inter- spersed with stages of reduction. It is also clearly suggested that at certain points the metabolism may become recessive and syn- METABOLISM OF CARBO-HYDRATES 545 theses be started, which may go far to retrace the steps of the pre- ceding katabolism in respect to a portion of the dextrose. Thus, lactic acid can be retransformed into dextrose (Mandel and Lusk), and this, of course, into glycogen. Pyruvic acid can be changed into lactic acid by reduction, and dextrose can in this way be again produced. There is evidence also that under certain conditions pyruvic acid can yield dextrose in the organism by a different reaction, being changed into acetaldehyde, which can then undergo transformations leading back to dextrose (Ringer). The formation of fat from sugar may also start from some of the stages displayed in the scheme, for it is only a short step to obtain by the reduction of glyceric aldehyde its alcohol glycerin. And from acetaldehyde fatty acids can be derived. Not only does lactic acid afford a point of contact between the metabolism of carbo-hydrates and that of fats — a junction, so to speak, where these two great metabolic currents cross each other, and where material originating in the one may be shunted into the other — but it also affords a point of junction and interchange with the current of protein metabolism. For example, the amino-acid, alanin, yields as a decomposition product a compound called methylglyoxal (CH3.CO.CHO), a ketonic aldehyde, which by the assumption of a molecule of water can be changed into lactic acid. It may also be one of the intermediate stages in the decomposition of dextrose as a precursor of lactic acid, and one of the ways in which the conversion of amino-acids into dextrose is accomplished may be through this link. The presence of a ferment glyoxalase, or probably more than one ferment which rapidly changes methyl- glyoxal into lactic acid, has been demonstrated in tissue extracts and in leucocytes. The same change is effected when blood con- taining methylglyoxal is perfused through an excised surviving liver. The reaction can be reversed for methylglyoxal like lactic acid when given to an animal rendered diabetic by phlorhizin can be shown to yield dextrose, possibly being first converted into glyceric aldehyde. The conversion of methylglyoxal into lactic acid is also a reversible reaction, for in vitro, at any rate, lactic acid readily yields methylglyoxal (Dakin). Pyruvic acid is another possible link. As has just been mentioned, it probably forms a stage in the decomposition of dextrose, and has, in addition, chemical relations on the one hand to certain of the amino- acids, especially to alanin, and on the other to glycerin and even to fatty acids. Thus — H.NH2 + COOH Alanin (a-amino-pro- pionic acid. CH3 « CO 4-NHa COOH Pyruvic acid. CHo.OH * CH CH.OH + 2O = CO + aHJD CH2.OH Glycerin. COOH Pyruvic acid. 35 546 METABOLISM, NUTRITION AND DIETETICS The following scheme, doubtless incomplete, illustrates a probable chemical sequence through which the interconversion of alanin, lactic acid, methylglyoxal and dextrose may be brought about by reactions only involving the addition or removal of water or ammonia (Dakin) : Dextrose Lactic acid Zl Methylglyoxal Z± Alanin (CH3.CHOH.COOH) (CH3.CO.CHO) (CH3.CH.NH2.COOH) Since pyruvic acid can be reduced to lactic acid, any reaction in which it plays a part in the intermediary metabolism of carbohydrates or proteins can also be fitted into the scheme. The more completely the various steps in the metabolism of the three great groups of food substances are unveiled, the more clearly does it appear that, far from being independent circuits, the three currents are constantly exchanging materials with each other. It is to be supposed that in many of these transformations enzymes are concerned, although comparatively little is definitely known as to this. Normal blood itself has been credited with a ferment which has the power of destroying sugar (glycolysis). But with rigid aseptic precautions the loss of sugar, even in several hours, is small, and it is doubtful whether such a ferment exists. Even under the most favourable circumstances the quantity of dextrose which blood can destroy is so small a fraction of that which disappears in the same time in the body, that it is probably of no importance in carbo-hydrate metabolism (Macleod). On the other hand, Cohnheim stated that while no glycolytic ferment can be demonstrated in the pancreas, and only an exceedingly weak glycolytic action in muscular tissue (Brunton), by combining ex- tracts of pancreas and extracts of muscles, distinct glycolysis, due to a ferment action, could be produced. He suggested that the glycolytic ferment is activated by another substance, as trypsinogen is activated by enterokinase (p. 372). This announcement aroused great interest, since it is known that the pancreas is intimately concerned in the metabolism of sugar. That sugar disappears under the conditions of Cohnheim's experiments has been confirmed by a number of observers. But his interpretation of his results has not been generally accepted. According to Levene and Meyer, the dextrose, far from being burnt, seems to be condensed to a poly- saccharide, and can be recovered by hydrolysing this compound when the mixture is acted on by dilute acid. The action of the pancreas- muscle mixture is, therefore, not a true glycolysis. In- deed, of all the tissues investigated by Levene, leucocytes alone can be credited with a real glycolytic action. Excision of the pancreas in dogs causes permanent glycosuria (pancreatic diabetes) (v. Mering and Minkowski), which is prevented if a portion of the pancreas be METABOLISM OF CARBO-HYDRATES— GLYCOSURI AS 547 left (p. 636). Diabetes in man is known to be frequently associated with pancreatic lesions. Although much still remains obscure, the study of this pathological form of glycosuria and of the experimental glycosurias has thrown light upon the normal metabolism of carbo- hydrates and upon those regulative mechanisms whose breakdown is responsible for the excretion of sugar. It will be best to discuss the experimental glycosurias first, and to begin with the form which probably is better understood than any other, the so-called punc- ture glycosuria. Puncture Glycosuria — Sugar-Regulating Mechanism. — An arti- ficial and temporary glycosuria, in which the sugar in the urine un- doubtedly arises from the hepatic glycogen, can be caused by punc- turing the medulla oblongata in a rabbit — for example, at a level between the origins of the auditory nerves and the vagi. It is stated that a puncture of the thalamencephalon, or 'tween-brain (p. 850), produces the same effect. If the animal has been previously fed with a diet rich in carbo-hydrates — that is, if it has been put under con- ditions in which the liver contains much glycogen — the quantity of sugar excreted by the kidneys will be large. The immediate cause of the glycosuria is an increase in the sugar content of the blood (hyperglycaemia), an increase which is most pronounced in the blood of the hepatic vein. If, on the other hand, the animal has been starved before the operation, so that the liver is free, or almost free, from glycogen, the puncture will cause little or no sugar to appear in the urine, and the proportion of sugar in the blood will remain normal. That nervous influences are in some way involved in the mobilization of the glycogen reserve of the liver is shown by the absence of glycosuria if the splanchnic nerves, or the spinal cord above the third or fourth dorsal vertebra, be cut before the puncture is made. But sometimes these operations are themselves followed by temporary glycosuria, due, it is believed, to irritation of the same efferent nervous path whose elimination when the splanchnics are divided prevents the glycosuria. The simplest explanation of the phenomena is that a ' sugar centre ' — that is to say, a centre which has the important office of regulating the sugar content of the blood by governing the rate at which glycogen is built up and de- composed in the liver, as the salivary centre regulates the rate at which the constituents of saliva are formed and discharged— has been injured or irritated by the puncture. If a nervous centre does in fact preside over this internal secretion of the liver, it will, of course, be connected with efferent and afferent nerves. The former, as defined by the experiments alluded to, seem to be confined to the splanchnic nerves; the latter are believed to run especially, though not exclusively, in the vagus. Section of the vagi has no effect either in causing glycosuria of itself or in preventing the ' puncture ' glycosuria, but stimulation of the central ends of these 548 METABOLISM, NUTRITION AND DIETETICS and of other afferent nerves may cause sugar to appear in the urine, although not, it is said, if precautions are taken to prevent any degree of asphyxia. Asphyxia produces an increase in the sugar content of the blood, an increase in the flow of urine and glycosuria. It has usually been assumed that this action of asphyxia is due to the effect upon the centre of blood over-rich in carbon dioxide (and other metabolic products) or impoverished as regards oxygen. But there is some evidence that the altered blood may also affect the liver-cells directly, or, what comes to the same thing in the long-run, that interference with the internal respiration of the hepatic tissue, operating, it may be, through an increase in the concentration of the hydrogen ions, upsets the equilibrium of those intracellular reactions by which glycogen is formed from dextrose and dextrose from glycogen. In like manner it may be supposed that under normal conditions the rate of transformation of the hepatic glycogen into dextrose is adjusted to the dextrose content of the blood, not only by reflex nervous impulses passing through the sugar-regulating centre, but also by the direct influence of the dextrose itself circu- lating in the blood, upon whose concentration the reaction of the centre on the one hand and of the liver-cells on the other may in part depend. So that when the proportion of sugar in the blood tends to sink we may perhaps picture the centre as sending impulses to the liver which increase the rate at which the glycogen is hydro- lysed ; and when the proportion tends to rise, we may think of it as sending impulses which inhibit the hydrolysis, both effects being accentuated by the direct influence of the changes of concen- tration on the hepatic cells. Whatever the mechanism may be through which the puncture hastens the transformation of glycogen into dextrose in the liver, there is no evidence that the amount of the enzyme which hydrolyses the glycpgen is increased. Whether the action of the enzyme is favoured in some other way — e.g., by the production of a co-ferment or by some change in the condition of the glycogen which renders it more open to attack — is unknown. Certain facts have recently been brought forward which go to show that the action of the splanchnic fibres on the liver is not a direct action, but that in some way or other the concomitant activity of the adrenal glands is essential. For if the adrenals have been previously extirpated, the puncture does not cause glycosuria. It was at first thought that the reason for this was that the removal of - the adrenals is itself followed by the disappearance of glycogen from the liver, and, as has been pointed out, the presence of glycogen in the liver is essential to the success of the puncture experiment. The matter, however, is not so simple. For although in certain animals — e.g., the dog — the liver does lose all its glycogen when the adrenals have been taken away, this is not the case in the rabbit, METABOLISM OF CARBO-HYDRATES— GLYCOSURI AS 549 and yet in the rabbit also the urine remains free from sugar after puncture in the absence of the adrenal glands. In some way or other, then, the adrenals do intervene in the production of puncture glycosuria. The observation, which is easily confirmed, that the injection of adrenalin (or epinephrin) (p. 550) under the skin or into the blood, or into one of the serous sacs, does cause a pronounced increase in the sugar content of the blood, and the appearance of dextrose in the urine, seemed at first to supply the missing link in the chain of evidence. What could be simpler than the assumpt:, n that the splanchnic fibres stimulated in the puncture experiment were fibres going not to the liver, but to the adrenals, which occa-' sioned an outpouring of adrenalin into the blood^and that puncture glycosuria was therefore merely a particular case"of adrenalin glyco- suria ? It is known that excitation of the splanchnic nerves cauces the passage of adrenalin into the blood of the adrenal veins (p. 661). It is known that puncture of the medulla oblongata diminishes the epinephrin content of the adrenal glands. The argument seemed straightforward, and the adrenal hypothesis of puncture glycosr.ria triumphant. As soon, however, as the matter was put to the test of quantitative experiments, the hypothesis began to crumble. It was shown, for example, that during a stimulation of the splanchnic nerves sufficient to cause a decided increase in the dextrose content of the blood, a quantity of adrenalin was given off. to the Adrenal veins, which, when mingled with the rest of the blood on its way to the liver, could not possibly amount to more than one in a hundred million parts of blood, a concentration in which adrenalin, when introduced artificially into the blood-stream, produces no glycosuna whatever. Still more significant is the fact that, after destroying the hepatic plexus, stimulation of the splanchnic nerves causes no in- crease in the blood-sugar in spite of the increased output of adrenalin by the way of the adrenal veins. On the other hand, excitation of the hepatic plexus causes hyperglycsemia (Macleod and Pearce). It is not, then, a direct action on the liver of epinephrin secreted in response to stimulation of splanchnic fibres supplying the adrenal glands which is responsible for the increase in the dextrose content of the blood. The adrenals, however, play some part. For in their absence stimulation of the hepatic plexus is not followed by hyper- glycaemia. But whether this is due to general derangement of the normal carbo-hydrate metabolism in their absence, or to the loss of . some special influence on the liver, without which stimulation of the hepatic plexus is ineffective, is unknown. Although several of the operations which lead to temporary glycosuria undoubtedly bring about changes in the hepatic circula- tion, it is as yet impossible to say whether vaso-motor effects con- tribute essentially to the result, or whether it is due entirely to nervous stimulation of the liver-cells, or to withdrawal of such 550 METABOLISM, NUTRITION AND DIETETICS stimulation or control (see also p. 509). There is some evidence that excitation of the uncut great splanchnic nerve (on the left side) in dogs may cause hyperglycsemia, diuresis, and glycosuria, even under conditions in which as far as possible circulatory effects are eliminated. Contrariwise, when in the puncture experiment on an unnarcotized animal the small instrument does not wound the medulla oblongata in the right place, a rise of blood-pressure due to excitation of the vaso-motor centre may occur without any glycosuria. But absolute proof of the existence of glycogenolytic nerve fibres going to the liver — that is, fibres whose stimulation accelerates the hydrolysis of glycogen into dextrose (Macleod) — has not yet been brought forward. Adrenalin Glycosuria.— In adrenalin glycosuria the sugar-content of the blood is increased. A given quantity of adrenalin introduced subcutaneously produces a more marked hyperglycaemia and glycosuria than the same amount injected into a vein or into muscle. The best evidence is that the glycosuria is produced by some action on the liver, possibly through the excitation of sympathetic fibres controlling the production of dextrose from glycogen (Underbill and Closson), or by a direct effect on the hepatic cells, which hastens the normal transformation of glycogen into dextrose, or hinders the normal transformation of dextrose into glycogen. It has been stated that in the isolated surviving liver of the frog adrenalin causes the glycogen to be rapidly converted into dextrose. While this confirms the view that experimental adrenalin glycosuria is due to an action on the liver which increases the sugar-content of the blood, it does not necessarily show that the action is exerted directly on the hepatic cells without the intervention of nerve fibres. For the sympathetic nerve-endings may survive a considerable time. The theory that epinephrin causes glycosuria by inhibiting the internal secretion of the. pancreas, and that the condition is therefore a par- ticular variety of pancreatic diabetes, is erroneous. Adrenalin glycosuria does not seem to be in any way related to true diabetes. The complete metabolism of dextrose is not interfered with. Indeed, a much larger proportion of the total heat produced comes from the destruction of sugar after the subcutaneous injection of epinephrin into dogs than in the normal animals (Lusk and Riche). If in spite of this glycosuria ensues, it is only because the carbo- hydrate reserve of the body is mobilized so rapidly that it cannot possibly be all consumed. Nor does epinephrin cause any increased production of sugar from protein or from fat. For in dogs rendered diabetic by phlorhizin and freed from glycogen by shivering, injec- tion of epinephrin is not followed by an' increase of either sugar or nitrogen in the urine (Ringer). After repeated injections of adren- alin, a tolerance for it is established, and glycosuria is no longer caused. METABOLISM OF CARBO-HYDRATES-GLYCOSURIAS 551 Fhlorhizin Glycosuria, produced by subcutaneous injection of the glucoside phlorhizin, agrees with pancreatic, but differs from punc- ture diabetes in this, that it can be produced in an animal free from glycogen, and is accompanied by extensive destruction of proteins. It differs from other forms of diabetes in being associated, not with an increase, but with a diminution, in the sugar of the blood. This is best explained by supposing that the phlorhizin acts on the kidney in such a way as to increase the permeability of the glomeru- lar epithelium for sugar, or (in terms of the secretion theory of urine formation) in such a way as to increase its sensitiveness to the stimulus of sugar circulating in the blood. The sugar is therefore rapidly swept out of the circulation, and this leads secondarily to an increased production of sugar to make good the loss. In addi- tion, within certain limits there is a total inability on the part of the body to consume dextrose. After the preliminary sweeping out of the sugar already in the body, a definite ratio is established between the dextrose and the nitrogen eliminated in the urine (dextrose : nitrogen : : 3~6or 37 : i). The sugar at this stage is produced entirely from proteins, and not at all from fat. It is a fact of considerable interest that, if small quantities of dextrose are now given, the amount of protein de- stroyed is reduced to some extent, although all of the dextrose is excreted, and none of it is burnt (Ringer). This supports the hypothesis of Landergren that in starvation some of the protein is metabolized for the formation of the indispensable dextrose, and that this fraction can be ' spared ' by carbohydrate, though not by fat. The protein metabolized is so much increased under the influence of phlorhizin that it exceeds the starvation requirement by a greater amount than in pancreatic diabetes, perhaps because the diminished content of sugar in the blood constitutes a more insistent call upon the proteins to produce sugar. In pancreatic diabetes, where hyper glycsemia exists, there can at least be no reason for the formation of sugar from protein for the maintenance of the normal sugar-content of the blood, and it is interesting that in this condition the giving of dextrose does not seem to spare any protein (p. 606). The degree of intolerance for carbo-hydrates in pathological diabetes may be arrived at by putting the patient on a diet of protein and fat (rich cream, meat, butter, and eggs), and determining the ratio of dextrose to nitrogen excreted. If it as 3-6 or 3-7: i, intolerance is complete, none of the dextrose produced from protein being burned (Lusk and Mandel). Glycosuria can be caused in many other ways than those already mentioned. Sometimes the action seems to be a direct one on the sugar-regulating centre — e.g., in concussion of the brain, occlusion and subsequent release of the arteries supplying the brain and cervical cord, and acute haemorrhage. Carbon monoxide has a 552 METABOLISM, NUTRITION AND DIETETICS similar action owing to the deficiency of oxygen occasioned by it. Many drugs also cause glycosuria, including curara, morphine, strychnine, phosphorus, chloroform, ether, and other substances, some of which may act on the ' sugar centre,' although others — e.g., phosphorus and chloroform— are poisons which can affect the liver directly. It is probable that some of the experimental hyper- glycaemias are due to an associated acidosis. For when the hydrogen- ion concentration of the blood is increased the transformation of glycogen into dextrose in the liver is accelerated. The adminis- tration of alkali is said to have a beneficial influence upon the oxidation of dextrose in dogs after total or partial extirpation of the pancreas (Murlin and Kramer). Injection of water or physiological salt solution into the bile-ducts, or into the mesenteric veins, or of salt solution in considerable amount into the general circulation, is followed by glycosuria (Fischer, etc.). It is a mistake to apply the term diabetes to most of the forms of artificial hyperglycaemia and glycosuria. The condition produced by removal of the pan- creas, which will be returned to in Chapter XI., is, however, a true diabetes, a derangement of metabolism of the sam° general nature as that which underlies human diabetes. Diabetes Mellitus. — In the natural diabetes of man, as in all the forms of glycosuria mentioned, with the exception of that pro- duced by phlorhizin, the immediate cause of the glycosuria is the in- crease of sugar in the blood. Instead of the i part per 1,000, or a little more or less, which constitutes the normal proportion in a healthy man, in diabetes 3 or 4 parts, and in exceptional cases even 7 to 10 parts per 1,000 may be present. The riddle of diabetes is the explanation of this persistent hyperglycaemia. Innumerable hypo- theses have been framed to account for this, but on the whole three possibilities have been emphasized: (i) That the power of temporarily storing carbohydrates is deranged; (2) that the power of the tissues to utilize carbo-hydrates (i.e., eventually dextrose) is diminished or abolished ; (3) that too much sugar is produced in the body. In addition, some writers have postulated a fourth factor to explain certain cases (of so-called ' renal diabetes ') — to wit, an increase in the permeability of the kidneys for sugar, as in phlorhizin glycosuria. Lest the student should be bewildered amongst all these theories, he should take note that while the second factor is now recognized as the essential one, there is some reason to believe that diabetes mellitus is not in every case a single patho- logical condition, but may comprise a group of such conditions. Some cases may therefore present a picture conforming closely to one or to another of the experimental glycosurias, but others a picture compounded of features characteristic of two or of several of these experimental conditions. METABOLISM OF CARBO-HYDRATES— GLYCOSURI AS 553 It is possible that in some cases the sugar coming from the ali- mentary canal passes entirely or in too large amount through the liver, owing to a deficiency in its power of forming glycogen. But although in certain cases of diabetes specimens of the hepatic cells, obtained by plunging a trocar into the liver, have been found free, from glycogen, in others glycogen has been present. The muscles also are usually stated to be much poorer in glycogen than normal muscles, but this might just as well be the case because glycogen was being transformed into sugar with abnormal ease as because there was interference with glycogen formation. Indeed, it is said that in the heart muscle of depancreatized dogs there is more glycc- gen than in normal heart muscle. It must be carefully remembered that the amount of glycogen present in a tissue gives no information as to the rate at which it is being formed or decomposed. And if the cause of the supposed defect in glycogen-forming power be the absence of a glycogen-forming ferment, or its production in too small an amount, the same circumstance may occasion a too tardy transformation into sugar of whatever glycogen happens to be present. In this case the sugar-regulating function of the glyco- gen store would be equally lost, whether the storehouses were permanently filled with long-formed glycogen or only half-filled or empty. In addition to an interference with the due and regulated storage of the surplus sugar as glycogen, it has usually been thought neces- sary for a rational explanation of the facts of diabetes, to assume that from some change in the tissues sugar has ceased to be a food for them, or is used up in smaller amount than in the healthy body. More and more the evidence points to this as the fundamental change both in the human disease and in experimental pancreatic diabetes. Why the tissues cannot decompose and utilize dextrose as they normally do, if it be really the case that they fail in this regard, is a question of great interest, but as yet no satisfactory answer can be given. It appears probable that the failure occurs at one or more of the earliest stages in the intermediate metabolism of carbo- hydrates (p. 542) or in the preliminary processes, whatever they may be, which, without profoundly altering the dextrose molecule, prepare it for the series of decompositions, in the course of which it eventually parts with all iis chemical energy. For it has been shown that many of the products of the cleavage or oxidation of sugar, even those in which the decomposition has proceeded but a little way — e.g., glyconic and glycuronic acids (p. 543) — are com- pletely utilized by the tissues of diabetics and of depancreatized dogs. And the derangement in the normal sequence of events, of whatever nature it may be, is not so deep-reaching as to prevent 554 METABOLISM, NUTRITION AND DIETETICS the retracing of the chemical steps by which sugar is synthesized from such derivatives of the proteins as amino-acids or their further degradation products. As to the actual cause of the alleged in- capacity of the tissues to consume dextrose, the change has by some been supposed to be the loss or diminution of a glycolytic ferment or a substance necessary for the activation of such a ferment. And although the sugar-destroying power of blood from diabetic patients, or from animals in which glycosuria has been caused by phlorhizin, is not at all inferior to that of healthy blood, it has been maintained that the intracellular glycolytic ferments, if such really exist, are much less active, especially in the more severe forms of the disease, which conform so closely in their clinical manifestations to the pic- ture presented by the depancreatized animal. Nevertheless, up to the present ah1 attempts to satisfactorily demonstrate for isolated tissues a loss or even a diminution in the capacity to utilize dextrose have broken down. In eviscerated dogs, for example — that is, in preparations consisting mainly of skeletal muscle — it has been found impossible to make out any deficiency as compared with normal animals in the amount of dextrose disappearing in a given time from the blood, even when the animals have been deprived of the pancreas as long as a week before the experiment, and therefore exhibit the condition of pancreatic diabetes in full intensity (Macleod and Pearce). This conclusion has been confirmed for the isolated heart -lung preparation. As regards the hypothesis that an increased production of sugar from proteins, or it may be from fat, is the essential proximate cause of the hyperglycaemia and the glycosuria, there is no good evidence that this factor, acting by itself in the absence of a derangement of the regulative influence of the glycogen store, and in the absence of a derangement of the normal katabolism of dextrose, is ever respon- sible for pathological diabetes. But a secondary overproduction of sugar unquestionably occurs in many cases. The tissues, bathed as they are in liquids rich in dextrose, are nevertheless starving for sugary if they cannot use what is offered to them, and the body labours to avert the famine by increasing its production of sugar, the sugar- forming tissues being stimulated to their task either through nervous influences or by chemical messengers circulating in the blood. In depancreatized dogs, and in dogs under the influence of phlorhizin, glycerin, given by the mouth, causes an increase in the excretion of sugar up to two or three times the original amount. The giving of fat does not increase the amount of sugar excreted, which, however, is increased by such substances as egg-yolk, which contain lecithin. These should accordingly be avoided in cases in which a strictly antidiabetic diet is desired. It is much more im- portant to exclude carbo-hydrates largely or entirely from the food, METABOLISM OF CARBO-HYDRATES— GLYCOSURI AS 555 although oatmeal and potatoes are said to occupy an exceptional position, and have even been recommended as beneficial. Calcium chloride has been stated to diminish the sugar excretion in diabetes (Boigey), and it has a similar effect in certain of the artificial glyco- surias (Brown, Fischer). In many cases, even when carbo-hydrates are completely, or almost completely, omitted from the food, sugar, derived from the breaking-down of proteins, and possibly to some extent from fats, still continues to be excreted, although in smaller quantity. Other products formed or imperfectly transformed in the deranged meta- bolism, especially of fats, such as acetone, aceto-acetic acid, and oxybutyric acid (the so-called acetone bodies), may also appear in the urine (ketonuria), or, accumulating in the blood, may, by uniting with its alkalies, seriously diminish the quantity of carbon dioxide which that liquid is capable of carrying, and thus lead to the condition known as diabetic coma. The small amount of carbon dioxide in the venous blood may also be partly due to the hyperpnoea, marked by increased depth of the respiratory movements produced by stimulation of the respiratory centre by other substances than carbon dioxide. The increased ventilation causes a fall in the carbon dioxide pressure in the alveolar air, and therefore an increased elimination of that gas from the blood. This form of coma appears to be really in part an acid-poisoning comparable to the condition produced in animals by doses of mineral acids too large to be neutralized by the ammonia split off from the proteins. The ad- ministration of very large doses of alkalies (sodium bicarbonate, for instance, to the amount even of hundreds of grammes) has been recommended for the treatment of this serious complication, and in some cases i.t is successful in staving it off for a time. Often, however, in spite of a prolonged course of treatment, during which the urine has continued distinctly alkaline, fatal coma eventually occurs. The coma then is not merely a symptom of acidosis, but is also due to the specific toxic effects of the acids even when neutralized. Other toxic products may also be formed in the deranged metabolism. The appearance of the acetone bodies in diabetes presents a problem which cannot be said to have been as yet completely solved. Oxybutyric acid, from which aceto-acetic acid and acetone are easily derived (p. 567), seems to be one of the intermediate steps in the normal metabolism of fats. But whereas under ordinary circumstances it is readily oxidized in the body, in diabetes the power of the tissues to burn oxybutyric acid seems to suffer just as does the power to utilize dextrose. The suggestion that in diabetes the abnormally great consumption of fat entailed by the loss of availability on the part of the carbo-hydrates causes the intermediary metabolism of fats to be scamped, as it were, is not satisfactory. For many animals and some races of men dwelling 5j6 METABOLISM, NUTRITION AND DIETETICS in cold climates consume with impunity much greater quantities of fat than any diabetic organism.* N II. — THE METABOLISM OF FAT. Chemistry of Fats. — The fats are compounds (esters) of an alcohol with fatty acids, and can be split, with assumption of water, into these constituents by the action of acids or alkalies or of enzymes (lipases). In the majority of the ordinary fats, and in all those which are of physiological importance (the triglyce rides), the alcohol is glycerin. The fatty acid components which may be united with the glycerin are very numerous, and the physical properties of the different fats — e.g., their melting-points and solubilities — are closely related to the physical properties of the corresponding fatty acids. Thus palmitic and stearic acids are solid at ordinary temperatures, and so are palmitin and stearin, the glycerin esters or fats formed with these acids. Olcic acid, on the contrary, is fluid at the ordinary temperature, and the corresponding fat, olein, is a liquid fat or oil. On the chemical side the fatty acids can be distinguished as saturated and unsaturated. The fatty acids of the series CnH^+x-COOH are saturated acids. Where n is o we have formic acid, H.COOH; where n is i, acetic acid, CH3.COOH; where n is 2, propionic acid, CH3.CHa.COOH ; where n is 3, butyric acid, CH3.CH2.CHa.COOH, and so on, each acid in the series differing from the one immediately preceding it in possessing an additional CHa group. In the case of the higher members of the series these carbon chains become very long. In palmitic acid, for instance, CHS.(CH2)14.COOH, there are fourteen CHa groups, and in stearic acid, CHs.tCH^.COOH, sixteen. Oleic acid,C8217"X=C/™ . COOH is a representative of a series of unsaturated fatty acids whose general formula is CnH^-i-COOH. As the formula of oleic acid shows, the unsaturated fatty acids contain hi their molecule two carbon atoms united by a double link, and one of these valencies can be occupied by halogens (e.g., chlorine) or by oxygen. Erucic acid, a fatty acid occur- ring in certain vegetable oils — for example, in rape oil — also belongs to this series, and linolic acid, found in linseed oil, to another series of unsaturated fatty acids. Then there are the so-called oxyfatty acids, which in their turn comprise saturated and unsaturated acids. They differ from the ordinary fatty acids in containing one or more OH groups. Thus a dioxystearic acid, C17H33(OH)2.COOH, in which two of the H atoms in stearic acid are replaced by OH, is found in castor-oil. It is clear, from the great variety of the fatty acids, that by their union with glycerin (with loss of water) a very large number of different fats can be formed . Thus, when all the OH groups in the trivalent alcohol are replaced by palmitic acid we have tripalmitin ; when they are replaced by stearic acid, tristearin ; when they are replaced by oleic acid, triolein ; and so on. As a group such fats may be termed homo-acid fats, since all the OH groups are replaced by the same fatty acid. Thus — CHj.OH C^H^.COOH CH.OH + CH CHj.OH C^H Glycerin. 3 molecules stearic acid. Tristearin. \Vatet. * The importance of controlling the diet, even to the extent of introducing periods of fasting, so as to keep the diabetic free from acidosis and glycosuria, has been recently emphasised (Allen) . THE METABOLISM OF FAT 557 But it is not necessary that each OH group in the alcohol should be replaced by the same fatty acid, and when this does not occur we have hetero-acid fats. For instance, one can be replaced by steeiric acid, and the remaining two by palmitic acid, yielding a fat called ' stearo- dipalmitin.' Conversely, one OH may be replaced by palmitic and two by stearic acid, forming palmito-distearin. Similarly, a dioleo-stearin (glycerin combined with two molecules of oleic and one of stearic acid), and an oleo-distearin (glycerin combined with two molecules of stearic and one of oleic acid) are known. Such compounds have been isolated from the fat of animals, and also formed synthetically. Again, each of the OH groups in the alcohol can be replaced by a different fatty acid. It is obvious, then — and this is the point to which these chemical details are intended to lead up — that the number of different fats which the animal organism has at its disposal for concocting those varied mixtures designated as body fat is very great, and that there is room for a considerable degree of specificity in the fat stores of different animals, and it may be in the fat contained in different organs of the same animal, even if this specificity is not as marked as in the case of the proteins. It may be added, in connection with the composition of the body fat, that small quantities of free fatty acids and of glycerin may be present ; but there is reason to believe that these are simply the surplus of raw materials which is about to be synthetized to neutral fat, or the surplus of decomposition products of the neutral fat which have not yet left the fat depots to take their place in the metabolism of the tissues. The discussion of the metabolism of fat involves a study — (i) of the transformations and migrations of the food fat before it begins to be utilized ; (2) of the possible production of fat from other con- stituents of the food ; (3) of the processes and the stages by which fat, whatever its origin, undergoes katabolism to its end products. The fat of the food, passing along the thoracic duct into the blood- stream, is soon removed from the circulation, for normal blood contains only traces, except during digestion. Where does it go ? What is its fate ? Transformation and Migration of the Food Fat. — The presence of adipose tissue in the body might suggest a ready answer to these questions. The fat-cells of adipose tissue are ordinary fixed connective-tissue cells which have become filled with fat, the protoplasm being reduced to a narrow ring, in which the nucleus is set like a stone. It would, at first thought, seem natural to suppose that the fat of the food is rapidly separated by these cells from the blood, and slowly given up again as the needs of 'the organism require, just as carbo-hydrate is stored in the liver for gradual use. And it has been found that a lean dog, fed with a diet containing much fat and little protein, puts on more fat, as estimated by direct analysis, or keeps back more carbon, as esti- mated by measurements of the respiratory exchange, than can be accounted for on the supposition that even the whole of the carbon of the broken-down protein corresponding to the excreted nitrogen 558 METABOLISM. NUTRITION AND DIETETICS has been laid up in the form of fat. Even with a diet of pure fat — • and with such a diet digestion and absorption are carried on under unfavourable conditions — more carbon is retained than can have come from the metabolism of the proteins of the body, as measured by the nitrogen given off in the urine and faeces: the fat passes rapidly from the blood into the organs, and especially into the liver (Hofmann, Pettenkofer and Voit). It is thus certain that some of the absorbed fat may be stored up as fat in the body. This is borne out by the careful experiments of Munk and Lebe- deff, who found that, when dogs are fed with excess of foreign fat (linseed oil, rape oil, mutton fat), a fat is laid down which is quite different from dog's fat, and has the greatest resemblance to the fat of the food. Thus, when rape oil, which contains a fatty acid, erucic acid, not found in animal fat, was given, erucic acid could be detected in the fat laid on. When the dogs were fed with mutton fat, whose melting-point is much higher than that of dog's fat, the fat laid on did not melt till it was heated to 40° C. or more. When they were fed with linseed oil, the body-fat was found liquid even at o° C. We have already referred (p. 444) to the fact that neutral fat can be built up in the wall of the intestine from fatty acids given in the food. Munk has shown that fat formed in this way can also be laid down as body-fat. But besides the fat and fatty acids of the food, the fat of the body has other sources, and some of it is produced by more complex processes. The fat of a dog consists of a mixture of palmitin, olein, and stearin. When a starved dog was fed on lean meat and a fat con- taining palmitin and olein, but no stearin, the fat put on contained all three, and did not sensibly differ in its composition from the normal fat of the dog (Subbotin). Stearin must, therefore, have been formed in some way or other in the body. If it was produced from the olein and palmitin of the food, the portion of these deposited in the cells of the adipose tissue must have undergone changes before reaching this comparatively fixed position. But there is conclusive evidence that fat may be derived from other sources, certainly from carbo-hydrates, and probably from proteins; and the stearin may have been formed from the carbo-hydrates or proteins of the food or tissues, and not directly from fat. And if the stearin was produced from proteins or carbo-hydrates, it is evident that the olein and palmitin might have been formed in this way too, the portion of the carbo-hydrate or protein devoted to this purpose being sheltered from oxidation by the combustion of the fats of the food. It is well known that not only neutral fats, but also fatty acids, exert such a ' protein-sparing ' action. It is possible also that the fat which is normally excreted into the intestine (p. 443), and which is perhaps derived from broken-down proteins, may be re- absorbed, and take its place among the fat ' put on.' THE METABOLISM OF FAT 559 At this point in the discussion it is necessary to remark that a distinction ought to be established between that store of surplus fat laid down in the connective tissue which, in order to avoid com- plicating the matter unduly, has hitherto been referred to as if it constituted the whole of the body-fat, and the fat which is contained in greater or less amount in all the tissue cells. The fat con- tained in the tissue elements — e.g., in the liver cells — in the visible form of droplets, and which can be easily extracted from them by solvents such as chloroform, should also be distinguished from the fat which is so intimately incorporated or combined with the cell substance that it can only be extracted after this has been digested by the aid of proteolytic ferments or acids. The latter fraction of the body-fat is probably an integral and indispensable constituent of the protoplasm. Now, it is in the great fat depots of the sub- cutaneous tissue and the mesentery and omentum that variations in the proportions of the various fatty acids corresponding to varia- tions in the nature of the food-fat are most easily produced, or, at least, most easily observed. These depots are laid down, not in the interest of the fat cells themselves, but to serve the purpose of a reserve of fat which may be drawn upon for the nutrition of the body as a whole, just as the glycogen store of the liver forms a general carbo-hydrate reserve. The free fat in the cells of the organs is superficially analogous to the glycogen reserves of such tissues as muscles and glands, and certain facts are known which might be interpreted as indicating that this fraction of the body-fat, like the fat of the connective tissue, is not a definite and specific mixture of fats with an unvarying composition for each kind of animal, but a mixture whose composition can be made to vary by altering the nature of the fats in the food. On the other hand, the fat combined in the tissues appears to preserve a certain specificity which is inde- pendent of the fats supplied in the food. Thus, wh'en dogs were fed with rape oil, and had accumulated considerable quantities of this fat of low melting-point in the subcutaneous and other fat depots, \ he fat combined in the organs remained in all respects the same as normal dog's fat. This was also the case with animals fed on fat of high melting-point, such as sheep's tallow (Abderhalden). Although the liver appears to have a special relation to the metabo- lism of fat, it is not known whether any particular organ is more than the rest responsible for the manufacture of this specific mix- ture of fats. It appears more probable that each cell has the power of forming for itself the characteristic fats from the crude materials represented by the food-fat directly absorbed from the tissue lymph, or the fat of the depots after it has been mobilized and has found its way again into the blood, or, finally, from other materials than fats, such as dextrose or some of its decomposition products. Even in the case of the subcutaneous and similar collections of 560 METABOLISM. NUTRITION AND DIETETICS fat, it must be noted that upon the whole, under normal conditions, it is their specificity of composition rather than their dependence upon the composition of the fat mixture in the food which is the striking fact, and undue weight can easily be given to the results of feeding experiments where great quantities of quite foreign fats are administered. When small quantities of fats very far removed in their properties from the normal fat of an animal are given in the food, they are either completely utilized before reaching the fat depots, or transformed into normal body- fat, since no change whatever can be detected in the latter. If they have been utilized, then it may be that a corresponding amount of fat, formed, say, from dextrose, has been laid down in the fat stores. If this fat is formed from dextrose, it will, of course, be the kind of fat which the particular animal is accustomed to form from dextrose — that is, the fat characteristic of the animal. If the foreign fat is itself transformed into body-fat when given in small amount, this same feat can without doubt be gradually accomplished in the case of the surplus of foreign fat laid down in the depots when a large quantity of it is given in the food. Formation of Fat from Other Sources than the Fat of the Food — (i) From Carbo- Hydrates. — It has been found that the addition of protein to a diet of fat, and especially to a diet of carbo-hydrate, in larger amount than is just necessary for nitrogenous equilibrium (p. 602), leads to a more rapid increase in the carbon deficit — that is, in the fat put on — than if the minimum quantity of protein required for nitrogenous equilibrium had been given. From this it is inferred that the carbonaceous residue of the broken-down protein is shielded from oxidation by the fat, and to a still greater extent by the carbo-hydrates, and so retained in the body as fat. And there is little doubt that the high repute of carbo-hydrates as fattening agents is in part due to their taking the place of proteins and fats in ordinary ' current ' metabolism, and so allowing body-fat to be laid down from these. Voit, indeed, has gone so far as to assert that this is the only sense in which carbo-hydrates can be said to form fat, and that, in carnivorous animals at least, a direct con- version never occurs. But the experiments of Rubner have shown that in a dog fed with a diet rich in carbo-hydrates, and containing but little fat and no proteins at all, the carbon deficit was greater than could be accounted for by the proteins being broken down in the body and the fat of the food. In the pig and goose, too, the direct formation of fat from carbo-hydrates has been demonstrated. For example, in an experiment by Tscherwinsky two young pigs of the same litter were taken. They weighed respectively 7,300 grammes and 7,290 grammes. One was killed, and the amount of fat and nitrogen in its body directly estimated. From the nitrogen the maximum quantity of protein which could be present was calculated. The other pig was fed for four months with barley, which was analyzed. The excreta were also analyzed to determine the amount of unabsorbed THE METABOLISM OF FAT 56* fat and protein. At the end of the iour months the pig was killed. It now weighed 24 kilogrammes, and contained 2-52 kilogrammes protein and 9-25 kilogrammes fat. Subtracting the protein (0-96 kilo- gramme) and fat (0-69 kilogramme) originally present, 1-56 kilogrammes of protein and 8-56 kilogrammes of fat must have been put on. The amount of protein taken in the food was 7-49 kilogrammes, and of fat 0-66 kilogramme. Therefore, 5-93 kilogrammes of protein must have been used up, and 7-90 kilogrammes of fat laid on. At least 5 kilo- grammes of this fat must have come from the carbo-hydrate of the food. Only a small amount of the fat put on could possibly have come from the protein. The production of wax by bees, which used to be given as a proof of the formation of fat from sugar, is not decisive, for in raw honey proteins are present ; and even when bees fed on pure honey or sugar manufacture wax, it may be derived from the broken-down proteins of their own bodies. It is probable that in the formation of fats the carbo-hydrates are first split up to some extent, and that the fats are then con- structed from their decomposition products, oxygen being lost in the process, since fat is much poorer in oxygen than carbo-hydrate. But the chemistry of the transformation as it takes place in the body is still imperfectly known, and all that can be done here is to indicate one or two of the ways in which chemists conceive that it may occur. The formation of the glycerin component of the neutral fats from carbo-hydrates would appear to present little difficulty. In dis- cussing the formation of glycogen from glycerin (p-536), it was stated that two molecules of glycerose (glycerin aldehyde), a triose or sugar with three carbon atoms, can be condensed to form a hexbse or sugar with six carbon atoms like dextrose, from the condensation or union of a number of molecules of which, with abstraction of water, glycogen is built up. The reaction can be worked equally well in the reverse direction — that is, from the hexose dextrose two molecules of glycerin aldehyde can be formed, and then from each molecule of the alde- hyde, by reduction, a molecule of the alcohol glycerin. As a matter of fact, it has been demonstrated that glycerin is produced when the cor- responding aldehyde is brought into contact with minced liver. As regards the fatty acid components of the fats, it will be seen from the schematic representation of the katabolism of dextrose on p. 544 that acetic acid, a fatty acid, is represented at one of the stages as being formed by the oxidation of a molecule of acetaldehyde. Lactic acid is represented in the same scheme as a previous stage in the decom- position of dextrose, and lactic acid can be converted into acetaldehyde and formic acid, the lowest of the same series of fatty acids of which acetic acid is the next highest member. Thus : H.COOH Lactic acid. Acetaldehyde. Formic acid. Aldehydes (as well as ketones) have a great capacity for entering into reactions with other substances, and their molecules show also a marked tendency to combine with one another, forming new compounds by their condensation. Thus, from two molecules of acetaldehyde one 36 362 METABOLISM, NUTRITION AND DIETETICS molecule of aldol is formed, which by transposition of certain groups, becomes butyric acid, the fourth member of the fatty acid series of which acetic acid is the second member, and palmitic and stearic acids, which form such important constituents of the ordinary body - fats, the sixteenth and eighteenth members respectively. By oxidation aldol becomes /3-oxybutyric acid, which by further oxidation yields aceto- acetic acid, compounds already referred to in connection with diabetes (p. 555). The following equations illustrate these reactions: = CH3.CH(OH) .CH Acctaldehyde. Acetaldehyde. Aldol. CH3.CH (OH) .CH2.CHO - CH3.CH2.CHa.COOH Aldol. Butyric acid. CH3.CH(OH).CHa.CHO+ O =CH3.CH(OH).CH,.COOH Aldol. j8-oxybutyric acid. CHS.CH (OH) .CH2.COOH + O = (CH3.CO) .CHa.COOH + H2O j8-oxybutyric acid. Oxygen. Aceto-acetic acid. Water. By reduction aceto-acetic acid is reconverted into /3-oxybutyric acid. Other aldehydes can react in similar ways, and thus many of the other fatty acids can be formed. It may be added that acetone (another of the so-called acetone bodies which appear in the urine in diabetes mellitus) is easily obtained from aceto-acetic acid by the splitting off of carbon dioxide. Thus : (CH3.CO).CHa.COOH =CH3.CO.CH3 + CO2 Aceto-acetic acid. Acetone. Carbon dioxide. Formation of Fat — (2) From Protein. — Dry protein contains on the average 16 per cent, of nitrogen and 50 per cent, of carbon, and urea contains 46 per cent, of nitrogen and 20 per cent, of carbon. Urea is therefore three times as rich in nitrogen as the protein from which it is derived, but two and a half times poorer in carbon ; and less than one-seventh of the carbon of protein will be eliminated in a quantity of urea sufficient to carry off all the nitrogen. It is probable that a portion of the remaining carbon may, after passing through various stages, take its place as the carbon of fat. We have seen that certain amino-acids derived from proteins can be converted into dextrose, and that dextrose can be converted into fat. So that the mere question whether carbon atoms or carbon chains originally present in protein molecules are ever capable of appearing in fat molecules can be straightway answered in the affirmative. But it is still in doubt whether amino-acids can be transformed into glycerin or into fatty acids, or into both, by processes which do not involve the production of dextrose from them. And in any case proof is required that the extent of the transformation, let the steps be what they may, is great enough to be satisfactorily demonstrated. In regard to this point it must be said that absolutely flawless experiments to prove the direct production of fat from protein seem still to be wanting. Phosphorus Poisoning and Migration of Fat. — In the experiments of Bauer, the amount of oxygen consumed and of carbon dioxide and nitrogen excreted was determined in starving dogs. Phosphorus, which, as is well known, causes extensive fatty changes in the organs, was then administered in small doses for several days. The excretion of nitrogen was doubled, the excretion of carbon dioxide and the consumption of oxygen diminished to one-half. When the animals died, in a few days, the organs were all found loaded with fat. In one case 42-4 per cent, of the solids of the muscles and 30 per cent, of the solids of the liver consisted of fat. This is much more than the normal amount. It was assumed that the fat could not have been simply transferred from the adipose tissue, since the dog had been starved for twelve days before the phosphorus was given, and died on the twentieth day of starvation. Now, after such a period of hunger the amount of fat in the adipose tissue is greatly reduced. It was there- fore concluded that the source of the fat could cnly have been the broken-down pro-tern. Since the nitrogen excretion was increased, while the carbon excretion was diminished, it was supposed that a residue rich in carbon must have been split off from the proteins, and, remaining unburnt in the body, must have been converted into fat. Experiments of this kind are open to criticism on several grounds, but especially on this : that unless the fat-content of the whole body before the adminis- tration of the poison is known, it is impossible to be sure that the fat in a particular tissue has not been increased simply by the transportation of fat from some other tissue. It has been conclusively shown that migration of preformed fat does occur, and on an extensive scale, in phosphorus poisoning. For example, a dog was fed for a time with sheep's tallow, and fat was laid down in its adipose tissue with the physical and chemical characters, not of dog's, but of sheep's fat. The animal was then poisoned with phosphorus, and the fat which accumu- lated in the liver examined. It also resembled sheep's fat,, as it should have done had it migrated from the adipose tissue, and not dog's fat, as it might have been expected to do had it been formed in the hepatic cells from protein. The ease with which connective-tissue fat — i.e., food fat — migrates to the liver suggests, with other facts, that the liver has a special relation to the transformation of this fat into the fat of the organs. This ' organized ' intracellular fat differs in various ways from the fats of adipose tissue. Its ' iodine value ' (p. 4) is higher (Leathes), and a large proportion of it consists of phosphatide lipoids (p. 571.) The most convincing evidence that fat is not produced in increased amount under the influence of phosphorus has been obtained by deter- mining by actual analysis the total fat in animals, then poisoning similar animals with phosphorus and again estimating the total fat. Far from being increased, the fat may even be decreased in the poisoned animals (Taylor, etc.). There is no ground, then, for the assumption that phosphorus and other substances, like arsenic, antimony, etc., which bring about so-called ' fatty degeneration ' of the organs, act by causing or accelerating the transformation of protein into fat. Yet there is good evidence that they do accelerate the decomposition of protein, or at least interfere with its normal metabolism, for after phosphorus poison- ing amino-acids (leucin, tyrosin, glycin) appear in the urine. The observations of Lusk and his pupils indicate that phosphorus does not directly increase the amount of protein broken down, but does so indirectly, by favouring the conversion of the carbohydrate -like radicle of the protein molecule into leucin, tyrosin, and perhaps fat, and thereby necessitating an increased consumption of protein. A celebrated experiment, performed nearly forty years ago, was long 564 METABOLISM, NUTRITION AND DIETETICS supposed to furnish an absolute proof of the formation of fat from protein, under strictly physiological conditions, although in a humble form of animal life. Maggots were allowed to develop from the egg on blood containing a known amount of fat. The quantity of fat in the eggs was also known. After the maggots had grown, ten times as much fat was found in them as had been contained in the blood and eggs together. The trifling quantity of sugar in the blood was utterly inadequate to account for the fat, which, it was concluded, must there- fore have come from the proteins of the blood (Hofmann). It can be objected to this experiment that no precautions were taken to prevent the growth of micro-organisms on the blood, and fat might have been formed by them from the proteins. Further, the fat estimations would scarcely pass muster according to the present standards. The experiments of Pettenkofer and Voit, which were supposed to have demonstrated that in the higher animals also fat is formed from proteins under normal conditions, are in the same position. According to them, a dog fed for a time on a liberal diet of lean meat may go on excreting a quantity of nitrogen equal to that in the food, while there is a deficiency in the carbon given off. Or if the dog is not in nitrog- enous equilibrium (p. 602), but putting on nitrogen in the form of ' flesh,' the deficiency in the carbon given off may be too great in pro- portion to the nitrogen deficit to warrant the assumption that all the retained carbon has been put on in the form of protein. In either case, carbon in large amount can only come from the proteins of the food, and can only be stored up in the body in the form of fat. For lean meat contains but a trifling quantity of carbon in any other proximate principle than protein, and the non-protein carbon of the animal body is only to a very small extent contained in carbo-hydrates or other substances than fat. Pfliiger has criticized these experiments, and has shown that lean meat contains more fat than was supposed, and this is now generally admitted. He has endeavoured to show that the fat and glycogen in the meat given to the animals fully accounts for the carbon retained. Pfliiger, indeed, takes up the position that the fat of the body comes exclusively from the carbo-hydrates and fats of the food, and not at all from the proteins. But there is little doubt that in this he has gone too far, although his criticism has rendered it impossible any longer to appeal to Pettenkofer and Voit's results as good evidence on the other side. If none of the supposed quantitative proofs of the conversion of proteins into fat which have hitherto been brought forward are free from flaw, the same is true of the alleged qualitative indications of its possibility and of its actual occurrence. The accumulation of fat between the hepatic cells caused by phlorhizin is, at the best, no better evidence than the accumulation within the cells in phos- phorus poisoning. The formation of adipocere (a cheesy substance, rich in fatty acids united with calcium or ammonium), sometimes seen in dead bodies which have remained a long time under water or in moist graveyards, is largely, if not entirely, due to the fat akeady present in the parts which have undergone the change, or to fat removed by the water from other parts of the body. If any portion of the adipocere represents fat formed from protein, this transformation may well be credited to the numerous micro- organisms present, and throws no light upon the question of fat formation in the normal organism. The fat in the cells of the THE METABOLISM OF FAT 5^5 sebaceous glands, and of the mammary glands, may be produced from protein by a transformation of the cell-substance. But abso- lutely convincing proof is wanting. The old idea that the cells of these glands underwent a physiological process of transformation into fat analogous to the fatty degeneration of pathology, and then broke down bodily into the secretion, has been long since disproved for milk formation, and is probably erroneous also as regards the secretion of sebum. The rule which experience has taught, that a woman during lactation requires an excess of proteins in her food corresponding not only to the proteins, but also to the fat given off in the milk, suggests such an origin for the milk-fat, but does not prove it. Other fat-containing secretions are the ear-wax formed by glands in the wall of the external auditory meatus, and the smegma formed by the glands of the prepuce, but nothing is known of the sources from which the fatty substances are derived. The Intermediary Metabolism of Fat. — The mechanism and the stages of the transformation, including the migration, of fats is not well understood — indeed, not as well as that of the carbo- hydrates. Many of the tissues contain intracellular, soluble, fat-splitting ferments called Hpases, especially the liver, the active mammary gland, and the intestinal mucosa. We have already seen that there is evidence that these lipases, like some other enzymes, have a reversible action. They are either fat- splitting or fat-forming ferments, according to the conditions (Kastle and Loevenhart). It is stated that the perfectly aseptic blood does not split ordinary neutral fats, although it contains a ferment which splits up monobutyrin (glycerin butyrate) into glycerin and butyric acid. The question how the fat, after absorption from the intestine, passes from the blood into the cells, and how it is enabled again to pass out of the fat-cells when the needs of the tissues call for its mobilization, cannot at present be definitely answered. It is possible that just as fat is split in the lumen of the intestine before being absorbed, and then rebuilt in the epithelium, so it is split in the blood or in the lymph before being taken up by the fat-cells. The lipase in these cells would then be capable of synthetizing the glycerin and fatty acids to fat in their interior. When the fat is about to pass out of the cells in response to the call, of whatever nature it is, of the tissues for fat, it may again be split, resynthetized in the blood, and again hydrolysed for entrance into the tissue cells. Or it may be carried to the cells in the form of glycerin and fatty acids, or soaps, in such small concentration as to be harmless, and there built up again into the original fat, or transformed into other fats characteristic of the particular tissues, including the fatty acid components of the phosphatides, or utilized without synthesis into fat. An alternative hypothesis avoids this series of decomposi- 566 METABOLISM, NUTRITION AND DIETETICS tions and syntheses by assuming that the fat passes in the lorni of very fine droplets through the walls of the cells and of the capil- laries. The reader will observe that we seem to be discussing again, and almost in the same terms, the question of the absorption of fat from the intestine. It is indeed at bottom the same question, and it might be argued that by analogy it should receive the same solu- tion. Analogy, however, is a dangerous guide in such matters, and it is even more difficult to secure an unambiguous experimental test of the manner in which the internal migration of fat is accomplished than to secure the like for its absorption from the digestive tube. As to the ultimate fate of the fat, from whatever source it may be derived, our knowledge may be compressed into very few sen- tences : Sooner or later it is split and oxidized to carbon dioxide and water, its energy being converted into heat or, directly or indirectly, into mechanical or other functional work ; some of the fat absorbed from the intestine rapidly undergoes this change without entering the fat-cells of the adipose tissue. A portion of the fat may be changed into carbo- hydrt&s. This has been proved for the glycerin component ; its possi- bility must be admitted for the fatty acids, but proof has not yet been given. Of the intermediate stages by which the fatty acids are degraded into the simple end products but little is surely known. Included among these stages must be the compounds with which the forma- tion of the acetone bodies (p. 562) starts, if and in so far as their formation is a normal event which is merely unveiled by the dis- turbance of the ordinary course of the metabolism in diabetes. Among these intermediate stages must also be included, it is to be supposed, the compounds, whatever they may be, which act as connecting links between the currents of fatty acid and of carbo- hydrate metabolism, and with which the transformation of fatty acids into carbo-hydrates commences, if this occurs at all. According to the observations of Knoop, the saturated as well as some of the other series of fatty acids when oxidized decompose in a very characteristic way. As already remarked, these acids are made up of a larger or smaller number of CH8 groups forming a chain which at one end terminates with a carboxyl (COOH) group, and at the other with a CH3 group. The carbon atoms in the chain are designated by Greek letters, c, j8, etc., the a position being that next the carboxyl group, the /3 position one remove from the carboxyl group, and so on. Accord- ing to Knoop, the oxidation of the fatty acid chain takes place in such a way that the chain is shortened by the cutting off from the carboxyl end the a CH2 group along with the carboxyl group, while in place of the /3CH2 group there is left a carboxyl group, an operation which might be fancifully compared to the naval manoeuvre of breaking the enemy's ., CHo.CHo.CHa-CHo. CH..COOH, " line. Thus from caproic acid e £ y p a we get by oxidation butyric acid, CHs-CH2.CH,.O >H, carbon dioxide and water. It appears that the oxidation proceeds in two stages, the hydrogen of the /3 group being first oxidized with formation of an THE METABOLISM OF FA T 567 oxyacid oxycaproic acid, C^.CHa.CHg.CHOH.CHg.COOH, which is then by further oxidation converted, with loss of two carbon atoms, into butyric acid. The oxidation process may then start afresh on the /3 group of butyric acid. On the long carbon chains of the higher fatty acids this operation may be repeated again and again, the chain losing two atoms of carbon at each attack. If this represents what occurs in the normal metabolism, the groups cut off may then and there undergo the fate of the ships isolated by a successful application ofjthe manoeuvre alluded to, complete destruction — that is to say, oxidation to the end products carbon dioxide and water, a portion of the energy of the fatty acid being thus liberated at each oxidation of the /3 group. Eventually a fatty acid or acids with very few carbon atoms will be left. There is some reason to think that acetic acid (and perhaps similar simple acids) may be one of the normal stages in the decom- position. Thus, butyric acid may first yield by oxidation of the /3 group the oxyacid /3-oxybutyric acid, CH3.CHOH.CHa.COOH, which by further oxidation of the /3 group and the cutting off of the a and carboxyl groups would give CH3.COOH, or acetic acid. If this is the general course of the oxidation of the fatty acids in the body, it is to be assumed that numerous intermediate stages unrepresented in such a simple scheme may exist. Thus it is known, as has been mentioned more than once in other connections (p. 562), that /3-oxybutyric acid by oxidation yields aceto-acetic acid, by losing from the /3 group two atoms of hydrogen which unite with oxygen to form water. A molecule of aceto-acetic acid contains the elements of two molecules of acetic acid minus the elements of one molecule of water. It is therefore possible that aceto-acetic acid, if it is a normal stage in the katabolism of fatty acids, yields by its hydrolysis as a further step acetic acid, according to the equation CH3.CO.CH2.COOH + H2O=2(CH3.COOH). Aceto-acetic acid. Acetic acid. It is worth while, perhaps, to point out once more that even the relatively simple products now arrived at are not necessarily at once completely oxidized to their end products. That, it is to be assumed, will depend upon the needs of the organism. Acetic acid, for example, when added to blood and perfused through the sur- viving liver, can be transformed into aceto-acetic acid, and may thus become the starting-point of new syntheses. The Liver and Fats. — The liver seemstoplay an important part in the metabolism of fat, as it does in the metabolism of carbo-hydrates and of proteins. It contains an oxidizing ferment, /3-oxybutyrase (or j8-hydroxybutyrase) , which transforms /3-oxybutyric acid into aceto-acetic acid (Dakin). This oxidation appears to occur in the normal as well as in the diabetic organism. The liver seems also to possess the power of transforming aceto-acetic acid into acetone, a reaction which does not involve an oxidation, and this may also be accomplished by means of an enzyme. But it is not at all likely 5<>8 METABOLISM, NUTRITION AND DIETETICS that acetone forms a stage in the normal katabolism of the fatty acids or of the /3-oxyacids derived from them. The importance of the liver in the metabolism of fats is further indicated by the extent of the migration of fat to that organ when the fat stores are mobilized in unusual amount (p. 563). The reason for this migration seems to be that the fats undergo preparatory changes which facilitate their utilization by the tissues. For example, there is evidence that saturated fatty acids are changed in the liver into unsaturated acids, which are then carried to the organs to be metabolized. The desaturation may serve the purpose of facilitating the rupture of the long carbon chains, or their capacity for entering into reac- tions with other substances, at the points where double links exist between carbon atoms (p. 556). Non-Nutritive Functions of Fat. — In connection with the metabo- lism of fat, it ought to be noted that, in addition to their value as reserve material for the nutrition of the body, the deposits of fat under the skin and in other situations perform important functions in protecting delicate structures from mechanical injury, in facili- tating their movements upon each other, and in hindering the loss of heat. It would doubtless be a gross exaggeration to say that the mechanical and physical properties of the fat depots are as im- portant in comparison to their chemical relations as is the case for the bones and ligaments, but it would be an error not less gross to consider them as of little account. It will even, perhaps, be thought not unworthy of mention, from the point of view of the propagation of the race, that in the human species, at least, the amount and distribution of the cutaneous fat play a part of some consequence in the aggregate of qualities which determine the physical attractiveness of the individual, especially of the female, although the standard in this regard varies widely in different communities. It is perhaps partly because the fat depots have important mechanical functions that the fat reserve is far less mobile than the glycogen reserve. The semi-solid panniculus adiposus, the fatty tissue around the great nerve trunks, between the muscles, around the eyeball, on the soles of the feet, etc., possesses as a protective packing the good qualities of a water cushion with none of its dis- advantages. But if the fat-cells were subject to sudden depletion, as the hepatic cells are — nay, in still greater degree, since they contain hardly any protoplasm — they would never serve for such a function. Of course, in the emergency of starvation, when even the glands and the muscles themselves are wasting, the fat reserves are neces- sarily mobilized, let their mechanical functions suffer as they may. Obesity. — The proportion of the total mass of the body which is made tip of fat varies greatly in different individuals, and often in the same individual at different stages in life. When the accumulation of fat THE METABOLISM OF FAT 569 passes beyond a certain point it causes obvious changes in the contours of the body, and often some embarrassment in its movements. This condition is termed obesity. It is extremely difficult to say when the condition oversteps the physiological boundary and becomes actually pathological. Some individuals who are notoriously stout are noted also for their intellectual activity, and may not fall below the average even in the ordinary kinds of physical effort. It would be an exaggera- tion to speak of such persons as suffering from a disease. In other cases the pathological stamp is clearly imprinted upon the metabolic anomaly which leads to the overfilling of the fat depots. This is perhaps best illustrated in those cases of extreme obesity in children where, in spite of the intense metabolism associated with growth, with the restless muscular activity characteristic of that age, and with the relatively great surface through which heat is lost, great quantities of fat continue to be put on. Muscular activity by itself is no certain antidote to or • prophylactic against obesity, and it is a mistake to suppose that the condition is exceedingly rare among manual workers sufficiently well paid to be able to gratify their tastes in the quality and quantity of their food. Statistics or rough estimates covering the whole of the hand-workers of a country throw no light on such a question, for few indeed are the lands where the masses of the people have such well-filled purses that they are able to nourish themselves according to their wishes. While it is true that the great majority of normal individuals (although not all, since even in the fattening of stock for market some animals are rejected as bad feeders) can be compelled to lay on fat when overfed with fat and especially with carbo-hydrates, and prevented from taking much exercise or from losing heat freely, the most important factor in the excessive storing of fat by human beings leading a free life seems to be an anomaly in the metabolism which permits the machine to be run on less than the usual amount of fuel. From the point of view of thermo- dynamics the fat man, in very many instances at least, grows fat and fatter because his body is a machine whose ' efficiency ' is greater than the normal — that is to say, a machine which is capable of doing a given amount of work and of keeping itself in repair with a food intake of smaller heat value than is usually needed. Whether this anomaly is to be considered a metabolic fault or a metabolic virtue depends largely upon the ease with which the intake is adjusted to the actual require- ment of the body. If the adjustment is rendered accurate, the man with the anomalous tendency to put on fat, the adiposophil, ashe might be called, is in all probability just as well off in every physiolog ical sense on a smaller diet than a so-called normal individual of the same age, weight, and daily routine, on a larger quantity of food, and on this smaller diet he does not become fat. In this connection it may be recalled that, in speaking of the blood-flow in the hands and feet (p. 127), which are in this relation to be regarded as essentially an ' outcrop ' , of the cutaneous circulation, it was pointed out that some healthy persons have habitually small flows and a habitually cool skin which perspires little, in comparison with others living practically the sam e life; It was suggested that this difference in the blood-flow through the skin, which of course would correspond with a difference in the rate of heat loss, and therefore in the rate of heat production, may be correlated with a difference in the intensity of the metabolism and the intake of food. The difficulty of adjusting the appetite to the actual physiological requirement is perhaps the real anomaly in adiposophilia. Several factors seem to be involved in the group of sensations comprised under appetite and hunger (Chapter XVIII), and the onset and intensity of these sensations are unquestionably influenced by habit. The real 570 METABOLISM, NUTRITION AND DIETETICS question in many cases of obesity may be not why the metabolism is managed so parsimoniously — that is, in the physiological sense, so thriftily — but why the fat man or the man tending to become fat still experiences so strong a desire for food after he has eaten what in pro- portion to his metabolic wants is enough, whereas the man with no tendency to obesity is no longer hungry after he has eaten an amount of food sufficient for the requirements of his tissues. Is there here perhaps an anomaly in the nervous mechanism in virtue of which, for instance, the gastric hunger contractions are more readily initiated and less easily stilled than in the normal person ? It is recognized that in the usually much more serious anomaly of the carbo-hydrate metabolism, diabetes mellitus, the nervous element may be important. The influence of the loss of certain of the internal secretions on the deposit of fat will be alluded to in the next chapter. In the treatment of obesity the factor of appetite and hunger control has to be specially kept in mind. Bulky but comparatively innu- tritious food, such as green vegetables, e.g., in the form of salads, should form an important constituent of the dietary, since the mere distension of the stomach staves off hunger. The total heat value of the food must be reduced gradually. Carbo-hydrates must be largely excluded, and also fats, although a certain amount of fat, say in the form of butter, is permissible and even beneficial as aiding in the passage of the food along the digestive tube. Alcoholic beverages are in general contra-indicated, because alcohol, as an easily oxidizable substance, protects the carbo-hydrates and fats from oxidation, and perhaps also because the normal oxidative power of the tissues may be depressed by its habitual use. On the other hand, tobacco smoking, which has some power of inhibiting the gastric hunger contractions, may be permitted. Muscular exercise, cold baths, light clothing both during the day and at night, and a cool environment, are favourable to the reduction of fat by increasing the consumption of material and the loss of heat, just as a sedentary life in an overheated house in a person predisposed to obesity, and eating too much for his requirements, favours the putting on of fat. But if the appetite of the patient is allowed to govern the intake of food, the increased decomposition brought about by exercise, etc., is very likely to be balanced by an increased ingestion, and no progress will be made. Metabolism of Sterins or Sterols. — It has been previously stated that cholesterin appears to be the only representative of the sterins in the higher animals. Its source and function have been much discussed of late years. As to its source, there seems to be no reason to believe that any part of the cholesterin of the tissues is formed from decomposition products of ordinary fats, carbo- hydrates, or proteins. It is probably entirely derived from the cholesterins of animal, and the phytosterins of vegetable food. On this assumption, its metabolism, unlike that of the great groups of food substances, is carried on in a closed circuit. Evidence that it can be synthesized from other substances in the body is lacking. No increase in the cholesterin has been observed during the develop- ment of eggs, and the cholesterin content of growing chickens appears to correspond to the sterins taken in the food (Gardner). The portion of the cholesterin which is ingested in the form of esters THE METABOLISM OF FAT 571 is probably split, with liberation of the fatty acid, in the course of digestion. But if this be so, cholesterin esters are again formed in the tissues, for the cells and the blood contain both cholesterin esters and free cholesterin. While some cholesterin is excreted in the faeces (p. 423), there is evidence that a portion of the cholesterin of the bile may be reabsorbed, a ' circulation ' of cholesterin taking place analogous to the circulation of bile-salts. The appearance of cholesterin in the bile has been connected by some writers with the destruction of erythrocytes in the liver, or the conveyance of the products of their decomposition to that organ (p. 21), but there are no means of distinguishing between the cholesterin set free from blood-corpuscles and that liberated from other cells. Since it is contained in all cells, every cell may be supposed to contribute something from time to time to the cholesterin excretion. As to the office of the tissue-cholesterin, it can only be suggested that a substance so ubiquitous must be important. There is some evidence that cholesterin, free or combined, plays a part in con- ferring on the cells those peculiarities in their permeability upon which their functions, and indeed their integrity, depend. Free cholesterin, for instance, hinders the haemolytic action of the saponins (p. 28), apparently by forming compounds with them. Whether it or its esters are actually concentrated at the surface of the cell, and contribute to the formation there of the so-called ' lipoid ' envelope, is not definitely known, although there are facts in favour of this idea. Metabolism of Phosphatides. — The lecithins, which are the best- known members of this class of compounds, have been already described (p. 366). They are built up of glycerin, fatty acids, phosphoric acid in the form of glyceryl-phosphoric acid, and a nitrogenous base cholin. There is some reason to think that the lecithins of the tissues are, in part at least, not free, but combined with proteins or with carbo-hydrates. Other bodies belonging to the phosphatide group are kephalin, a constituent of nervous tissue and of yolk of egg, cuorin found in heart muscle, etc. It is probable, as stated in the chapter on Digestion, that the phosphatides of the food are hydrolysed in the alimentary canal with liberation of the glycerin, fatty acids, and the other com- ponents. It is not known whether they are resynthesized in the intestinal wall, but it is more probable that they pass directly to the tissues, where they can be utilized for building up the phos- " phatides of the cells. Cholin is found in small quantities free in the tissues, and also, it is said, in the blood-plasma. Glyceryl-phos- phoric acid has also been obtained in small amount from various tissues. The other components of lecithin are, of course, never wanting, and there can be no doubt that the cells possess the power of reconstructing phosphatides from such materials. They can do 572 METABOLISM, NUTRITION AND DIETETICS more than this: they can prepare the 'building-stones 'themselves. For even when the ingestion of phosphatides in the food is excluded, or the intake is so small as to be negligible, the formation of phos- phatides in the body goes on apparently without check. An instance of this will be given on a future page in discussing experiments on the relative value of different proteins for nutrition and growth. A very striking observation has been recorded by McCollom, who fed three hens on a diet almost free from fat. In about three and a half months they laid fifty-seven eggs, containing over 9 per cent, of phosphatides. Calculation showed that here, first of all, fats or their components must have been constructed from carbo-hydrates. Then the nitrogenous component of the phosphatides (cholin in the case of lecithin, at least) must have been obtained from some source, possibly from an amino-acid by the addition of methyl groups (CH3), of which cholin, OH.H2C.H2C -N<3 (trimethyl-oxyethyl- \ ^ OH ammonium hydroxide) contains three. SECTION III. — METABOLISM OF PROTEINS. Blood-Proteins. — The two chief proteins of the plasma, serum- globulin and serum-albumin,* must, as has been already pointed out, be recruited from proteins absorbed from the intestine and for the most part, at any rate, profoundly altered in its lumen and in their passage through the epithelium which lines it. Even when proteins are being actively absorbed, the plasma, after the blood-proteins have been separated, contains no substances which give the biuret reaction (p. 449). So that the peptones, which can be demonstrated in the intestinal contents, suffer great changes before or during their absorption. The physiological reasons for this alteration are in a measure known, and have already been alluded to in connection with the digestion of proteins. No doubt the far-reaching decom- position of the protein molecule may to some extent facilitate the absorption of protein food. No doubt also it is imperative that such comparatively slightly hydrolysed products as peptone, and particularly proteose, should not appear in quantity in the blood, for when injected they cause profound changes in that liquid, one expression of which is the loss of its power of coagulation, and are rapidly excreted by the kidneys, or separated out into the lymph. But the passage of the food from the stomach is so gradual an affair, the quantity of digesting protein present at one time in any loop of intestine is so small, and the rush of blood which irrigates the * It is probable that plasma contains a mixture of different albumins anf1 globulins. METABOLISM OF PROTEINS 573 active mucosa is so large, that the concentration of peptone or proteose necessary to produce injurious effects could hardly in any case be realized. Again, there is no evidence thai the simpler decomposition products of further hydrolysis are not in equal con- centration as poisonous as proteose and peptone. Apart from any influence which it may have in favouring absorp- tion, the complete shattering of the protein molecule has a double significance. In the first place, as already pointed out, the food- proteins cannot be used directly in the upbuilding and repair of the protoplasm (p. 448), since the tissue-proteins differ from them and from each other in the amount and nature of the amino-acids and other groups in their molecule (p. 2). Secondly, under ordinary dietetic conditions a surplus of nitrogen in the protein food has to be got rid of by being converted into urea without being built up into the tissue substance. Here we come upon the fundamental fact that the protein katabolism is not a single uniform process. Two forms may be distinguished which are essentially independent in course and character. One kind varies extremely in its quantita- tive relations, according to the amount of protein in the food. Its chief end-products are urea, representing the nitrogen, and inorganic sulphates, representing the sulphur of the proteins. Since this form of katabolism, as we shall see directly, is not essentially connected with the life and nutrition of the living substance, it is termed exogenous. The other variety is practically constant in amount for one and the same individual, and independent of the quantity of protein in the food. Its characteristic end-products are creatinin and neutral sulphur. This form of protein katabolism is essentially an expression of the waste of the living substance itself, and is therefore spoken of as endogenous. Some have supposed that the intestinal mucosa has as one of its special functions the resynthesis of a great part of the digestive decomposition products into the proteins of the blood-plasma. If this is the case, these proteins must be again decomposed in the cells of the various tissues in order that the ' building-stones ' may be recombined to form the tissue-proteins. For the proteins of the organs are not the same as those of the blood, and the proteins of different organs differ characteristically from each other. The significance of the synthetic function of the intestinal wall would then lie in this: that from the varying mixture of amino-acids, etc., derived from the food-proteins an always uniform and suitable' protein mixture (the blood-proteins) is fabricated for the feeding of the tissues. Experiments intended to test this hypothesis have hitherto yielded a negative result. No accumulation of protein in the wall either of the intestine in situ or of the isolated surviving intestine has been detected during absorption of the decomposition products of protein. An alternative assumption, and superficially at least a simpler one, is that no more extensive synthesis of proteins 574 METABOLISM, NUTRITION AND DIETETICS occurs in the wall of the alimentary canal than is necessary for the needs of the tissues composing it, and, perhaps in addition, for the maintenance of the normal composition of the plasma, and that the decomposition products of the proteins are mainly absorbed as such, and pass in the blood to the tissues for which they are destined. If this is the case, the blood-proteins can no longer be looked upon as representing the main current of protein supply for the organs, but rather the store of protein material proper to the circulating tissue blood itself, and which confers on it certain chemical and physio- chemical properties (e.g., the due degree of viscosity) necessary for its function. Slowly accumulated, under ordinary conditions, and slowly consumed, this protein store may, of course, be at the dis- posal of the organs in an emergency — for instance, in starvation — or may be rapidly recruited from the organ-proteins, as after haemorrhage, just as in prolonged hunger the proteins of skeletal muscle may be utilized to feed the heart. That the blood-proteins can serve as nutritive material for the cells without undergoing digestion in the alimentary canal is well shown by the observations of Carrell and Burrows on the growth of isolated tissues in a medium composed of clotted blood-plasma. But, as previously pointed out in another connection (p. 449), a portion, and probably a large portion, of the digested protein is absorbed from the intestine by the blood in the form of amino-acids. Considerable quantities of these compounds can be separated by dialysis from blood drawn off during the absorption of proteins or by the process of vivi- diffusion (p. 48) (Abel). Among these amino-acids, glycocoll, alanin, glutaminic acid, and leucin, have been identified. While the quantity of amino-acids in the blood, which is very small in the fasting animal, is decidedly increased during protein digestion, it is probable that even in starvation amino-acids derived from the decomposition of the body-proteins are not entirely lacking. The normal concentration of amino-acids in the general blood of man and of the dog is about o-i per cent. — i.e., about the same as that of dextrose, and it may be nearly twice as great in the portal blood of dogs after a heavy protein meal. The amino-acids are very rapidly absorbed by the tissues from the blood, and can be demon- strated in muscles and other organs as free amino-acids. They accumulate there in much greater concentration than that in which they exist in the blood. Accordingly, the tissues take them up from the blood by some other process than simple diffusion (Van Slyke) . It has been surmised that amino-acids constitute the form in which proteins are transported from tissue to tissue, as well as the form in which proteins are normally utilized by the cells.* Although this cannot be regarded as yet established, there is reason to believe that • Recent experiments of Abel tend to rehabilitate the old view that some protein is absorbed as proteoses, which can also be isolated from the tissues- METABOLISM OF PROTEINS 573 the amino-acids play a great part in protein metabolism, perhaps as great a part as the dextrose does in the metabolism of the carbo- hydrates. There is some evidence that serum-albumin is more directly related to the proteins of the food than serum- globulin. And it is said that during starvation the albumin is relatively diminished, and the globulin relatively increased. It is, of course, not at all improbable that the plasma-proteins have a double source — organ-proteins on the one hand, food-proteins on the other. In any case, it is certain that serum-albumin and serum-globulin cannot be interchangeable without far-reaching decomposition, for their composition is very different. The globulin, e.g., yields glyco- coll, but the albumin does not. That the plasma-protein mixture maintains a very constant composition in the face of wide variations in the composition of the food-protein is indicated by the following experiment : A horse fed mainly on hay and oats was bled to the amount of 6 litres, and in the total protein of the serum the content of tyrosin and glutaminic acid was determined. In order to eliminate the influence of remains of the food in the digestive canal, nothing was given to the animal for a week. Then 6 litres of blood were again removed, and the tyrosin and glutaminic acid in the serum-protein again estimated. The horse was now fed with gliadin (one of the prolamins or alcohol- soluble proteins obtained from flour), a substance which contains 36-5 per cent, glutaminic acid and 2-37 per cent, tyrosin — that is, about the same amount of tyrosin as the serum-protein, but about four times as much glutaminic acid. The serum-protein was again analyzed for the two amino-acids after this diet. The results of one experiment are shown in the table : Normal. After 8 Days' Hunger. After Feeding with 1,500 Grammes Gliadin. After Feeding again with 1,500 Grammes Gliadin. Tyrosin - Glutaminic acid 2-43 8-85 2 -60 8-20 2-24 7-88 2-52 8-25 No increase in the glutaminic acid content of the serum-protein occurred, although, owing to the loss of blood, much new serum-protein must have been formed. If the amino-acids of the gliadin were used without change to build up the new serum-protein, three-quarters of the glutaminic acid must have been superfluous, and the nitrogen of this portion may have been straightway changed into urea and excreted. But the possibility that the glutaminic acid, or a portion of it, may- have been changed into other amino-acids in the body cannot be excluded. In the case of some of the amino-acjds it has been shown that such a transformation occurs (p. 613) (Abderhalden). The high degree of independence of the food and body proteins is still more clearly exhibited in the table from Abderhalden on p. 576, in which the proteins of milk are compared with some of the proteins which must be formed from them in the body of the suckling. The numbers represent percentages of the weight of each protein. 576 METABOLISM, NUTRITION AND DIETETICS Living and Dead Proteins. — Carried to the tissues, the decomposition products of the food -proteins, or the regenerated proteins of the plasma, which in ordinary language are still to be regarded as dead material, are built up into the living protoplasm, at any rate to the extent neces- sary to make good its waste. In this form they sojourn for a time within the cells, and then they become dead material again. The nature of this tremendous transformation has, of course, been the subject of speculation, but the truth is that we do not understand wherein the difference between a living and a dead cell, between a living and a dead particle in one and the same cell, really consists. All we know is that now and again a protein molecule or an aggregate of such molecules incorporated in the colloid mass which constitutes the proto- plasm of a muscle-fibre, or a gland-cell, or a nerve-cell, must fall to pieces. Now and again a molecule of protein, hitherto dead (or perhaps, to speak moBe correctly, hitherto not a constituent of living protoplasm. Since protoplasm is certainly more than protein), or a molecule of a particular amino-acid, or perhaps a polypeptide group intermediate in complexity between amino-acid and protein, coming within the grasp of the molecular forces or chemical affinities of the living substance, is caught up by it, takes on its peculiar motions, acquires its special powers, and is, as we phrase it, made alive. Each cell has the power of selecting and, if necessary, further decomposing or further synthesizing the protein materials offered to it ; so that a particle of serum-albumin or a mixture of amino-acids may chance to take its place in a liver-cell and help to form bile, while an exactly similar particle or mixture may furnish constituents to an endothelial scale of a capillary and assist in forming lymph, or to a muscular fibre of the heart and help to drive on the blood, or to a spermatozoon and aid in transferring the peculiarities of the father to the offspring. And just as a tomb and a lighthouse, a palace and a church, may be, and have been, built with the same kind of material, or even in succession with the very same stones, so every organ builds up its own characteristic structure from the common quarry of the blood. i ! Milk- Albumin. Serum- Albumin. Serum- Globulin. O 1 E Histon (Thymus). _d 1 3 M Glycocoll - O O 0 3'5 O 3-o o-5 26-0 4'7 Alanin 0-9 2'5 2-7 2'2 4'2 3-6 3'5 6-6 1 "5 Valin I'O 0-9 some some I'O I'O 0-9 Leucin 10-5 19-5 20 -o 18-7 29-0 15-0 ii-8 21-4 7-1 Serin 0'2 — 0-6 0-6 some — — Cystin O'O7 — 2 '3 0-7 0-3 — — — 0-6 Asparaginic acid Glutaminic acid I '2 I I'O I'O IO'O 7-7 4'4 1-7 2-0 8-0 ~. 0-8 IO'O 3'7 Lysin 5-8 — — 4'3 — 6-9 — Arginin 4-8 — — — 5 '4 — *5'5 0-3 — Phenylalanin - 2'4 3-1 3-8 2'O 2 '2 3-9 — Tyrosin - 4'5 I'O 2'I 2*5 i «5 3'5 5*2 °'34 3'2 Prolin 4-0 I'O 2 '8 2*3 i'5 1-7 3-4 Histidin - 2-6 — — I I'O — METABOLISM OF PROTEINS 577 It is not any difference in the kind of protein offered them which determines the difference in structure and action between one organ and another. In this quarry alongside of the plasma proteins the tissue cells find what is probably more important for their individual nutrition, the building-stones of the shattered food-protein molecules. They are only under exceptional circumstances confronted with intact molecules of food -proteins. ' The body cells do not know what the kind of food was ' (Abderhalden). In the case of the more highly developed tissues, at least, no mere change of food will radically alter the structure of the cells, nor even, as we have seen, the composition of the tissue proteins. A cell may be fed with different kinds of food, it may be overfed, it may be ill-fed, it may be starved; but its essential peculiarities remain as long as it continues to live. What may be called its organization, per- haps at bottom a more or less metaphorical expression for its essential physico-chemical make-up, dominates its nutrition and function. ' We must assume that many of the enigmatical properties of living matter depend upon the activity of intact protein molecules. We can obtain some idea of the possible variety in "the combinations of the '* building-stones " of the proteins by recalling the fact that they are as numerous as the letters of the alphabet, which are capable of expressing an infinite number of thoughts. Every peculiarity of species and every occurrence affecting the individual may be indicated by special com- binations of the "building-stones " — that is to say, by specific proteins. Consequently we may readily understand how peculiarity of species may find expression in the chemical nature of the proteins constituting living matter, and how they may be transmitted through the material contained in the generative cells ' (Kossel). Add to the great variety of compounds rendered possible by the enormous number of permuta- tions and combinations of the protein ' building-stones,'* the still greater variety rendered possible by the fact that the quantitative relations of given amino-acids may vary greatly in different proteins, and it will be seen what a practically infinite power of functional adjustment and reaction, correlated with a practically infinite variety of chemical changes in the midst of which the cell still preserves its specificity through and through, may be conferred upon the living substance by its content of protein. Some have supposed that the protein of the living substance is es- sentially different from dead protein, especially in possessing a character- istic instability, a prodigious power of dissociation and reconstruction. All the older theories which attempted to explain this alleged difference require revision in accordance with the newer chemistry of proteins, and speculations on the subject are probably in any case premature till the constitution of the proteins is thoroughly understood. In the meantime it is enough to say that the velocity of the reactions into which the proteins of living protoplasm or their constituent amino- acids may enter must depend upon intracellular conditions, which may vary rapidly and within wide limits. For example, enzymes may be present in greater or smaller concentration, or be activated and aided more or less powerfully by other substances, or by a more or less favour- able chemical reaction of the medium. The protein itself, too, or such part of it as is ready for decomposition, may exist in a physical con- dition now more and again less favourable to the attack of the enzymes. * Twenty different amino-acids, each used only once, but in a different order, would be capable of forming about 2,000,000,000,000,000,000 (two thousand million times a thousand millions) of different polypeptides, all containing the various amino-acids in the same proportions (Abderhalden) . 578 METABOLISM, NUTRITION AND DIETETICS It may not be superfluous at this point to again warn the reader that protoplasm and tissue -profeins are by no means synonymous. The physical, physico-chemical, and chemical changes involved in the katabolism of the colloid aggregates, including water, salts, phosphatides, sterins, and probably fats and dextrose as well as proteins, to which the term protoplasm is applied, may be many and complex before the individual proteins known to the chemist come face to face in the interior of the cells with the ferments which decompose them. On the other hand, it has not been proved that in the katabolic processes of the living substance isolated proteins ever form a stage. It may well be that without the complete decomposition of the protein molecules, or even without their complete detachment from the protoplasm, indi- vidual amino-acids or mixtures of amino-acids, or polypeptide groups, are cut out of the protoplasmic mass. It is now necessary to follow, as far as is possible, the steps in the degradation of the body-proteins. Since there is reason to believe that these, like the food-proteins, are first split up into the amino-acids from which they were originally synthesized before undergoing further decomposition, a study of protein metabolism is to a great extent a study of the metabolism of amino-acids. In this study it is for most purposes impracticable, even if it were desirable, to distinguish between amino-acids directly derived from the food, and which have not yet been, and may never be, built up into tissue-proteins, and those derived from the tissue proteins. There is nothing to indicate that the fate of a given amino-acid, once it has reached the blood, depends in the least upon its source. It may be said at once that the katabolism of the amino-acids is not a single and uniform process, one step in which inevitably follows another till the final end-products are reached. On the contrary, certain of the stages may become the starting-points of syntheses, which may lead back to the original or to another protein, or it may be to sugar or to fat. The extent of such synthesis, and even in some degree the stage from which it starts, may be assumed to depend upon the needs of the tissues and the relative abundance of protein and of other foods. Formation of Amino-Acids from Tissue-Proteins. — That amino- acids are formed in the metabolism of the cells and by the action of intracellular enzymes is indicated by the fact that proteolytic en- zymes (proteases) are invariably present in the tissues, and can be obtained from them by appropriate methods — e.g., by subjecting the organ in a finely divided state to a high pressure and collecting the expressed juice. Not only do unicellular organisms, like leuco- cytes, yeast cells, and bacteria, which must naturally depend upon themselves alone for all enzymatic reactions, yield ferments which have the power of splitting proteins, peptones, and polypeptides into amino-acids, but their existence has been demonstrated in practically all the organs of the higher animals and man. When a piece of liver, e.g., is removed with aseptic precautions and kept at 579 body-temperature, extensive auto-digestion occurs, and ammonia and other basic substances, glycin, and tryptophane, appear among the products. Tyrosin appears so early that it is scarcely possible to doubt that it must be a product of protein decomposition in the liver-cells under normal conditions — a decomposition which could be observed also in the organ in situ were the circumstances as favourable. The circumstances are less favourable in an organ whose circulation is going on, because the amino-acids are removed by the blood as they are formed. Further, it is to be assumed that the regulation of the ferment action, which is a characteristic property of the normal cell, becomes feebler the longer it is with- drawn from normal conditions. Similar autolytic processes have been observed in the spleen, muscle, lymph-glands, kidneys, lungs, stomach wall (independently of pepsin), thymus, and placenta; also in pathological new growths like carcinoma, in the breaking down of which and in the removal of such exudations as occur in the alveoli in pneumonia, these proteolytic ferments seem to play a part. It is to be assumed that the syntheses of the proteins or their products, which are scarcely less characteristic of the tissue cells than the decompositions effected by them, are also due to the action of separate intracellular ferments or upon the reversed activity of the proteolytic ferments. More direct proofs of the production of amino-acids in the tissues are not lacking. A rare condition known as cystinuria has been alluded to on a previous page (p. 488). Here there is a continuous excretion of the sulphur-containing amino-acid cystin in the urine. Sometimes the cystin is accompanied by other amino-acids, as leucin and tyrosin, a condition which might be called amino- aciduria. Cystinuria, while of course resulting from a gross anomaly in metabolism, is of little clinical importance unless the sparingly soluble cystin should form calculi somewhere in the urinary tract. The most plausible explanation of the condition is that in the normal course of the metabolism each of the ' building- stones ' of the proteins is sooner or later further decomposed by special fer- ments, and that for some reason the ferment which acts on cystin is absent, or if it is still produced, the conditions are more or less unfavourable to its action. Unable to take its place in the meta- bolic current, except in so far as it can be utilized to form taurin, and therefore taurocholic acid — for this property it has not lost — cystin becomes a chemical outcast in the body of the cystinuric individual, and is got rid of by the kidneys as an ' unemployable.' By comparing the amount of cystin excreted with the amount ingested in the food-proteins and with the (undiminished) amount contained in the tissues (especially the hair and the nails, since keratin is exceptionally rich in cystin), it has been shown that a portion of the cystin in the urine must have come from the tissues 5So METABOLISM, NUTRITION AND DIETETICS (Abderhalden). Observations on animals in prolonged starvation afford additional evidence. The hair and nails continue to grow and to maintain the high cystin content, and taurocholic acid con- tinues to be excreted. In like manner glycin continues to be pro- duced and to unite with cholic acid to form the glycocholic acid of the bile. There are other and more striking proofs that glycin can be formed in the body in large amount. For example, as already stated (p. 481), when benzoic acid is ingested it is not excreted as such in the urine, but coupled with glycocoll as hippuric acid. Thus — C6H5.COOH + CH2(NH2).COOH - H2O =C6H5.CO.NH.CHa.COOH Benzoic acid. Glycocoll. Hippuric acid. Benzoic acid, therefore, meets glycin in the body, and combines with it, as fatty acids meet glycerin and combine with it. Even starving animals fed with benzoic acid excrete large quantities of hippuric acid. Yet their tissues, as shown by analysis after death, yield as much glycocoll as starving animals which have received no benzoic acid, and excreted little or no hippuric acid. Many other acids which are totally foreign to the body are, when ingested, paired in the same way with glycocoll and excreted in the urine. Even substances whose chemical nature does not permit of a direct union with the glycin are often altered by oxidation or reduction till they can unite with it, and then the coupling takes place, and the conjugated acid is eliminated by the kidneys. The paired or aromatic sulphuric acid which we have already recognized as a normal constituent of the urine affords another instance of this coupling. Cystein among the derivatives of proteins, and glycu- ronic acid (p. 482) among the derivatives of carbo-hydrates, can also unite in the same way with numerous compounds. There is some evidence that the physiological significance of this process is that the toxicity of the foreign substances, or, as in the case of the aromatic sulphuric acid of the urine, of substances formed by bacteria in the intestine, or even produced in the metabolism of the tissues, is diminished by the pairing. The place and manner of formation of hippuric acid have been investigated with the following result : If an excised kidney is per- fused with blood containing benzoic acid, or, better, benzoic acid and glycin, hippuric acid is formed. Oxygen is required, for if the blood is saturated with carbon monoxide, or if serum is employed for perfusion, the synthesis does not take place. The kidney cells must be intact, for if a mixture of blood, glycin, and benzoic acid be added to a minced kidney immediately after its removal from the body, hippuric acid is produced, but not if the kidney has been crushed in a mortar. Nevertheless, there is some evidence that a METABOLISM OF PROTEINS 581 ferment is concerned, and the known mechanism of similar reactions in the body scarcely permits the physiologist to acquiesce in any other explanation. It must not be forgotten that the urinary constituents which must come into contact with the ferment when the kidney is crushed may injure or inhibit the enzyme. In herbivora hippuric acid cannot normally be detected in the blood; it is present in large quantities in. the urine ; it must therefore be manufactured in the kidney, not merely separated by it. In certain animals, as the dog, the kidney is the sole seat of the production ol hippuric acid. But in the rabbit and the frog some of it must also be formed in other tissues, for after extirpation of the kidneys the administration of benzoic acid causes hippuric acid to appear in the blood. It has, indeed, been recently shown that when the rabbit's liver is perfused with blood containing benzoic acid, hip- puric acid is produced. The benzoic acid required for the normal excretion of hippuric acid comes mainly from substances of the aromatic group contained in vegetable food, but a small amount is produced in the body, since hippuric acid does not entirely dis- appear from the urine in starvation. The differences which may exist in the metabolism of different groups of animals is well illustrated by the fact that in birds, when benzoic acid is given in the food, it unites not with glycin, but with ornithin, a derivative of arginin, forming not hippuric acid, but ornithuric acid (dibenzoyl-ornithin) . A much more important instance of such a difference will be seen when we come to consider the formation of urea and uric acid. The method by which the presence and the production of glyco- coll in the body are demonstrated by coupling it with benzoic acid, and so saving it from decomposition and bringing it to excretion, can also be applied to other amino-acids. If instead of fishing with the bait benzoic acid we fish with a bait caUed brom-benzol or bromo-benzene (C6H5.Br), a substance derived from benzol by the substitution of an atom of bromine for an atom of hydrogen, we capture the amino-acid cystein in the form of a compound called mercapturic acid, produced by the union of brom-benzol with cystein and acetic acid, with oxidation and loss of water. In other words, when this substance is administered, mercapturic acid is excreted in the urine, the cystein, which is very unstable and readily changes into cystin (p. 360), being thus preserved from decom- position. Another instance in which amino-acids (tyrosin and phenylalanin), which would normally be decomposed and so escape detection, come to the surface by being excreted in the urine, has a'ready been alluded to in connection with alkaptonuria (p. 483). In this condition, which seems to have no serious significance so far as the well-being of the patient is concerned, not only does the taking of food containing the aforesaid amino-acids lead to an 582 METABOLISM, NUTRITION AND DIETETICS increased excretion of homogentisinic acid, but even in starvation this substance still continues to appear in the urine. Since homo- gentisinic acid is undoubtedly formed from tyrosin and from phenyl- alanin, this observation constitutes a convincing proof that these amino-acids are produced from tissue-proteins. A striking illus- tration of the fact that the amino-acids of the tissues are not simply a store of reserve material derived directly from the alimentary canal is the failure of a period of starvation to make any impression upon their amount. They normally constitute 2 to 4 per cent, of the dry weight of the tissues, and if anything, tend somewhat to increase in the tissues of a starving animal. Fate of Amino- Acids in the Body. — The problem of the kata- bolism of proteins is thus reduced to the question, What becomes of the amino-acids ? Where, how, by what stages, and to what end- products are they decomposed ? Some, possible or probable steps in their metabolic history have been already suggested in dealing with the intermediary metabolism of carbohydrates (p. 542). Something more will have to be said of the where and the how of their chemical degradation in treating of the place and manner of urea formation. As to the end-products by which they are repre- sented in the final balance-sheet of the bodily economy, the answer is easy. The amino-acids, whatever intermediate stages they may pass through, whatever cleavages, oxidations, or reductions they may undergo, yield eventually carbon dioxide, water, and comparatively simple nitrogen-containing substances, which after further changes appear in the urine principally as urea, and in birds and reptiles as uric acid. When amino-acids are fed to mammals or introduced parenterally, a very large proportion of the nitrogen appears in the urine as urea. The same is true when, instead of simple amino- acids, polypeptides, like glycyl-glycin, alanyl-alanin, or leucyl-leucin, are given. When amino-acids are administered to birds, the great bulk of the nitrogen is excreted in the form of uric acid. Whether in mammals, and if so to what extent, uric acid is also one of the nitrogenous end-products of the decomposition of ordinary proteins or of the amino-acids which they yield, are moot questions. In any case, the most important and characteristic source of the uric acid in mammals and the other groups of animals whose chief nitrogenous end-product is urea, is not the ordinary proteins, but the nucleins which form constituents of the nucleo-proteins. We have no definite information as to the production of water from the hydrogen of the tissues, except what can be theoretically deduced from the statistics of nutrition (p. 620). A few words will be said a little farther on about the production of carbon dioxide from proteins; we have now to consider the seat and manner of formation of the nitrogenous metabolites. And since in man and the other mammals urea contains, under ordinary conditions, by far the METABOLISM OF PROTEINS 583 greater part of the excreted nitrogen, it will be well to take it first. Formation of Urea. — The starting-point of all inquiries as to the place of formation of urea is the fact that it occurs in the blood in small amount (4 to 6 parts per 10,000 in man; 3 to 15 parts per 10,000 in the dog), the largest quantity being found when the food contains most protein and at the height of digestion, the smallest quantity in hunger (Schondorff). Evidently, then, some, at least, of the urea excreted in the urine may be simply separated by the kidney from the blood; and analysis shows that this is actually the case, for the blood of the renal vein is poorer in urea than that of the renal artery, containing only one-third to one-half as much. If we knew the exact quantity of blood passing through the kidneys of an animal in twenty-four hours, and the average difference in the percentage of urea in the blood coming to and leaving them, we should at once be able to decide whether the whole of fhe urea in the urine reaches the kidneys ready made, or whether a portion of it is formed by the renal tissue. Although data of this kind are as yet inexact and incomplete, it is not difficult to see that all, or most of, the urea may be simply separated by the kidney. If we take the weight of the kidneys of a dog of 35 kilos at 160 grammes (^Oth of the body-weight is the mean result of a great number of observations in man), and the average quantity of blood in them at rather less than one-fourth of their weight, or 35 grammes, and con- sider that this quantity of blood passes through them in the average time required to complete the circulation from renal artery to renal vein, or, say, ten seconds, we get about 300 kilos of blood as the flow through the kidneys in twenty-four hours. Even at 0-3 per 1,000, the urea in 300 kilos of blood would amount to 90 grammes. Now, Voit found that a dog of 35 kilos body-weight, on the minimum protein diet (450 to 500 grammes of lean meat per day) which sufficed to maintain its weight, excreted 35 to 40 grammes of urea in the twenty-four hours. If, then, the renal epithelium separated somewhat less than half of the 90 grammes urea offered to it in the circulating blood, the whole excre- tion in the urine could be accounted for, and the blood of the renal vein would still contain more than half as much urea as that of the renal artery. So that the whole of the urea in the urine may be simply separated by the kidney from the ready-made urea of the blood. Another line of evidence leads to the same conclusion : that the kidney is, at all events, not an important seat of urea-formation. When both renal arteries are tied, or both kidneys extirpated, in a dog, urea accumulates in the blood and tissues; and, upon the whole, as much urea is formed during the first twenty-four hours of the short period of life which remains to the animal as would under normal circumstances have been excreted in the urine. Where, then, is urea chiefly formed ? The answer to this question is that, while some urea is probably produced from amino-acids in all the tissues, one organ is particularly associated with this function — namely, the liver. 584 METABOLISM, NUTRITION AND DIETETICS There is no reason to suppose that the hepatic cells, so far as the repair of their own protoplasm or the supply of energy for their own special work is concerned, require to metabolize particularly large quantities of amino-acids as compared, for instance, with the muscles. Glycin, however, they must have for the manufacture of glycocholic.andcystin for the manufacture of thetaurin of taurocholic acid. In addition, the liver is known to possess the power of utilizing amino-acids for the formation of dextrose and eventually of glyco- gen, and a portion of the surplus amino-acids of the food may be withdrawn from the blood of the portal vein for this purpose, just as the surplus of dextrose is withdrawn. The liver contains a relatively large amount of urea, and there is strong evidence that it is the manufactory in which a great part of the nitrogenous relics of broken-down proteins, including amino- acids, reach the final stage of urea. This evidence may be summed up as follows: (1) An excised ' surviving ' liver forms urea from ammonium carbonate mixed with the blood passed through its vessels, while no urea is formed when blood containing ammonium carbonate is sent through the kidney or through muscles. Other salts of am- monium, such as the lactate, the formate, and the carbonate, under- go a like transformation in the liver. It is difficult, in the light of this experiment, to resist the conclusion that the increase in the excretion of urea in man, when salts of ammonia are taken by the mouth, is due to a similar action of the hepatic cells. (2) If blood from a dog killed during digestion is perfused through an excised liver, some urea is formed, which cannot be simply washed out of the liver-cells, because when the blood of a fasting animal is treated in the same way there is no apparent formation of urea (v. Schroeder). This suggests that during digestion certain substances which the liver is capable of changing into urea enter the blood in such amount that a surplus remains for, a time unaltered. These substances may come directly from the intestine ; or they may be products of general metabolism, which is increased while digestion is going on ; or they may arise both in the intestine and in the tissues. Leucin — which, as we have seen, is constantly, or, at least, very frequently, present in the intestine during digestion — can certainly be changed into urea in the body. So can other amino-acids of the fatty series, like glycocoll or glycin, and aspartic acid, and it has been shown by perfusion experiments that this change can take place in the liver. Further, the blood of the portal vein during digestion contains several times as much ammonia as the arterial blood, and the excess disappears in the liver. (3) During digestion the blood loses a greater proportion of its amino-nitrogen (amino-acids) in passing through the liver than in passing through other organs, as shown by the fact that the excess METABOLISM OF PROTEINS 585 in the portal blood, as compared with the blood of the hepatic vein, is greater than the difference between the arterial and the vena cava blood. Now it has been proved that the liver does not store amino-acids to an appreciable extent as such, and therefore it must either have destroyed or condensed them into proteins. There is no evidence that the liver forms a store of reserve protein from amino- acids as it does of glycogen from carbohydrates, although this is possible. But there is good reason to believe that a portion at least of the amino-acids taken up by the liver is quickly broken down, and- that urea is formed from them. For when amino-acids were injected into a vein in such amount that the content of amino-acids in all the tissues was considerably increased, they disappeared far more rapidly from the liver than from muscle or kidney, and their disappearance from the liver was accompanied by an increase in the urea content of the blood (Van Slyke) . (4) Uric acid — which in birds is the chief end-product of protein metabolism, as urea is in mammals — is formed in the goose largely, and almost exclusively, in the liver. This has been most clearly shown by the experiments of Minkowski, who took advantage of the communication between the portal and renal-portal veins (p. 385) to extirpate the liver in geese. When the portal is ligated the blood from the alimentary canal can still pass by the round- about road of the kidney to the inferior cava, and the animals survive for six to twenty hours. While in the normal goose 50 to 60 per cent, of the total nitrogen is eliminated as uric acid, in the urine, and only 9 to 18 per cent, as ammonia, in the operated goose uric acid represents only 3 to 6 per cent, of the total nitrogen, and ammonia 50 to 60 per cent. A quantity of lactic acid equivalent to the ammonia appears in the urine of the operated animal, none at all in the urine of the normal bird. The small amount of urea in the normal urine of the goose is not affected by extirpation of the liver. And while urea, when injected into the blood, is in the normal goose excreted as uric acid, it is in the animal that has lost its liver eliminated in the urine unchanged. (5) After removal of the liver in frogs, or in dogs which have survived the previous connection of the portal vein with the inferior vena cava by an Eck's fistula (p. 385), the quantity of urea excreted is markedly diminished, and the ammonium salts in the urine are increased. When the Eck's fistula is established and the portal vein tied, without any further interference with the hepatic circula- tion, the amount of urea in the urine is not lessened to nearly the same extent, evidently because the substances from which urea is formed still, for the most part, gain access to the liver through the hepatic artery and by means of the back-flow which is known to take place through the hepatic vein. Yet while in normal dogs the proportion of ammonia to urea in the urine is only 1 :22 to i : 73, 586 METABOLISM, NUTRITION AND DIETETICS in dogs with Eck's fistula it rises to i : 8 to 1 : 33. If the animals are kept on a diet poor in proteins, no symptcms may develop for a considerable time. But if much protein is given, characteristic symptoms, including convulsions, always appear. These may be produced by the saturation of the organism with ammonia com- pounds, which are formed from the proteins as in the normal animal, but which the liver, with its circulation crippled, is unable to cope with, and to completely change into urea, although the statement has been made that when ammonia or ammonium salts are injected into the blood larger quantities must be present to produce these symptoms than are found in animals with the Eck's fistula. Although the portal vein carries to the liver a greater supply of blood than the hepatic artery (about twice as much, according to Opitz and Macleod and Pearce), ligation of the latter causes a greater diminution in the ratio of the amount of urea to the total nitrogen in the urine than ligation of the former. While the blood of the hepatic artery is, of course, nearly saturated with oxygen, that of the portal vein is not half saturated, so that the total quan- tity of oxygen transported to the liver by the hepatic artery is actually greater than the quantity transported by the portal vein. This indicates that a good supply of oxygen is an important factor in the formation of urea in the liver (Doyen and Dufourt). But this is no proof that the process by which it is formed is an oxidation. The work of the liver, like that of other tissues, is no doubt deranged by lack of oxygen. (6) In acute yellow atrophy, and in extensive fatty degeneration of the liver, urea may almost disappear from the urine, and leucin, tyrosin, and other amino-acids may appear in it along with a much larger amount of ammonia than normal. Here it may be supposed that the amino-acids and ammonia formed in the intestine in the digestion and absorption of proteins, perhaps also amino-groups formed in the tissues which would normally be culled from the blood by the hepatic cells for the manufacture of urea, pass unchanged through the degenerated liver, and are excreted by the kidney. It would, however, be very easy to overdo this argument ; for it is sometimes observed that in pathological and experimental condi- tions in which the liver has suffered severely considerable quantities of urea continue to be excreted. Urea does not entirely cease to be produced even when the liver is removed; and it must again be pointed out that there is reason to believe that the formation of urea is not a function peculiar to the liver, but one shared prob- ably with all tissues. The liver certainly does not arrest the whole of the amino-acids coming from the alimentary canal ; for the non- protein nitrogen in the muscles is distinctly increased during the absorption of amino-acids, and the muscular tissue, even when freed from blood, contains some urea, which in all probability is formed METABOLISM OF PROTEINS 587 there in the decomposition of amino-acids. Some writers, indeed, take the view that the muscles, containing as they do three-fourths of the proteins of the body, and utilizing as they appear to do a large proportion of the amino-acids of the food-protein, are more important seats of urea formation than the liver. Yet the fact that it is far easier to demonstrate this power — e.g., by perfusion experi- ments— for the liver than for such tissues as the muscles, renders it difficult to avoid the conclusion that in the preparation of an end- product so important as that in which the great bulk of the nitrogen leaves the body, a certain degree of specialization has been developed, and that this preparation has been largely entrusted to a special organ. And, while it may be true that larger amounts of amino-acids are taken up and utilized by the muscles than by the liver under certain conditions, this does not show that amino-groups removed from the amino-acids in the muscles may not be largely transferred to the liver before being changed into urea. Further, the transforma- tion of amino-acids into dextrose (and glycogen) may be assumed to entail a considerable absorption of amino-acids by the hepatic cells. Processes by which Urea is formed. — In the case of only one of the amino-acids derived from proteins can urea be obtained by a simple process of hydrolytic cleavage. This is arginin (o-amino-5-guanidin- w-valerianic acid) — that is to say, normal valerianic acid, CH3.CH2.CHa.CHa.COOH, i y ft a in which an amino group is attached to the o carbon atdm, while guanidin NrT2^C.NH is attached to the S carbon atom (p. 566)- When arginin is hydrolysed by barium hydroxide it yields urea and ornithin (diamino-valerianic acid), half of the nitrogen of the arginin appearing in each. Thus, NHa aO > Arginin. NHa .CH.C =O+ NH2.CHa.CHa.CHa.CH.COOH UreaT Ornithin. The amount of arginin, and therefore the amount of urea which can be artificially obtained in this way, varies extremely with the different proteins. Thus, salmin, a protamin (p. 2), prepared from the milt of salmon, yields 84-3 per cent, of its weight of arginin, while the casein of cow's milk yields only 4-8 per cent., and gluten-fibrin, one of the proteins of wheat, only 3 per cent. In the body the hydrolysis of arginin to urea and ornithin is accomplished by the ferment arginase (Kossel and Dakin). This ferment is found in the liver, and also in many other organs. The urea formed in this way appears very rapidly in the urine. The ornithin itself is then more slowly transformed into urea. Since the ordinary food-proteins are poor in arginin, the 588 METABOLISM, NUTRITION AND DIETETICS amount of urea which can possibly be formed in mammalian metabolism by this process cannot be large, even if most of the arginin, as is the case when it is fed to an animal, is transformed into urea. There is no reason to suppose that urea can be directly split off from the other amino-acids with which we are concerned. A comparison of their constitutional formulae with that of urea (or with that of uric acid) shows that a more far-reaching decomposition must take place before products are obtained from which urea (or uric acid) can be formed. Urea has been artificially obtained from protein by oxidation with an ammoniacal solution of permanganate at body-temperature. When the protein is first split into its cleavage products, and these are then oxidized, a very large amount of urea is produced — e.g., as much as 3 grammes of urea from 10 grammes of glycin. While these facts suggest possible ways of formation of urea in the body, we cannot assume that what happens in the test-tube must happen in the tissues. The best evidence is to the effect that in the body the removal of the amino-group (NH8) in the form of ammonia from the amino-acids is the essential step in the formation of at least a great part of the urea, which is then synthesized from ammonia and carbonic acid . The possibility exists that this deamini zation (or deamidi- zation) of the amino-bodies is the result of hydrolysis, or of oxidation, or of reduction, or of a combination of these processes. Evidence has been found that in the case of some of the amino-acids the deaminization is associated not with hydrolysis in which hydroxyl is substituted for the NH2 group, but with oxidation (Neubauer) . But this has not been shown to hold good for all amino-acids. In either case, however, whether the deaminization process is oxidative or hydrolytic the nitrogen is split off as ammonia, and it is to such ammonium compounds as have been already mentioned as being transformed into urea when circulated through an excised liver (p. 584) that we have to look for the source of, at any rate, a large portion, of the urea. Ammonia in the form of carbonate or carbamate is constantly found in the blood (p. 585). The excess of ammonia in the portal blood, which, however, is not admitted by all observers to be very large or very constant, has been interpreted as indicating that a considerable decomposition of iamino-acids with liberation of the amino-groups occurs in the intestinal lumen or the intestinal wall. It is not established beyond doubt that ammonia is itself present in the protein molecule, or that its liberation in the hydrolysis of proteins can take place except at the expense of further decomposition of amino-bodies. It has been shown, however, that a great part of the ammonia in the blood is produced in the decomposition of protein in the digestive tube by putrefactive bacteria (Folin and Denis). This is a necessary part of the reaction by which phenol and indol are formed in the intestine. It has been generally taught that the deaminization of the surplus of amino-bodies takes place chiefly in the liver, the extra nitrogen being thus ' shunted ' out of the blood-stream before it has had a chance to reach the tissues. It would seem more advantageous in the light of our present knowledge that a large and, so to say, a miscellaneous assortment of amino-bodies should be placed at the disposal of the tissues to facilitate the selection of those which are indispensable. We have seen that tissues such as muscle can and do take up amino- acids when protein is digested in the intestine, and it is very probable that they take up not merely the relatively small amount necessary to replace their wear and tear, but also a portion of the surplus, which after deaminization in the cells takes its place as a source of energy to drive the machine. The nitrogen in the form of ammonia may pass METABOLISM OF PROTEINS 589 back into the blood, and may thus be carried to the liver for conversion into urea. It is not necessary, however, to suppose that all of the nitrogen must perforce make this journey before being changed into urea. There is evidence that all the tissues share to some extent with the liver the power of forming urea just as they share with the liver the power of splitting off NH2 from the amino-bodies. It may be that the liver surpasses other tissues in its deamidizing power just as it seems to surpass other tissues in its power of transforming ammonium com- pounds into urea. But this does not prevent at least a considerable proportion of the amino-substances absorbed from the intestine from passing into the general circulation. It is of importance to remark that such hydrolytic cleavages as are associated with the splitting of protein into amino-acids, etc., only slightly reduce the available energy of the compounds. In so far as the liberation of the nitrogen from the amino-acids is also accomplished by hydrolytic cleavage (sup- posedly by a ferment desaminase), the residue, relatively rich in carbon, will still be available for yielding to the body by its oxidatien an amount of energy not much less than could be obtained from the original protein. The combination of ammonia with carbon dioxide and the conversion of the carbonate into urea, perhaps through the intermediate stage of ammonium carbamate, does not require any oxidation. Thus, /OH /O.NH, /O.NH4 /NHfi , \OH \O.NH4 \NH2 \NH2 Carbonic acid. Ammonium Ammonium Urea. carbonate. carbamate. Another way in which some of the urea may be produced is by the direct formation of ammonium carbamate in the katabolism of amino- acids without the preliminary liberation of ammonia. By the loss of a molecule of water the carbamate would then become urea. But if, as there is every reason to believe, a part of the carbonaceous residue is converted into carbo-hydrate, a certain amount of oxidation must occur in the transformation. Such compounds as guanin, sarkin or hypoxanthin, xanthin, uric acid, and creatin, used to be cited as among the possible intermediate substances between protein and urea. But while there is now complete evidence that the first three bodies can be and are converted into uric acid, there is nothing at all to indicate that they are stages on the way to urea. Uric acid is, indeed, very closely related to urea, and can be made to yield it by oxidation outside the body. Not only so, but it is, in part at least, excreted as urea when given to a mammal by the mouth and it replaces urea as the great end-product of nitro- genous metabolism almost wholly in the urine of birds and reptiles. But none of these things can be admitted as evidence that in the normal metabolism of mammals uric acid lies on the direct line from protein to urea. Creatin exists in the body in greater amount than any of these, muscle containing from 0-2 to 0-4 per cent, of it; and the total quantity of nitrogen present at any given time as creatin is not only greater than that of the nitrogen present in urea, but greater than the whole excretion of nitrogen in twenty-four hours. But although there are facts which indicate that creatin is an important derivative of the decomposed tissue proteins (p. 597) there is no evidence that it is related to urea formation. Summary : Here may be summed up the most probable view of the normal decomposition in the body of the amino-acids, whether they undergo decomposition without being incorporated into the tissue proteins or proto- 590 METABOLISM, NUTRITION AND DIETETICS plasm, or are liberated as the protoplasm breaks down. The first step may be conceived of as the splitting off of the NH2 group, which yields ammonia. The ammonia is transformed into urea. The non-nitrogenous residue after removal of the amino-group differs according to the amino-acid. Some of the amino-acids, such as alanin, glycin, prolin and aspartic acid, yield substances which can be changed into dextrose in the body, as shown by the increased amount of dextrose excreted by phlorhizinized animals when the amino-acids are fed to them.* Other amino-acids — e.g., lysin, histidin and tyrosin, phenylalanin, etc., have such a chemical structure that the non-nitrogenous compounds derived from them cannot form dextrose. From certain amino-acids, such as histidin, phenylalanin, tyrosin and leucin, it has been shown that acetone bodies can be derived. These facts explain how sugar can be formed and the acidosis associated with the acetone bodies developed in diabetes, even when the diet consists of protein only or when the patient is living on his own tissues. Formation of Uric Acid. — Uric acid, like urea, is separated from the blood by the kidneys, not to any appreciable extent formed in them. In birds, and often in man, it can be detected in normal blood. It is present in increased amount in the blood and transuda- tions of gouty patients, in whose joints and ear-cartilages it often forms concretions. ' Chalk-stones ' may contain more than half their weight of sodium urate. As to the place and manner of formation of uric acid, it has already been stated that in birds, after extirpation of the liver, the uric acid excretion is greatly diminished, and that ammonium lactate appears instead in the urine. The simplest interpretation of this result is, that ammonia and lactic acid pass into the urine because they can no longer be utilized for the synthesis of uric acid. Chemical schemata can indeed be constructed, which show more or less plausibly how lactic acid, pyruvic acid (p. 545) and other substances reacting with ammonia or with the urea derived from it (and birds form some urea) might yield uric acid. It has been further stated that when blood containing ammonium lactate is circulated through the surviving liver of the goose, an increase in the uric acid content of the blood occurs. As demonstrated by control experiments, this increase is too great to be due merely to the sweeping out of pre- viously formed uric acid from the hepatic cells ; also the feeding of lactic acid, pyruvic acid, and other organic acids leads to an increased output of uric acid. The story seems fairly complete, although criticisms have not been lacking. It has been suggested, for instance, that for some reason the loss of the liver leads to acidosis, an in- creased production of acids, especially lactic acid, in the organism ; that ammonia, which would otherwise be employed in the formation of uric acid, is needed to neutralize these acids, and that the appear- ance of this ammonia in the urine is only a secondary consequence of the elimination of the liver. The deficiency in the uric acid * In animals under the influence of phlorhizin sugar is formed from all substances capable of producing it in the organism. METABOLISM OF PROTEINS 59* excreted, it is said, is therefore due, not to inability on the part of the remaining tissues to form uric acid, but to the absence of the ammonia which they require for its formation. This criticism, if it were admitted as against the current interpretation of such ob- servations on the bird's liver, could scarcely be denied some validity as against the current interpretation of similar observations on the results of interference with the mammalian liver. It is therefore important to point out that there is still the same deficiency of uric acid when alkali is administered to neutralize the acids, although ammonia ought now to be available. There can be no question, then, that the liver in birds is the seat of an extensive synthesis of uric acid, and there is little doubt that ammonia compounds are essentially concerned in the process, whatever the role of the lactic or other acids may be. A similar synthetic formation of uric acid from ammonia and a derivative of lactic acid may take place in mammals, and probably exclusively in the liver, but it is of much less importance. Another way in which uric acid arises both in mammals and in birds is by the splitting and oxidation of nucleins. This is by far the most important mode of formation in mammals, as synthesis is the chief mode of formation in birds. In both groups of animals the oxidative production of uric acid takes place, not in any particular organ, but in the tissues in general, including the liver. It has been shown that when air is blown through a mixture of splenic pulp and blood, uric acid is formed from purin bodies already present in the spleen. When the quantity of these is increased by the decomposition of nucleins induced by slight putrefaction, the yield of uric acid is also increased. Uric acid is also formed by the perfectly fresh surviving spleen, liver, and thymus in the presence of oxygen, and the quantity is increased when purin bodies are artificially added. Sources of the Uric Acid. — It is well established that in the bird it arises both from amino-acids derived from the hydrolysis of protein and from nuclein compounds and their derivatives in the food and tissues. The amino-acids constitute by far the greatest source of uric acid in these animals and in the reptiles, and it is practically certain that the course of the decomposition of the amino-acids and the form in which nitrogen is liberated from them in its transformation into this end-product are not essentially differ- ent from what obtains in the formation of urea in the mammal and the amphibian. This is sufficiently illustrated by the rdle played by ammonia and ammonia compounds in the production of uric acid in the birds and their congeners. In the mammal, the taking of food rich in nucleated cells, and therefore in nucleo-proteins and nucleins, the characteristic conjugated proteins of nuclei (thymus gland, pig's pancreas, and herring roe), or of food rich in purin bases (Liebig's meat extract), increases the quantity of uric acid in 592 METABOLISM, NUTRITION AXD DIETETICS the urine. The increase is mainly due to the production of uric acid from the nuclein substances of the food. But this is not the only source of the uric acid, since extracts of the thymus gland containing only traces of nucleins or nucleic acid cause, when in- jected, a characteristic increase in the uric acid excretion, just as the entire gland does when taken by the mouth. And during the period of increased nitrogen excretion occasioned by a meal contain- ing protein, the increase in the uric acid occurs particularly in the hours immediately following the ingestion of the food, and does not last so long as the increase in the urea. Now, the nucleins of the food are comparatively little affected during the earlier stages of digestion (Hopkins and Hope). Whether in mammals any portion of the uric acid comes from amino-acids is still in doubt, but there are facts which indicate that a fraction of it may do so. We may conclude, therefore, that in the mammal, as well as in the bird, a portion of the uric acid, although certainly a far smaller portion in the mammal, is derived from bodies other than the nuclein substances of the food — that is io say, from the nuclein substances of the tissues contained particularly in the cell-nuclei and probably from the ordinary proteins of both food and tissues. The portion derived irom the proteins may be assumed to be that small fraction which has already been spoken of as synthetically formed. Metabolism of the Nucleic Acids and Purin Bases. — Our know- ledge of the metabolism of the nucleo-proteins and nucleins has been greatly augmented in recent years. When nucleo-protein is digested by gastric juice, a certain amount of protein is easily split off and hydrolysed to peptone and the other ordinary products of proteolysis. An insoluble residue of nuclein remains. This is acted upon with difficulty by gastric juice, although eventually an active juice will split it up also. By the action of pancreatic juice, or by heating with dilute acids, it is more easily hydrolysed, yielding a further quantity of protein along with nucleic acid. This second fraction of protein, which is split off with so much more difficulty than the first, undergoes proteolysis in the usual way. The result- ing amino-acids no doubt take their place in the general metabolism precisely like the amino-acids derived from ordinary proteins, and yield the same end-products. As regards the nucleic acid (or rather acids, since different nucleo-proteins contain different nucleic acids), pancreatic juice is practically inert, although succus entericus can effect a partial hydrolysis. For their complete decomposition more drastic treatment is required — namely, heating with hydrochloric acid in a sealed tube. Thus treated, nucleic acids yield a number of components, out of which they may be assumed to be built up, as the proteins are built up out of amino-acids, etc. The charac- teristic components are purin bases (adenin, C5H3N4.NH2; guanin, C6H,N4O.NH2; hypoxanthin, C6H4N4O; and xanthin, C5H4N4O2); METABOLISM OF PROTEINS 593 pyrimidin bases (uracil, C4H4N2O2; cytosin, C4H3N2O.NHa; thymin C4H3NaOa.CH3); phosphoric acid and a carbo-hydrate group. Some of the nucleic acids contain all these components; they are sometimes spoken of as the true nucleic acids. In others certain of the components are absent, and to these nucleic acids the name nucleotids has been applied. The purin bases are always present. The carbo- hydrate group varies in different nucleic acids, being in some a hexose (p. 537), in others a pentose (p. 487). The pentose d-ribose is especially often met with. It is probable that the nucleotids are merely simpler decomposition products of the true nucleic acids. Thus, inosinic acid, a nucleotid first isolated from meat extract, yields phosphoric acid, <2-ribose, and the purin base hypo xan thin. The nucleotid guanylic acid found in the pancreas yields phosphoric acid, d-ribose, and the purin base guanin. There is evidence that nucleic acids may be built up out of a number of nucleotid groups, and for this reason they have been termed polynucleotids (Levene). The purin bases have a very close chemical relationship to uric acid, which, like them, is characterized by the possession of a group called the purin nucleus. For convenience of reference the atoms composing the purin nucleus are numbered, and the purin bodies are named with reference to the position of the carbon atom or atoms at which oxygen or the amino-group (NH2) is introduced. Purin consists of the nucleus with H atoms introduced at the points shown in the constitutional formula. Adenin is a 6-a.mino- purin — i.e., purin in which NH2 replaces the H attached to C(«j. Guanin is 2-amino-6-oxypurin, NH2 being united with C@) and oxygen with C«j) in purin. Uric acid is 2, 6, 8-trioxypurin — i.e., purin in which oxygen is united to the carbon atoms 2, 6, and 8. Hypoxanthin is 6-0 xy purin, oxygen being introduced at the position of C(6) in purin. By removal of the amino-group from, adenin hypo xan thin is formed. Xan thin is 2, 6, dioxypurin, oxygen being introduced at C(2> and C(6) in the purin molecule. Xan thin can be derived from guanin in the same way as hypoxanthin from adenin. C(6) - N(T) N(3) C,4)— N(9) N=CH HC C— NH ;CH N— C— N Purin nucleus. Purin. N =C.NH2 I I HC C— NH >H N— C— N Adenin. NH2.C NH— CO L NH NH— CO Jo c!-: N=C.OH NH— CO •NH HC C— NH o NH CH CO N — C— N Guanin. NH— C— NH Uric acid. > CH N— C— N Hypoxanthin. NH— C— N X anthin. )CH Besides the purin bases combined in the nuclein substances, purin bases and uric acid are widely spread in the tissues in the free state, although in very small amounts. A portion of the intake of purin bodies is therefore ready formed, especially in the animal constituents of the food, and does not require the decomposition of nucleic acid for its liberation. The nuclei of vegetable cells contain nucleo-proteins, and accordingly can contribute to the purin intake. The most interesting contribution of vegetable 38 594 METABOLISM, NUTRITION AND DIETETICS origin has been previously alluded to (p. 481) — namely, the methyl purins forming the active "principles of tea, coffee, and cocoa, caffein, or i, 3, 7-trimethylxanthin (C8H10N4O2), theobromin, or 3, y-dimethyl- xanthin (C7H8N4O2), and theophyllin, or i, 3 - dimethylxanthin (C6H8N402). CH3.N— CO CO C— N.CH3 >H CH,.N— C— N Caffein. NH— CO :o CH3.N — C Theobromin. CH3.N— CO . 482), and its complete independence of the changes in the total nitrogen excretion, show that it has a. different significance in protein metabolism from the urea. Evidence is accumu- lating that it is especially in the metabolism of the organized or tissue protein that the product eventually excreted as creatinin arises ; in other words, that it represents especially the nitrogenous waste connected with the wear and tear of the bodily machinery, while urea represents also, and under ordinary conditions of diet chiefly, the nitrogen of the surplus amino-acids which are not utilized in the building of new or the repair of old tissue elements. The fact that the amount of creatinin excreted by different persons seems to be related to the weight of active tissue in the body, excluding fat, is in favour of this suggestion, and METABOLISM OF PROTEINS 599 there is other evidence pointing in the same direction ; for example in ordinary circumstances creatin is either absent from the urine or present in very small amount, except in young children. When, how- ever, the decomposition of tissue-protein is abnormally increased, as in starvation, in fevers, in women after delivery, while involution of the uterus and the associated destruction of a considerable mass of smooth muscle is taking place, creatin appears in larger quantities in the urine, perhaps because it can no longer be all converted into creatinin. Now, the increased excretion of creatin in starvation can be prevented by giving carbo-hydrate food, which is known (p. 606) to lead to sparing of tissue-protein (Mendel and Rose). That the depletion of the body of carbohydrate is in some way related to the elimination of creatin is further shown by the fact that creatinuria is associated with diabetes and with the action of substances like phlorhizin, hydrazin, phosphorus, etc., all of which cause carbo-hydrate deficiency. Something more is involved, however, for under conditions of diet which produce acidosis (an increase in the hydrogen-ion concentration of the blood), creatin appears in the urine no matter how rich the food may be in carbo-hy- drate. For example, a diet of oats and maize, which are typical acid- producing foods, causes creatinuria in rabbits, which promptly disappears when carrots, a base-producing food, are added (Underhill). The statement that the content of the urine in creatinin is increased by muscular work may indicate that the muscular machine wears out* faster during activity than during rest, or perhaps only that already- formed creatin leaves the muscles in greater amount when the blood-flow in increased; but recent observations tend to show that this statement may require revision. As to the manner in which creatin is changed into creatinin in the body, a highly suggestive fact is the presence of ferments in various organs which possess this power. Ferments also exist which can decompose both creatin and creatinin. The existence of such enzymes is presumptive evidence that the changes which they are capable of producing actually occur in the organism ; but the seat of the changes if they do take place, and their metabolic significance, are unknown. Creatin when given by the mouth or injected into the blood does not cause any increase in the urinary creatinin, nor when administered in moderate quantities does it seem to be excreted as creatin. Like urea, creatinin, and amino-acids, it is taken up very rapidly by the muscles. Creatinin, on the other hand, when added to the food, causes an increase in the creatinin of the urine. Intracellular Ferments — Autolysis. — As to the agencies by which the decomposition of the proteins is carried out in the cells, we have already spoken of the oxidizing cell ferments, or oxydases (p. 272). Reducing ferments, or reductases, are also known, and can be ex- tracted from most organs, if not all. Like oxydases, they act in a weakly alkaline medium, causing in the presence of hydrogen such reductions as the formation of nitrites from nitrates. There is some evidence that one and the same ferment may act as an oxydase or a reductase according to the conditions. Recent researches have brought to light in addition hydrolytic intracellular ferments, which split up proteins very much in the same way as the proteolytic ferments of the digestive juices. The significance of these autolytic enzymes in the normal metabo- 6oo METABOLISM, NUTRITION AND DIETETICS lism of proteins has been already discussed (p. 578) ; indeed, so many of the chemical reactions of the body have been found to depend upon enzymes that modern physiology may at first thought seem almost to have reverted to the position of Van Helmont and his school in the seventeenth century, who resolved all difficulties by murmuring the magic word ' ferment.' No fewer than eleven fer- ments have been stated to be present and active in the liver alone — viz., a proteolytic and a nuclein-splitting ferment, a ferment which splits off ammonia from amino-acids, a milk-curdling ferment, a fibrin ferment, a bactericidal ferment, an oxydase, a lipase, a maltase, a ferment called glycogenase, which changes glycogen into dextrose, and an autolytic ferment. In the presence of such an array of enzymes the organs might seem to be little more than incubators in which the ferments do their work. It must not be supposed, however, that the intracellular ferments, whether they cause decomposition or synthesis, oxidation or reduction, work in- dependently of what, for want of a better name, we must call the organization of the cell. We may be sure they are the servants and not the masters of the protoplasm, and that a drop of an extract containing intracellular ferments has very different powers from a living cell. ' It is not in the existence of the ferments, but in their combined action at the proper time and in the proper intensity, that the riddle of metabolism lies ' (Hober) . Summary. — At this point let us sum up what we have learnt as to the relation between the approximate principles of the tissues and the proximate principles of the food. Inside the body we recognize representatives of the three groups of organic food-substances in a typical diet — proteins, carbo-hydrates, and fats. But we should greatly err if we were to imagine that the three streams of food- materials have flowed from the intestines into the tissues each in its separate channel, neither giving to nor taking from the others. The fats of the body may, indeed, in part be composed of molecules which were present as fat in the food ; but they may also be formed from carbo-hydrates, and probably from proteins. The carbo-hydrates of the body — the glycogen of the liver and muscles, the sugar of the blood —may undoubtedly be derived from carbo-hydrates in the food, but they may also be derived from proteins and from fats (certainly from their glycerin constituent, perhaps from the fatty acids as well) . The pro- teins of the body come mainly, if not solely, from the proteins of the food. Although, of course, neither fats nor carbo-hydrates can by themselves form protein, being devoid of nitrogen, it is possible that products arising in the intermediary metabolism of either may, by combining with nitrogenous groups, be tranfsormed into amino-bodies, which can then take part in the synthesis of proteins. In any case there is no doubt that both carbo-hydrates and fats can economize proteins and shield them from an overhasty metabolism. STATISTICS OF NUTRITION 60 1 SECTION IV. — STATISTICS OF NUTRITION — THE INCOME AND EXPENDITURE OF THE BODY IN TERMS OF MATTER.* Preliminary Data. — The office of the food is to maintain the con- stituents of the body upon the whole in their normal proportions. A knowledge of the chemical composition of the body is, therefore, an important datum in the consideration of the statistics of its metabolism. The body of a man analyzed by Volkmann had the following composition : Inorganic substances Pn^lmatte; - - 6« per cent. {Carbon 1 8 -4 pe r cent . ^ NSro"!nn 11 I; [-9-7 .. Oxygen 6-0 ,, J The muscles, the adipose tissue, and the skeleton form nearly four- fifths of the total body-weight in the adult. The following table shows the percentage amount of each of these tissues in a man, a woman, and a child (Bischoff) : Man. Woman. New-born Child. Voluntary muscles Adipose tissue Skeleton Rest of body 41-8 18-2 15-9 24-I 35-8 28-2 15-1 20-9 23-5 13-5 I5'7 47'3 The nitrogen is contained chiefly in the muscles, glands, and nervous system, and in the constituents of the connective tissues, which yield gelatin, various mucoids, and elastin. The ordinary proteins make up about 9 per cent, of the weight of the body, or 22 per cent, of its solids ; the albuminoids or sclero-proteins (gelatin-yielding material, etc.) (p. 2) about 6 per cent, of the body-weight. Nitrogen exists in proteins to the extent of 16 per cent., so that the 6-5 kilos of protein of a yo-kilo body contain about I kilo of nitrogen. The carbon is contained chiefly in the fat, which forms a very large proportion of the water-free substance of the body, and in the proteins. A small amount is present as calcium carbonate in the bones. In the body of a strong young man weighing 68-6 kilos, Voit found the following quantities of dry fat in the various tissues : Adipose tissue Skeleton Muscles Brain and spinal cord Other organs Total - 8809-4 grammes. 2617-2 ,, 636-8 226-9 .» 73-2 - 12363-5 equivalent to 18 per cent, of the whole body-weight, or 44 per cent, of the solids. In dry fat rather more than 75 per cent, of carbon is present, and in protein about 50 to 55 per cent. ; so that while the fat of the body analyzed by Voit contained more than 9 kilos of carbon, only about a third of this amount would be found in the proteins. * The income and expenditure of the body in terms of energy are considered in Chapter XII. 602 METABOLISM, NUTRITION AND DIETETICS In the fat there is, roughly speaking, 12 per cent, of hydrogen, in proteins only 7 per cent. ; so that from three to four times as much hydrogen is contained in the fat of the body as in its proteins. Oxygen forms about 12 per cent, of fat, and 20 to 24 per cent, of proteins; the protein constituents of the body, therefore, contain about as much of its oxygen as the fat. Of the inorganic salts, calcium phosphate, Ca3(PO4)2, is much the most abundant, owing to the large amount of it in bone, in the ash of which it is found to the extent of 83 per cent., along with 13 per cent, of calcium carbonate. INCOME AND EXPENDITURE OF NITROGEN — THE NITROGEN BALANCE-SHEET. Nitrogenous Equilibrium. — It is a matter of common experi- ence that the weight of the body of an adult may remain approxi- mately constant for many months or years, even when the diet varies greatly in nature and amount. And not only may the weight remain constant, but the relative proportions of the various tissues of the body, so far as can be judged, may remain constant too. Here it is evident that the expenditure of the body must precisely balance its income: it must lose as much nitrogen as it takes in, otherwise it would put on flesh; it must lose as much carbon as it takes in, otherwise it would put on fat. Or, again, the body may be losing or gaining fat, giving off more or less carbon than it receives, while its ' flesh ' (its protein constituents) remains constant in amount, the expenditure of nitrogen being exactly equal to the income.* In both cases we say that the body is in nitrogenous equilibrium. A starving animal or a fever patient, on the other hand, is living upon capital, the former entirely, the latter in part ; the expenditure of nitrogen is greater than the income. A growing child is living below its income, is increasing its capital of flesh. In neither case is nitrogenous equilibrium present. The starving animal, as long as life lasts, excretes urea, kreatinin, and other nitrogenous substances, and gives off carbon dioxide; but its expenditure, and especially its expenditure of nitrogen, is pitched upon the lowest scale. It lives penuriously, it spins out its resources; its glycogen goes, its fat goes, a certain part of its protein goes, and when its weight has fallen from 25 to 50 per cent, it dies. At death the heart and central nervous system are found to have scarcely lost in weight ; the other organs have been sacrificed to feed them. Fig. 199 shows the percentage loss of weight and the proportion of the total loss which falls upon each of the organs of a cat in starvation (Voit). * For long experiments extending over many days the nitrogen balance may be considered as practically the same as the protein balance, but this is not necessarily true of short periods of time, since the stock of nitrogen present in the body in other forms than proteins, although relatively small, is subject to variations. STATISTICS OF NUTRITION 603 For the first day of starvation the excretion of urea in a dog or cat is not diminished; it takes about twenty- four hours for all the nitrogen corresponding to the proteins of the last meal to be elimin- ated. On the second day the quantity of urea sinks abruptly; then begins the true starvation period, during which the daily output of urea remains constant or diminishes very slowly until a short time before death, when it rapidly falls, and soon ceases altogether. An increase in the excretion may precede the final abrupt decline (pre- mortal increase). This seems to indicate the time at which all the available fat has been used up, and after which protein is no longer ' spared ' by the fat.* If the animal has little fat in its body to begin with, the rise in the urea excretion takes place even after the first few days. So long as the fat lasts the rate at which it is 42-1 Muscle Z6-Z Fat 8-7 Skin S-5 Bones 4- -8 Liver 3-6 Blood Z-0 Mestines 0-6 Kidneys • ^0-5 Spleen 0-3 Lungs 0-1 Pancreas Oil Testes (O-/ Brain & Cord {(?•/ Heart Fig. 198. — Diagram showing Loss of Weight of the Organs in Starvation. The numbers under I. are the percentages of the total loss of body-weight borne by the various organs and tissues. The numbers under II. give the percentage loss of weight of each organ calculated on its original weight as indicated by com- parison with the organs of a similar animal killed in good condition. destroyed — as estimated from the amount of carbon given off minus the carbon corresponding to the broken-down proteins — remains very nearly constant after the first day. The fat to a certain extent economizes the proteins of the starving body, but however much fat may be present, a steady waste of the tissue-proteins goes on. If non-nitrogenous food in the form of sugar is supplied to an other- wise starving animal, the premortal rise in the nitrogen excretion does not occur. By giving a sufficient quantity of sugar, or df sugar and fat, but practically no protein (so-called nitrogen starva- tion), the excretion of nitrogen may be reduced to one-third of its amount when no food at all is given. This is true both in animals * If the animal has been for some time on a diet containing an abundance of proteins, several days may elapse before the constant excretion of urea is reached; if the previous diet has been poor in protein, the constant star- vation output may be at once established. 604 METABOLISM, NUTRITION AND DIETETICS and man. In this way the daily excretion of nitrogen in a man has been reduced to 4 grammes. It is a remarkable fact that while a mixture of carbo-hydrate and fat will act just as well as carbo- hydrate alone in bringing about this reduction in the nitrogen output, fat without carbo-hydrate is much less effective. The hypothesis suggested by Landergren to explain this is alluded to on another page (p. 551). The results obtained on fasting men differ in some respects from those obtained on starving animals. In ten days of hunger, Cetti, a professional ' fasting man ' of meagre habit, excreted 112 grammes nitrogen, or an average of n grammes a day. The excretion was least on the eighth, ninth, and tenth days — namely, about 9 grammes a day. On the third day it was higher than on the second, and almost as high on the fourth as on the third. A similar rise in the nitrogen excretion on the second day has been observed in other fasting men, but is either rare or absent in fasting dogs. The explanation appar- _fl •_ ^al :„ tv.e Fig. 199.— Excretion of Urea in Starvation. A is a curv« ,ntiy 1 nai in me representing the quantity of urea excreted daily by a ordinary tOOd OI fat dog in a starvation period of sixty days. B is the man there is a curve of urea excretion in a lean young dog in a cn-Aafpr ahiinrlanrp starvation period of twenty-four days. Both are con- structed from Falck's numbers, but in A only every Of car DO - hydrates third day js put m> in order to save space. The num. and fats, the pro- bers along the vertical axis represent grammes of urea; tein - sparing action those along the aorizontal axis davs flom the beginning ,5 , ? of starvation, of which is most pronounced at the very beginning of the starvation period. The quantity of chlorine and alkalies in the urine was also diminished, while the phenol was increased. The respiratory quotient sank to 0-66 to 0-69 — even less than the quotient corresponding to oxida- tion of fats alone. The meaning of this, in all probability, is that some of the carbon of the broken-down proteins was laid up in the body as glycogen (Zuntz). In another professional fasting man (Succi) with a considerable amount of body-fat, the excretion of nitrogen was found to diminish continuously during a fast of thirty days, being less than 7 grammes on the tenth day. In another fast of twenty-one days by the same person it was a little less than 3 grammes on the last day. The surprisingly small nitrogenous waste in this case is perhaps to be accounted for by the protein- sparing action of the abundant body-fat. The nitrogenous metabo- STATISTICS OF NUTRITION 605 lism has also been investigated during long-continued hypnotic sleep (Hoover and Sollmann). The results were very much the same as in an ordinary starvation experiment. It might be supposed that' if an animal was given as much nitrogen in the food in the form of proteins as corresponded to its daily loss of nitrogen during starvation, this loss would be entirely prevented and nitrogenous equilibrium restored. The supposition would be very far from the reality. If a dog of 30 kilos weight, which on the tenth day of starvation excreted 11-4 grammes urea, had then received a daily quantity of protein equivalent to this amount — that is to say, about 34 grammes of dry protein, or 175 grammes of lean meat — the excretion of nitrogen would at once have leaped up to nearly double its starvation value. If the quantity of protein in the diet was progressively increased, the output of urea would increase along with it, but at an ever- slackening rate; and at length a condition would be reached in which the income of nitrogen exactly balanced the expenditure, and the animal neither lost nor gained flesh. In an experiment of Volt's, for instance, the calculated loss of flesh in a dog with no food at all was 190 grammes a day. The animal was now fed on a gradually increasing diet of lean meat, with the following result : Flesh in the Food. Flesh used up in the Body. Net Loss oi Body -flesh. O IQO 190 250 341 91 350 4II 61 4OO 454 54 450 471 21 480 492 12 The loss of nitrogen in the urine and faeces is what was measured. Knowing the average composition of ' body-flesh ' (muscles, glands, etc.), it is possible to translate results stated in terms of nitrogen into results stated in terms of ' flesh.' Muscle contains approximately 3-4 per cent, of nitrogen. Here, with a diet of 480 grammes of meat, the dog was still losing a little flesh; it would probably have required from 500 to 600 grammes for equilibrium. The results are graphically represented in Fig. 200, p. 607. The quantity of protein food necessary for nitrogenous equili- brium varies with the condition of the organism ; an emaciated body requires less than a muscular and well-nourished body. The least quantity which would suffice to maintain in nitrogenous equilibrium the famous 35 kilo dog of Voit, even in very meagre condition, was 480 grammes of lean meat, corresponding to 16 grammes of nitrogen, or 35 grammes of urea — that is, about three times the daily loss 606 METABOLISM, NUTRITION AND DIETETICS during starvation. From this lower limit up to 2,500 grammes oi meat a day nitrogenous equilibrium could always be attained, the animal putting on some flesh at each increase of diet, until at length the whole 2,500 grammes was regularly 'used up in the twenty- four hours. A further increase was only checked by digestive troubles. A man, or at least a civilized man, can consume a much smaller amount both absolutely and in proportion to the body- weight. Rubner, with a body-weight of 72 kilos, was able to digest and absorb over 1,400 grammes of lean meat; Ranke, with about the same body-weight, could only use up 1,300 grammes on the first day of his experiment, and less than 1,000 grammes on the third. But whether the surplus of protein food above the necessary minimum is great or small, nitrogen equilibrium is eventually attained, and thereafter all the nitrogen of the food regularly appears in the excreta; the explanation of this fact will be considered a little later (p. 609). So much for a purely protein diet. When fat is given in addition to protein, nitrogenous equilibrium is attained with a smaller quantity of the latter. A dog which, with protein food alone, is putting on flesh, will put on more of it before nitrogenous equilibrium is reached if a considerable quantity of fat be added to its diet. Fat, therefore, economizes protein to a certain extent, as we have already recog- nized in the case of the starving animal. On the other hand, when protein is given in large quantities to a fat animal, the consumption of fat is increased ; and if the food contains little or none, the body- fat will diminish, while at the same time ' flesh ' may be put on. The Banting cure for corpulence consists in putting the patient upon a diet containing much protein, but little fat or carbo-hydrate ; and the fact just mentioned throws light upon its action. All that we have here said of fat is true of carbo-hydrates. To a great extent these two kinds of food substances are complementary. Carbo-hydrates economize proteins as fat does, but to a greater extent, so that with an abundant supply of carbo-hydrate in the food the minimum protein requirement can be forced down much below what is possible on a diet of protein and fat alone. Carbo- hydrates also economize fat, so that when a sufficient quantity of starch or sugar is given to an otherwise starving animal, all loss of carbon from the body, except that which goes off in the urea, krea- tinin, etc., still excreted, can be prevented. Of course, the animal ultimately dies, because the continuous, though diminished, loss of protein cannot be made good. The fact that carbo-hydrates econo- mize proteins so much more efficiently than fat indicates that sugar is essential in the bodily metabolism, so that when carbo-hydrates are absent from the food some of the protein must be broken down so as to yield eventually the compounds necessary for the formation of carbo-hydrate. It is probable, indeed, that purified proteins, STATISTICS OF NUTRITION 607 vfOO 4.00 200 !OO 0 100 1$0 absolutely free from admixture with carbo-hydrates, which, of course, is not the case with the natural protein foods, will not per- manently suffice for nutrition, but that the protein must be supple- mented by a certain amount of carbo- hydrate in some form available for the tissues. It would appear, indeed, that fats are not absolutely indispensable either for maintenance or for growth. White rats have been seen to grow nor- mally over long periods with dietaries devoid of fat ; for example, mixtures of the purified protein edestin (from hemp seed) or casein with starch, sugar, and ' protein- free milk ' freed from fat by extraction with ether (Osborne and Mendel). While in these experiments the food might not have been free from the so-called ' lipoids/ it has been demonstrated that an impor- tant group of substances of this class, the phosphatides, can be synthesized in the body, the necessary phosphorus being ob- tainable even from inorganic phosphates (McCollom). Relation between Nitrogen excreted and the Quantity of Protein Food. — At this point we may consider a little more closely a phenomenon already alluded to, and to which much discussion used to be devoted by writers on metabolism. It has been stated that within the limits of nitrogenous equilibrium, which is the nor- mal state of the healthy adult, the body lives up to its income of nitrogen ; it lays by nothing for the future. In the actual pinch of starvation the organism, when its behaviour is tested by a comparison of the intake and excretion of nitrogen, appears to have become suddenly econo- mical. When a plentiful supply of protein is presented to the starving body, it seeins, judged by the same criterion, to pass at once from extreme frugality to luxury. Some flesh may be put on for a short time, some nitrogen may be stored up ; but the excretion of nitrogen is soon adjusted to the new scale of supply, and the protein income is apparently spent as freely as it is received. These facts were usually summed up in the Fig. 200. — Curves constructed to illustrate Nitrogenous Equilibrium (from an Ex- periment of Voit's). The loss of flesh in grammes is laid off along the horizontal axis. The income and expenditure corresponding to a given loss are laid off (in grammes of ' flesh ') along the vertical axis. The continuous curve is the curve of income ; the dotted curve, of expenditure. With no income at all the expen- diture is 190 grammes; with an income of 480 grammes the expenditure is 492 and the loss 12 grammes. Nitrogenous equilibrium is represented as being reached with an income of about 530 grammes; here the two curves cut one another. 608 METABOLISM. NUTRITION AND DIETETICS dictum, often dignified as a ' law ' of nitrogenous metabolism that : Consumption of protein is largely determined by supply (Practical Exercises, p. 720). To explain this many hypotheses were invented. The famous theory of Voit assumed that the food-protein after absorption (the so-called ' circulating-protein ') is carried to the tissues and taken up by the cells, where the greater part of it, without being incorporated with the protoplasm, is nevertheless acted upon, rendered unstable, shaken to pieces, as it were, by the whirl of life (by the intracellular enzymes we might now say less dramatically) in the organized framework, the interstices of which it fills. Pniiger, on the other hand, maintained that we have no right to draw a distinction between the consumption of organ- and circulating- protein; that the whole of the latter ultimately rises to the height of organ- or tissue -protein, and passes on to the downward stage of metabolism only through the topmost step of organization. An increase in the supply of nitrogenous material in the blood must, on this view, be accompanied with an increased tendency to the break-up, the dis- sociation, as Pfliiger put it, of the living substance. The actual organ- ized elements, however, the existing cells, were not supposed to be destroyed ; the building remained, for although stones were constantly crumbling in its walls, others were being constantly built in. A much less plausible view was that the tissue elements themselves are short-lived ; that the old cells disappear bodily and are replaced by new cells; and that the whole of the proteins of the food take part in this process of total ruin and reconstruction. Histo logical evidence, as soon as the methods of examining tissues with the microscope became sufficiently refined, told strongly against this idea. Although the cells of certain glands, such as the mammary, perhaps the mucous glands, and especially the sebaceous glands (p. 565), exhibit changes which, hastily interpreted, might seem to indicate that they break down bodily, as an incident of functional activity, no proof could be obtained of the production of new cells on the immense scale which this theory would require. The relatively small and constant amount of the endogenous metabolism indicates that the actual protoplasmic sub- stance, the living framework of the cell, is comparatively stable; that it does not break down rapidly; and that only a small and fairly constant amount of food- or circulating-protein, or of the decomposition products of protein, is required to supply the waste of the organ -protein. We have referred to these theories because there could scarcely be a more instructive instance of the way in which theories become obsolete with the advance of knowledge and of the way in which, with the advance of knowledge, a phenomenon which appears an absolute riddle to one generation may become fairly intelligible to the next, perhaps childishly simple to a third. The student will not derive much benefit from the perusal of this page should he fail to recognize that the hypotheses of the twentieth century are mortal too, and bound for the same bourne as those of the nineteenth. It is apparent in the first place from our study of the metabolism of the proteins that the conclusion, ' consumption of protein is pro- portional to supply,' cannot be drawn from the equality of nitrogen intake and nitrogen output. The amino-acids derived from proteins, except that relatively small fraction employed in repairing the STATISTICS OF NUTRITION 609 waste of the tissues, which in nitrogen equilibrium is exactly com- pensated for by a corresponding release of amino-acids or their equivalent from the cell-proteins, are indeed speedily deaminated and the nitrogen of the amino-groups excreted as urea (with ammonia compounds and creatinin). But, as we have seen, only a small proportion of the chemical energy of the amino-acids and only a small fraction of their carbon are liberated in this process. The carbon-containing residues are katabolized only to the extent re- quired by the momentary needs of the tissues, any balance being stored as part of the reserve of carbo-hydrate or of fat. The body does not possess the means of storing surplus amino-acids as such or even in the form of proteins, except to the small extent corre- sponding to any increase which may occur in the body-protein when the food-protein is increased beyond the minimum required for nitrogen equilibrium. Why the organism has not developed the capacity to store large quantities of protein is, of course, an interesting question, but it need scarcely be discussed here. One obvious reason is that protein is not a suitable source, nor are amino- acids apparently a suitable source of energy for the tissues until they have been deaminated and have probably undergone further decomposition and transformation. Therefore they are decomposed at once and their available residue stored, if it is in any case to be stored, in the more available form of carbo-hydrate (or fat). Where the food-proteins differ greatly from the body-proteins in the proportions of the various amino-acids, there would be no object in storing a great surplus of those which are most" plentiful in the food, if they were at the same time the scarcest in the tissues, or, in the case of gland-cells, the scarcest in the proteins which they manufacture for their secretions. At any moment the magnitude of this non-utilizable surplus will depend upon the quantity of that one of the indispensable amino- acids which is present in the smallest amount. For the proper proportion must be preserved between the different ' stones ' out of which the molecule is built. When a single amino-acid is intro- duced into the body, it is at once changed into urea and excreted, since it cannot be utilized by itself for building up protein. When the cells have once culled from the mixture circulating in the blood the amino-acids, a full supply of which they have most difficulty in obtaining, a residue, large or small, according to the quantity and quality of the protein intake, will be left, and this can only be utilized to supply energy or to add to the fat and carbo- hydrate stores. For these uses removal of the amino-group is an essential preliminary. The question whether the deamination of a large part of the amino-acids coming from the intestine takes place in the liver, so that the surplus nitrogen is shunted out of the main metabolic current at its very source, has been already touched upon 39 6io METABOLISM, NUTRITION AND DIETETICS (p. 588). Some writers conceive that in such a short-cut from pro- tein to urea we have a kind of physiological safety-valve to protect the tissues from the burden of an excessive metabolism. And il by this is meant that it is advantageous to the tissues that a special mechanism should exist to eliminate a surplus of nitrogen which they do not require, and which they cannot store, and to present them with a residue which they can utilize, the conception is certainly correct. But there is no good evidence that in the presence of an over-abundant supply of amino-acids the endogenous protein meta- bolism would be essentially modified. Relation of Nitrogenous Metabolism to Muscular Work. — This is another of those classical physiological problems which it is difficult to present properly apart from its historical setting. The general result of much experimental work and long-continued discussion is that when the work does not transgress what may be called ' normal limits,' the excretion of nitrogen is nearly independent of mus- cular work — that is to say, the quantity of nitrogen excreted by a man on a given diet is practically the same whether he rests or works. Before this was known it was maintained by Liebig that proteins alone could supply the energy of muscular contraction — that, in fact, proteins were solely used up in the nutrition and functional activity of the nitrogenous tissues, while the non-protein food yielded heat by its oxidation. As exact experiments multiplied, it was found that muscular work, the production of which is the function of by far the greatest mass of protein-containing tissue, had little or no effect upon the excretion of urea in the urine. More than this, it was shown that a certain amount of work accomplished (by Pick and Wislicenus in climbing a mountain) on a non-nitrog- enous diet had double the heat equivalent of the whole of the pro- tein consumed in the body, as estimated by the urea excreted during, and for a given time after, the work. On the assumption that all the urea corresponding to the protein broken down was eliminated during the time of this experiment, a part at least of the work must have been derived from the energy of non-nitrogenous material. And other experiments in which account was taken of the increase in the carbon dioxide given off (as conspicuous an accompaniment of muscular work as the constancy of the urea excretion), showed that during muscular exertion carbonaceous substances other than proteins — that is to say, fats and carbo-hydrates — are oxidized in gi eater amount than during rest. So the pendulum of physiological orthodoxy came full-swing to the other side. Liebig and his school had taught that proteins alone were consumed in functional activity; the majority of later physiologists following Voit denied to the proteins any share whatever in the energy which appears as muscular contraction. The proteins, they said, 1 repair the slow waste of the framework of the muscular machine, replace a loose rivet, a worn-out belt, as occasion may require ; the STATISTICS OF NUTRITION 611 carbo-hydrates and fats are the fuel which feeds the furnaces of life, the materal which, dead itself, is oxidized in the interstices of the living substance, and yields the energy for its work.' Now, it is a singular circumstance, and full of instruction for the ingenuous student of science, that the facts which were supposed absolutely to disprove the older theory, and- absolutely to establish its more modern rival, are now seen to do neither the one thing nor the other. The fact — and it is a fact — that the excretion of nitrogen is but little affected by muscular contraction, does not prove that none of the energy of muscular work comes from proteins; the fact that, under certain conditions, some of the muscular energy must apparently come from non-nitrogenous materials, does not prove that these are the normal source of it all. The distinction had again been made too absolute. The pendulum must again swing back a little; and the experiments of Pfliiger and his pupils were soon to set it moving. In the first place, it is not perfectly correct to say that work causes no increase in the excretion of nitrogen; excessive work in man, and work, severe but not excessive, in a flesh- fed dog (Pfliiger), do cause somewhat more nitrogen to be given off. On the first day of work the increase is always much less than on the second and third ; and on the first and second rest days, following work, the elimina- tion of nitrogen is still increased. After excessive exercise in man not only is the urea increased, but also the ammonia, kreatinin, and if the subject is in poor training, the uric acid and purin bases (Paton, Stockman, etc.). Moderate exercise causes no increase on the first day, but a slight increase on the second. The meaning of these facts seems to be that during muscular work the intensity of which does not exceed certain limits, the protein waste of the muscular substance itself is no greater than during rest. When, however, the machine is ' speeded up ' beyond a certain point the wear and tear is sensibly increased and an excess of tissue-protein is katabolized. There is no reason to suppose that the tissue-protein thus broken down will not yield energy for the muscular work by the oxidation of its non-nitrogenous residue, just as well as the surplus amino- bodies derived from food-protein. The muscular machine has the peculiarity that it is constructed of combustible material ; even the dust and the splinters, if we may so express it, which represent the wear and tear of the machine can be burnt in the furnace which keeps it going. In the second place, even if the excretion of nitrogen were entirely unaffected by work, this would not prove that none of the energy of the work comes from proteins. For, as we have seen, it is after the nitrogen has been split off and converted into urea that the energy of a great part of the food-protein is developed by oxidation. Further, since the animal body is a beautifully-balanced mechanism which constantly adapts itself to its conditions, it is conceivable that it may, when called upon to labour, save proteins from lower uses to devote them to muscular contraction. In this 612 METABOLISM. NUTRITION AND DIETETICS case the excretion of nitrogen would not necessarily be altered ; the proteins which, in the absence of work, would have been oxidized within the muscular substance or elsewhere, their energy appearing entirely as heat, may, when the call for protein to take the place of that broken down in muscular contraction arises, be diverted to this purpose. In any case, there is no doubt that a dog fed on lean meat may go on for a long time performing far more work than can be yielded by the energy of fat and carbo-hydrates occurring in traces in the food, or taken from the stock in the animal's body at the beginning of the period at work. A large portion, and perhaps the whole, of the work, must in this case be derived from the energy of the pro- teins (Pfliiger). On the other hand, it is well established that when fats and carbo-hydrates are present in sufficient quantity in the tissues or the food, they constitute the main source of the energy of muscular contraction (p. 772), and there is some evidence that of the two classes of food materials carbo-hydrates in the form of dextrose (or glycogen) is the material of election. The outcome, then, of this famous controversy is essentially a compromise. Everybody now admits that the muscular machine can and does utilize predominantly any one of the great groups of food substances, be it carbo-hydrate, fat, or protein, when the dietetic conditions are such that only one of these is offered to it in large amount, the others being either absent or offered in small amount. To be sure, amino-acids are not the first choice, but if it must do so the muscle can make shift with them, and can indeed make them serve excellently well. When all the food substances are present in abundance, carbo-hydrate is favoured above fat, and fat above protein. Experience has shown that the minimum quantity of nitrogen required in the food of a man whose daily work involves hard physical toil is higher than the minimum required by a person lead- ing an easy, sedentary life. This is evidently in accordance with the view that protein is actually used up in muscular contraction ; but it is not inconsistent with the opposite view. For the body of a man fit for continuous hard labour has a greater mass of muscle to feed than the body of a man who is only fit to handle a composing- stick, or drive a quill, or ply a needle ; and the greater the muscular mass, the greater the muscular waste. But if an animal just in nitrogenous equilibrium on a diet of lean meat when doing no work is made to labour day after day, it will lose flesh unless the diet be increased. This must mean that some of the protein is being diverted to muscular work, and that the balance is not sufficient to keep up the original mass of ' flesh ' (see p. 628). Relative Value of Different Proteins in Nutrition — Synthesis of Amino-Acids. — The fact that the various proteins differ quantita- STATISTICS OF NUTRITION 613 tively and qualitatively in respect to their amino-acids raises the ques- tion of the relative value of different proteins in nutrition. In this is involved the further question, whether the body can itself synthesize from other materials any of the amino-acids which may be deficient, or change one amino-acid into another. That the ' peptones ' derived from a protein which is itself capable of permanently supplying the whole nitrogenous intake of the organism can be substituted for the protein scarcely needs demonstration, since it is known that the protein is converted into peptones in digestion. Nevertheless, this has been proved conclusively by feeding experiments with peptones. It was to be expected, too, leaving out of account all consideration of the means of overcoming the repugnance of animals to accepting such unnatural food substances, that the further products of protein hydrolysis, the amino-acids, etc., could be substituted for the original proteins when these were themselves adequate. For it is in the form of amino-acids or at most of such relatively simple polypeptide groups as may still hang together after complete digestion and absorption, that the nitrogenous food substances are normally offered to the tissues. Experimental demonstration of the feasibility of this substitution has also been obtained. The split products of meat, for example, will keep an animal in nitrogen equilibrium as well as the meat from which they are derived. But what happens when one or more of the amino- acids found in the proteins of the body are missing from the protein of the food ? That the components of an amino-acid like arginin (ornithin and urea), into which it can be split not only by the possibly crude and violent methods of the chemical laboratory, but also by the delicate and precisely-adapted ' biological ' action of a special enzyme (arginase), should be able to replace the original amino-acid is a fact which does not greatly help towards an answer. For when these components, or ornithin alone, since urea is always present in the body, are given instead of arginin, the reversal of the enzyme reaction by which arginin is decomposed is all that is neces- sary for its synthesis, and the reversal of such a reaction is doubtless a very commonplace affair in tissue metabolism. The formation of one amino-acid from another, or from materials which do not originate exclusively from protein, is a different thing, and the answer to the question raised, so far as it can yet be given, is that the way in which the body deals with a deficiency in the protein ' building stones ' depends upon the nature of the missing amino- acids. Thus, the phospho-protein casein does not yield glycin on hydrolysis; yet it has been shown that casein is a perfectly adequate or complete protein food capable of covering the whole nitrogen requirement of the body over long periods. The same is true of the cleavage products of casein which has been subjected to pancreatic digestion. In an animal fed on no other protein than casein, with 614 METABOLISM, NUTRITION AND DIETETICS suitable quantities of carbo-hydrate and fat in addition, the glycin contained in certain of the body proteins must therefore have been produced in the body itself. We have already- seen (p. 580) that for the synthesis of hippuric acid after the administration of benzoic acid, glycin is necessary, and the quantity of hippuric acid which can be thus produced is so great that it is impossible to suppose that it all comes from glycin preformed in the body or from glycin in the food substances. It may accordingly be taken as proved that the tissues have the power of synthesizing at least this one of the amino-acids (amino-acetic acid), reckoned among the protein ' building stones. ' It is said that if the casein has been hydrolysed by acid, the products will not preserve nitrogen equilibrium, per- haps because the acid has broken up all the polypeptides (p. 2), some of which the cells may need as the starting-point of protein synthesis. This, however, is uncertain. Lysin also appears to be capable of being synthesized in the body, and protein foods free from lysin, or containing only a trace of it, may yet be adequate for nutrition and growth. Prolin, too, is not indispensable, and this is of special interest, for the amino-acids hitherto mentioned as capable of being built up in the tissues are all more or less directly related to each other, being derivatives of the series of saturated fatty acids. The task of changing one of these into another in which the food is deficient may, therefore, be considered a comparatively easy one. But prolin has no obvious relation to most of the other amino-acids. It is a-pyrrolidin carboxylic acid, CHg — CHj CH2 — CHg CH8 CH.COOH— i.e., pyrrolidin, CH2 CH2 NH NH in which H in one of the CH2 groups is replaced by carboxyi (COOH) . It has been suggested that prolin may be formed in the body from glutaminic (amino-glutaric) acid, which by loss of a molecule of water can be made to yield a-pyrrolidon carboxylic acid. Thus, CH8.CH2.CH.COOH CH2— CHt COOH NH2 -H2O = CO CH.COOH NH Glutaminic acid. a-pyrrolidon carboxylic acid. By reduction the latter compound might be changed into prolin. With proteins deficient in certain other amino-acids a totally different result has been obtained. Gelatin, for example, contains most of the amino-acids and other groups which compose the body proteins, but tyrosin, cystin, 'and tryptopha*ne are lacking in the STATISTICS OF NUTRITION 615 gelatin molecule. Zein, an alcohol-soluble protein or prolamin* derived from maize, yields no tryptophane, glycin, or lysin. Now, it is found that neither gelatin nor zein can replace the whole of the ordinary proteins in the food. When only enough protein is taken to prevent loss of nitrogen from the body, one-fifth of the necessary nitrogen can be supplied in the form of gelatin. When the food is much richer than this in ordinary protein, a correspond- ingly greater proportion of the protein can be replaced by gelatin. The surplus is not used in the endogenous metabolism of the cells (p. 573), but supplies energy to the body after the elimination of its nitrogen as urea, just as the surplus protein would do. Thus gelatin economizes protein in the same way that fat and carbo- hydrates do, but also to some extent in a different way by supplying ' building stones ' for the protoplasm. It is therefore an interesting question whether gelatin can fully replace protein when the missing substances are given in addition. Kauffmann has stated that his own nitrogen requirement (15 -2 grammes) was almost completely covered by a mixture containing 93 per cent, of the nitrogen in the form of gelatin, 4 per cent, as tyrosin, 2 per cent, as cystin, and i per cent, as tryptophane, in addition to the same amounts of carbo-hydrate and fatty food as in the comparison diet, in which the nitrogen was supplied in the form of plasmon, a commercial preparation of casein. Similar results have been reported in experiments on animals in which attempts have been made to ' complete ' such inadequate proteins by addition of the missing amino-bodies, with fair but, according to Osborne and Mendel, not entirely satisfactory results. The converse experiment, in which an amino-acid such as trypto- phane has been purposely eliminated from the food mixture, has also been tried, with the result that rapid deterioration in the condition of the animal ensued. It would seem, indeed, that whatever capacity the animal body may have for synthesizing certain of the amino-acids, this power does not extend to the cyclic compounds tryptophane, tyrosin, phenylalanin, and histidin, which must be supplied in the food. It has been suggested by Osborne that in this regard an essential difference exists between the animal and the plant, the latter alone being endowed with the function of ' cyclopoiesis,' or formation of substances of the cyclic type. It is not clearly understood as yet on what this difference really hinges, whether, as some have supposed, on the inability of the animal organism to form the appropriate fatty acid radicals, or on some * The prolamins are so called because on hydrolysis they yield exceptionally large amounts of prolin (p. 360) and ammonia. They are insoluble in water and absolute alcohol, but soluble in 70 to 80 per cent, alcohol and in dilute acids and alkalies. Besides zein they include gliadin (from wheat and rye), hordein (from barley), and bynin (from malt). They are extraordinarily rich in glutaminic acid, hordein yielding more than any protein hitherto investi- gated (over 41 per cent). 6ib METABOLISM, NUTRITION AND BIETETICS other limitation of its chemical powers. While the cyclic (and hetero- cyclic) compounds cannot be replaced by other ' Bausteine ' of the proteins, they may to some extent replace each other. Thus it would seem that tyrosin can be replaced by phenylalanin (Abderhalden) . While some of the food proteins like casein are sufficient by themselves to supply all the amino-bodies necessary not only for the maintenance, but also for the growth of the body, and can accordingly be termed adequate or complete protein food sub- stances, others, like gelatin, are insufficient by themselves to supply the protein required for mere maintenance, still less for growth, Fig. 201. — Two Female Rats of the Same Age (140 days). The upper one was fed on an ordinary diet, and is of the normal weight for its age. The lower one was fed on a diet composed of a mixture of isolated food-stuffs. Its weight is only equal to that of a normal rat thirty-six days old (Osborne and Mendel). and may be spoken of as inadequate or incomplete proteins. There is a third intermediate group, comprising proteins which suffice when given as the sole protein food to maintain the body for an indefinitely long period, and to repair the tissue waste without per- mitting growth of the animal to take place. Gliadin and hordein (see footnote, p. 615) are representatives of this group. The ex- periments of Osborne and Mendel with gliadin are of special interest, since this substance is very differently constituted from the ordinary food proteins, as well as from the tissue proteins of the animal body. While, as already stated, it yields very large quantities of glutaminic acid, prolin, and ammonia, it either contains no lysin STATISTICS OF NUTRITION 617 and no glycin, or yields too little to be detected with certainty. It also yields comparatively little histidin and arginin. Now, it has been found that a dietary containing carbo-hydrate in the form of starch, fats in the form of lard, and inorganic salts, but no pro- tein except gliadin* suffices to maintain adult rats in good condition for very long periods (up to 290 days), and also to maintain young rats in a stationary condition as regards growth, but in perfect health. The youthful appearance of the rats whose growth was thus inhibited was very striking, and corresponded with their size rather than with their age. The capacity for growth on a normal diet was apparently not in the least diminished; the growth pro- cesses simply remained in abeyance. ' In one rat, after a continuous suppression of growth lasting 277 days, when the animal was 314 days old — an age at which normally little or no growth takes place — satisfactory growth was resumed on a suitable diet.' A still more remarkable experiment was the following: ' A male rat, kept for 154 days with gliadin as the sole protein in the food, was paired with a female also on the gliadin diet. At the end of 178 days on the gliadin diet she gave birth to four young, which were satisfactorily nourished by the mother, still on gliadin, during the first month of their existence. After a month three of the young rats were removed from the mother and put on diets of casein food (i.e., casein plus suitable proportions of carbo-hydrate, fat, and inorganic materials), edestin food and milk food respectively. The fourth was left with the mother. The fourth rat began to evince a failure to grow at about the period (thirty days) when young rats are wont to depend upon extraneous food. The meaning of this last observation can only be that the young animal, when obliged to depend upon its share of the gliadin food, left with the mother in place of the milk formed by the mother from this same gliadin food mixture, showed the typical failure to grow on a diet inadequate as regards the power of producing growth in respect to the protein contained in it. On the other hand, in the body of the mother this inadequate diet had been so transformed that not only had she maintained her body-weight and repaired * Certain accessory substances of unknown nature, contained in some of the natural foods must also be supplied (p. 632). Osborne and Mendel, for instance, gave their animals a certain amount of ' protein-free milk,' contain- ing the salts of milk, but only traces of milk proteins. The important thing in the ' protein-free milk ' is neither the slight residue of protein nor the salt, but an unknown substance, a so-called 'water-soluble vitamine.' When this was supplied in the ' protein-free milk ' the animals were maintained on the artificial diet for very long periods, although they did not grow. It has since been rendered probable that in addition to the water-soluble vitamine, fat- soluble vitamine (McCollom) is necessary for full growth of rats and the main- tenance in health of the adults, upon otherwise complete artificial diets, for a great part of their natural life-span. Such fat-soluble vitamines may be supplied by replacing a part of the lard by butter-fat, egg-yolk fat, beef fat, or codliver oil. 618 METABOLISM, NUTRITION AND DIETETICS her tissue waste completely, but she had produced from it every- thing necessary for the development of the embryo rats up to full term, and after that everything necessary (in the form of milk) for their normal growth during the period of suckling. On the whole, a very large amount of body tissue in proportion to the original weight of the mother must have been formed or renewed in the 200 days or more during which the experiment continued, and during which gliadin was being steadily transmuted into tissue protein, and latterly into the proteins of milk as well, by what might almost be called a feat of chemical legerdemain. There must have occurred a synthesis ' not only of the Bausteine (the " building- stones ") deficient in the protein intake, but likewise of tissue and milk components like the nucleic acids (with their content of purins, pyrimidins, and organically combined phosphorus) and phospho- proteins like casein, etc., which were completely missing ' in the food. It has been suggested that the bacteria of the alimentary canal, which, of course, are plant cells, may have and may exercise on a large scale the power of building up new amino-acids from a variety of materials in the intestinal contents, and that they may thus be synthe- sizing agents, thanks to which inadequate proteins may be reshaped to proteins adequate to the needs of the body. It is precisely, however, in the case of incomplete proteins like gelatin deficient in cyclical compounds, that the body fails to effect the necessary transformation in spite of the fact that plant cells are supposed to be specially capable of forming these compounds. In any case bacterial action would not explain why proteins like gliadin and hordein are only adequate for the renewal of tissue, and not for its growth. This points rather to the possibility that the processes by which the nitrogenous compounds degraded in cellular metabolism are replaced are not of the same char- acter as the processes by which new nitrogenous complexes are built up into growing protoplasm. If, for instance, the protein molecule is not completely disrupted in ordinary metabolism, it will not need to be completely reconstructed, while in the formation of new tissue complete protein molecules will have to be synthesized. Incomplete proteins like gliadin may furnish building-stones adequate for repairing the house, but inadequate for building it from the foundations. Income and Expenditure of Carbon — The Carbon Balance-Sheet. — This division of the subject has been .necessarily referred to in treating of the nitrogen balance-sheet, and may now be formally completed. Carbon Equilibrium. — A body in nitrogenous equilibrium may or may not be in carbon equilibrium. It has been repeatedly pointed out that the continued loss or gain of carbon by an organism in nitrogenous equilibrium means the loss or gain of fat ; and, since the quantity of fat in the body may vary within wide limits without harm, carbon equilibrium is less important than nitrogen equili- brium. It is also less easily attained when the carbon of the food is increased, for the consumption of fat is not necessarily increased STATISTICS OF NUTRITION 619 with the supply of fat or fat-producing food, and there is by no means the same prompt adjustment of expenditure to income in the case of carbon as in the case of nitrogen. Carbon equilibrium can be obtained in a flesh-eating animal, like a dog, with an exclusively protein diet; but a far higher minimum is required than for nitrogenous equilibrium alone. Voit's dog required at least 1,500 grammes of meat in the twenty- four hours to prevent his body from losing carbon. For a man weighing 70 kilos, the daily excretion of carbon on an ordinary diet is 250 to 300 grammes. About 2,000 grammes of lean meat would be re- quired to yield this quantity of carbon; and, even if such a mass could be digested and absorbed, more than three times the necessary nitrogen would have to undergo preliminary cleavage and excretion as urea or be thrown upon the tissues. Not only may carbon equilibrium be maintained for a short time in a dog on a diet containing fat only, or fat and carbo-hydrates, but the expenditure of carbon may be less than the income, and fat may be stored up. But, of course, if this diet is continued, the animal ultimately dies of nitrogen starvation. So far we have spoken only of the income and expenditure of carbon and nitrogen; and from these data alone it is possible to deduce many important facts in metabolism, since, knowing the elementary composition of proteins, fats, and carbo-hydrates, we can, on certain assumptions, translate into terms of proteins pr fat the gain or loss of an organism in nitrogen and carbon, or in carbon alone. But the hydrogen and oxygen contained in the solids and water of the food, and the oxygen taken in by the lungs, are just as important as the carbon and nitrogen; it is just as necessary to take account of them in drawing up a complete and accurate balance- sheet of nutrition. Fortunately, however, it is permissible to devote much less time to them here, for when we have determined the quantitative relations of the absorption and excretion of the carbon and nitrogen, we have also to a large extent determined those of the oxygen and hydrogen. Income and Expenditure of Oxygen and Hydrogen. — The oxygen absorbed as gas and in the solids of the food is given off chiefly as carbon dioxide by the lungs; to a small extent as water by the lungs, kidneys, and skin; and as urea and other substances in the urine and faeces. The hydrogen of the solids of the food is excreted in part as urea, but in far larger amount as water. The hydrogen and oxygen of the ingested water pass off as water, without, so far as we know, undergoing any chemical change, or existing in any other form within the body. But it is important to recognize that although none of the water taken in as such is broken up, some water is manufactured in the tissues by the oxidation of hydrogen. 6ao METABOLISM, NUTRITION AND DIETETICS We have already considered (p. 241) the gaseous exchange in the lungs, and we have seen that all the oxygen taken in does not reappear as carbon dioxide. It was stated there that the missing oxygen goes to oxidize other elements than carbon, and especially to oxidize hydrogen. We have now to explain more fully the cause of this oxygen deficit The Oxygen Deficit. — The carbo-hydrates contain in themselves enough oxygen to form water with all their hydrogen ; they account for a part of the water-formation in the body, but for none of the oxygen deficit. The fats are very different ; their hydrogen can be nothing like com- pletely oxidized by their oxygen. A gramme of hydrogen is contained in 8-5 grammes of dry fat, and needs 8 grammes of oxygen for its com- plete combustion. Only i gramme of oxygen is yielded by the fat itself; so that if a man uses 100 grammes of fat in twenty -four hours, rather more than 80 grammes of the oxygen taken in must go to oxidize the hydrogen of the fat. The proteins also contribute to the deficit. In 100 grammes of dry proteins there are 15 grammes of nitrogen, 7 grammes of hydrogen, and 21 grammes of oxygen. The carbon does not concern us at present. The 33 grammes of urea, corresponding to 100 grammes of protein, contains 15 grammes of nitrogen, a little more than 2 grammes of hydrogen, and a little less than 9 grammes of oxygen. There remain 5 grammes of hydrogen and 12 grammes of oxygen. But 5 grammes of hydrogen needs for complete combustion 40 grammes of oxygen ; there- fore 28 grammes of the oxygen taken in must go to oxidize the hydrogen of 100 grammes of protein. Taking 140 grammes of protein as the amount in a liberal diet for a man, we get 39 grammes as the required quantity of oxygen. This, added to the 80 grammes needed for the hydrogen of the fat, makes a total of, say, 120 grammes, equivalent to about 85 litres of oxygen. A small amount of oxygen also goes to oxidize the sulphur of proteins. With a diet containing less fat and protein and more carbo-hydrate, the oxygen deficit would of course be less. The Production of Water in the Body.— One gramme of hydrogen corresponds to 9 grammes of water. In 140 grammes of proteins and 100 grammes of fat there are, in round numbers, 22 grammes of hydro- gen; in 350 grammes of starch, 21-5 grammes. With this diet, 43 '5 grammes of hydrogen is oxidized to water within the body in twenty-four hours, corresponding to a water production of 391 grammes, or 15 to 20 per cent, of the whole excretion of water. It has been observed that during starvation the tissues sometimes become richer in water, even when none is drunk. The only explanation is that the elimination of water does not keep pace with the rate at which it is produced from the hydrogen of the broken-down tissue-substances, or set free from the solids with which it is (physically ?) united. Inorganic Salts. — The inorganic salts of the excreta, like the water, are for the most part derived from the salts of the food, which do not in general undergo decomposition in the body. A portion of the chlorides, however, is broken up to yield the hydro- chloric acid of the gastric juice. Within the body some of the salts STATISTICS OF NUTRITION 621 are more or less intimately united to the proteins of the tissues and juices, some simply dissolved in the latter. The chlorides, phos- phates and carbonates are the most important ; the potassium salts belong especially to the organized tissue elements, the sodium salts to the liquids of the body; calcium phosphate and carbonate pre- dominate in the bones. The amount and composition of the ash of each organ only change within narrow limits. In hunger the organism clings to its inorganic materials, as it clings to its tissue- proteins; the former are just as essential to life as the latter. In a starving animal chlorine almost disappears from the urine at a time when there is still much chlorine in the body; only the inorganic salts which have been united to the used-up proteins are excreted, so that a starving animal never dies for want of salts. When sodium chloride is omitted as an addition to the food of man, the decomposition of protein seems to be slightly accelerated, but for a time, at least, there are no serious symptoms (Belli). It is a general rule that purely carnivorous animals do not desire salt, and the same is true of human beings living on a purely animal diet, while vegetable feeders eagerly seek it. On the other hand, when an animal, even a carnivore, is fed with a diet as far as possible artificially freed from salts, but otherwise sufficient, it dies of salt- hunger. The blood first loses inorganic material, then the organs. The total loss is very small in proportion to the quantity still retained in the body; but it is sufficient to cause the death of a pigeon in three weeks, and of a dog in six, with marked symptoms of muscular and nervous weakness. A deficiency of lime salts causes changes particularly in the skeleton, although the nutrition of the rest of the body is also interfered with. These changes are of course most marked in young animals, in which the bones are growing rapidly. In pigeons on a diet containing very little calcium the bones of the skull and sternum become extremely thin and riddled with holes, while the bones concerned in movement scarcely suffer at all (E. Voit). It is not indifferent in what form the calcium is taken, nor can it be replaced to any great extent by other earthy bases, as magnesium or strontium. Weiske fed five young rabbits of the same litter on oats, a food relatively poor in calcium. One of the rabbits received in addition calcium carbonate, another calcium sulphate, a third mag- nesium carbonate, and a fourth strontium carbonate. At the end of a certain time it was found that the skeleton of the rabbit fed with calcium carbonate was the heaviest and strongest of all, and contained the greatest proportion of mineral matter. Then came the rabbit fed with calcium sulphate. The animal which received only oats had the worst- developed skeleton; the condition of the animals fed with magnesium and strontium carbonates was but little better. Milk as a Food.— Milk is a food rich in calcium and also in phos- phorus, a circumstance evidently related to the rapid development 622 METABOLISM, NUTRITION AND DIETETICS of the skeleton in the young child. As in the other natural foods, the calcium and phosphorus are partly in the form of organic com- pounds, united with the proteins, the calcium especially with caseinogen, and partly in the form of inorganic salts. Both of these elements are more easily assimilated by the body in the organic than in the inorganic form. The same is true of iron, which exists in organic combination in the bran of wheat, in the haemoglobin of the blood and of muscular fibres, in the nuclei of most cells, vegetable and animal, and conspicuously in the nuclein compounds of the yolk of the egg. Attempts have been made to increase the amount of iron in hen's eggs by giving them food mixed with preparations of iron — e.g., iron citrate. An increase takes place, but only after a long time. Thus in one experiment 100 grammes of egg- substance contained 4-4 milligrammes of Fe2O3 before the administration of the iron was begun; after feeding with iron for three and a half weeks the amount was 4-5 milligrammes, after more than two months 7-4 milligrammes; and after a year only 7-3 milligrammes. Although, as we have seen, inorganic iron can be absorbed, it is certainly the case that under ordinary conditions all the iron that the body receives or needs is taken in the form of organic com- pounds, since there is no inorganic iron in the natural food sub- stances. Stockman, from careful estimations of the quantity of iron in a number of actual dietaries, finds that it only amounts to about 8 to 10 milligrammes a day. He concludes that the greater part of it must be retained in the body and used over and over again. Milk is poor in iron, but this does not hinder the development of the young child, except when it is weaned too late, when it is apt to become anaemic unless the milk is supplemented with a food rich in iron, such as yolk of egg. The explanation is that the foetus, especially in the last three months of intra-uterine life, accumulates a store of iron in the liver and other organs; so that, in proportion to its body-weight, it is at birth several times richer in iron than the adult. This iron, of course, all comes from the mother, and the loss is not exactly balanced by the excess of iron in her food ; certain of her organs, the spleen, for instance, though not apparently the liver, are impoverished as regards their content of iron. .SECTION V. — DIETETICS. There are two ways in which we can arrive at a knowledge of the amount of the various food substances necessary for an average man: (a) By considering the diet of large numbers of people doing fairly definite work, and sufficiently, but not extravagantly, fed — e.g., soldiers, gangs of navvies, or plantation labourers; (b) by making special experiments on one or more individuals. Voit, bringing together a large number of observations, concluded DIETETICS 623 that an ' average workman/ weighing 70 to 75 kilos, and working ten hours a day, required in the twenty-four hours 118 grammes protein, 56 grammes fat, and 500 grammes carbo-hydrate, corre- sponding to about 18-8 grammes* nitrogen, and at least 328 grammes carbon. Ranke found the following a sufficient diet for himself, with a body- weight of 74 kilos : Proteins - ... IOQ grammes. Fat - 100 Carbo-hydrates - 240 ,, This corresponds to only 16 grammes nitrogen and, say, 230 grammes carbon. A German soldier in the field receives on the average : Proteins - - - - - 151 grammes. Fat ... - 46 Carbo-hydrates - 522 representing about 24 grammes nitrogen and 340 grammes carbon. The average ration for four British regiments in peace-time con- tained 133 grammes protein, 115 grammes fat, and 424 grammes carbo-hydrate ( = 3,400 calories). But in addition the soldiers constantly obtained at their own expense a supper, generally com- prising meat (Pembrey). The Russian army war ration in/ the Manchurian campaign is said to have comprised 187 grammes protein and 775 grammes carbo-hydrate, but only 27 grammes fat ( = 4,900 calories). The diet of certain miners (Steinheil) and lum- berers (Liebig) contained respectively 133 and 112 grammes protein, 113 and 309 grammes fat, and 634 and 691 grammes carbo-hydrates. The diet of a Japanese jinricksha man with a body- weight of 62 kilos contained 158 grammes protein, and its total heat value was 5,050 calories. The work of these men in running long dis- tances with passengers is very laborious. They consume large amounts of fish, eggs, beef, and pork during their periods of rest, and large quantities of rice during their working periods (McCay). The diet of prize-fighters and of athletes in training is richer in protein than any of these. The members of two college football teams are stated to have consumed on the average 225 grammes protein, 334 grammes fat, and 633 grammes carbo-hydrates ( = 6,800 calories). Caspari, from a study of the phenomena -of training, concluded that continuous bodily work at a rate above the ordinary requires a large amount of protein (2 to 3 grammes a day per kilo of body- weight). But there seems to be a considerable difference between different individuals. So that a definite and typical diet for severe labour does not exist. And although perhaps the hardest physical work ever done in the world is to break athletic * Taking the percentage of nitrogen in protein at 16. 624 METABOLISM, NUTRITION AND DIETETICS records, to cut and handle timber, to mine^coal, and to make war, the diet on which these things are accomplished is very variable. Recent observations tend to reduce the amount of protein con- sidered necessary for a person under ordinary conditions. Siven remained in nitrogen equilibrium, for a time at least, with an intake of only 0-07 to 0-08 gramme of nitrogen (0-4 to 0-5 gramme of protein) per kilo of body-weight, or not much more than one-third of the amount in Ranke's diet. It is obvious that the endogenous protein katabolism sets the limit below which it must be impossible permanently to reduce the allowance of protein. But it would be very hazardous to assume that this theoretical minimum limit corresponds with the permissible physiological limit. From ex- periments on men of various callings extending over many months, Chittenden has concluded that the average man eats at least twice as much protein as he really requires. We have already seen that the amount of nitrogen required to repair the actual waste of the tissues is comparatively small, and that with the ordinary amount of protein in the food a very large fraction of the total nitrogen is rapidly excreted as urea. There is no doubt, also, that many persons consume too much protein, at any rate in the form of animal food, and would feel better, work better, and probably live longer, if they restricted themselves in this regard. But there is no evidence that the digestion of such quantities of protein as the average healthy man eats, or the elaboration and excretion of the corresponding amounts of urea, ' strain ' in the least the digestive apparatus, the liver, or the kidneys. And it may just as well be argued that it is advantageous that much more than the minimum protein requirement should be offered to the tissues, so that the appropriate amino-acids, even the scarcest of them, may be sure to be present in sufficient amount, rather than that the organs should be subjected to the unnecessary ' strain ' of reconstructing some of the amino-acids themselves, supposing that they possess this power. In a question of this sort the immemorial experience and instinct of mankind cannot be lightly waved aside. McCay points out that while Bengalis in Lower Bengal subsist on food containing only about one-third the amount of protein in such a ' standard ' diet as Voit's (6 to 7 grammes of nitrogen a day), and may therefore be supposed to be immune from the dangers of an excessive protein metabolism, the large intake of carbo-hydrate rendered necessar y by the poverty of the food in protein is associated with perhaps g reater evils, among them a marked predisposition to diabetes and renal troubles. Their weight, chest measurement, and muscular development are inferior to those of other Asiatics living in the same climate, but with dietetic habits or economic conditions which ensure them a larger supply of protein. Thus the natives of Behar, with a larger intake of nitrogen, derived from wheat, and the natives of Eastern Bengal with a larger intake of nitrogen, derived DIETETICS 625 from wheat and fish, are physically much superior to the rice-eating Bengalis of Lower Bengal, although all belong to the same race. If we decide the matter merely on physiological grounds, we may say that for a man of 70 kilos, doing fairly hard, but not excessive, work, 15 grammes nitrogen and 250 grammes carbon are a sufficient allowance. The 15 grammes nitrogen will be contained in 95 grammes dry protein, which will also yield 50 grammes of the required carbon. The balance of 200 grammes carbon could theoretically ~ be supplied either in 450 grammes starch or in 260 grammes fat. But it has been found by experiment and by experience (which is indeed a very complex and proverbially expen- sive form of experiment) that for civilized man a mixture of these is necessary for health, although the nomads of the Asian steppes, and the herdsmen of the Pampas, are said to subsist for long periods on flesh alone, and a dog can live very well on proteins* and fat. The proportion of fat and carbo-hydrates in a diet may, however, be varied within wide limits. Probably no ' work ' diet should contain much less than 40 grammes of fat, but twice this amount would be better; 80 grammes fat give about 60 grammes carbon, so that from proteins and fat we have now got no grammes of the necessary 250, leaving 140 grammes carbon to be taken in about 310 grammes starch, or an equivalent amount of cane-sugar or dextrose. Adding 30 grammes inorganic salts, we can put down as the solid portion of a normal diet sufficient from the physiological point of view for a man of 70 kilos : 95 grammes proteins - - =T&, 2 s s •a C «« >>o S = c« ^S S 1^ «-J8 S * 3 S g > §S S5 ^ c ^ o^-°e " ^So^ a*1*? •§§ -3 rt r- <" .: -5 M .5 w S -3 N C -™ , r* r-> *»^ *-* j *" " ilssia^; »-* « d » *S „ "C l^lm1 ?se« W^P§ 8) ** ** -el « O 3 |iP.5il.fti!| ADRENALS 663 the latent period of the secretion is so short that the reaction follows the commencement of stimulation of the nerve after a time-interval sensibly the same as when it is elicited by the corresponding amount Fig. 213. — -20, Pocket Experiment with Stimulation of ^Right Splanchnic in Abdo- men after Section of Both Splanchnics. 21, repetition of 20, but with a shorter time of stimulation. The epinephrin rise of blood-pressure after 20 is consider- ably greater than after 21. Time-trace, half-minutes (f reduction). of adrenalin injected at the level of the adrenal veins. Even when the glands have ceased to respond to stimulation of the nerves, epinephrin can still be liberated into the blood by massaging the glands with the fingers (Fig. 214). Fig. 214. — Showing the Effect of Massage in Liberating Epinephrin from the Left Adrenal whose Nerves had been Cut Five Weeks before. 27, cava pocket with massage, the left adrenal vein being open to the pocket. 47 to 49, pocket with massage, the left adrenal vein having been closed at 45 and opened at 48, after' closure of the pocket. Massage begun at 45-A, stopped at 46. Blood-pressure tracing from carotid artery. Time-trace, half-minutes (reduced to $). A centre for epinephrin secretion seems to exist in the upper part of the thoracic region of the spinal cord. After section of the cord in the cervical region, even as low as the level of the last cervical segment, the secretion persists, but it is abolished when the cord is divided at the third or fourth dorsal segment. 064 INTERNAL SECRETION— ENDOCRINE GLANDS The relation of the nervous system to the adrenal medulla has been further illustrated by comparison of the amount of epinephrin which can be extracted from the two adrenals when the nerves of one have been cut. Under the influence of anaesthetics and some other drugs, bac- terial toxins, such as diphtheria toxin, etc., the epinephrin store is markedly diminished in the gland whose nerve supply has been left intact as compared with the other. For example, in a cat the fibres coming from the sympathetic to the left semilunar ganglion, including the left splanchnic nerves, were cut, and the animal allowed to recover. Several days later when it is known that equality of epinephrin load in the two adrenals would have been restored, morphin was administered to the animal, and after a few hours it was killed, and the epinephrin in the adrenals assayed. The left gland contained 0-28 milligramme, the right, o-ii milligramme. If the animal had been suddenly killed without morphin, the two adrenals would have been found to contain equal amounts of epinephrin. There- fore the deficiency in the right gland must have been due to the fact that it was not protected by section of its nerves. A similar deficiency has been found in the unprotected gland aftei administration of a drug, /3-tetrahydronaphthylamine, after surgical operations (post-operative deficiency), and in many other conditions. When the animal recovers the deficiency is gradually made good, the adrenal having the power of forming and storing epinephrin up to a ce "tain point in the absence of its nerve supply, although it is unable to liberate it into the blood. There is no foundation for the statement that emotional disturb- ances, as fright or anger, cause a depletion of the epinephrin store of the adrenals. Summary : — The existence of a nervous mechanism through which the gland is stimulated to secrete epinephrin into the blood can there- fore be considered as definitely established. It is the function of the epinephrin, once it has entered the circulation that is involved in doubt. Although it is highly improbable that the concentrations necessary for direct stimulation of sympathetic endings ever exist in the general mass of the blood, it is quite possible that a ' sensitizing ' influence is exerted by epinephrin upon the sympathetic peripheral mechanisms which renders them more susceptible to nerve impulses originating in other ways. The possibility of a more direct action of epinephrin upon the chemistry of carbo-hydrate metabolism, particularly upon glycogenolvsis or even upon glycogen formation has been emphasized by some writers. But glycogen can be formed abundantly, and certain experimental hyperglyccemias which depend upon the rapid transformation of glycogen into dextrose can be produced after the epinephrin secretion of the adrenals has been abolished. Chemistry and Formation of Adrenalin. — It has been shown (Stolz, Dakin) that adrenalin (C9H33NO3) is a dioxyphenyl-ethanol-methy lamin, CH OH.C/\C.CH(OH).CH2NH.CH, OH.C\/CH CH ADRENALS 665 It has been prepared synthetically, and in the body appears to be formed, probably by the introduction of a methyl (CH3) group, from a compound arising from an aromatic amino-acid (tyrosin or phenyl- alanin). While the natural adrenalin rotates the plane of polarization to the left, the artificial substance is optically inactive. This is because it consists of equal parts of laevo-rotatory and dextro-rotatory adrenalin. The artificial adrenalin has approximately half the effect of the natural on the blood-pressure, from which it may be inferred that the dextro- rotatory isomer has only a very slight pressor effect. The left and right rotatory moieties have been separated. The former has exactly the same power of raising the blood-pressure as the natural adrenalin, the latter only ^ to -^s as much. Practically the same proportion holds when the power of the two isomers in producing glycosuria is compared. This constitutes important corroboration of the view already referred to (p. 550), that adrenalin glycosuria is caused by an action on the sympathetic system, for the effect on the blood-pressure is known to be thus produced (Cushny). It is in the medulla of the adrenals that the epinephrin is formed. The medullary cells contain a substance which gives a yellow or brown stain with chromic acid or chromates, and for this reason the cells are called chromaffin or chromaphil. Similar cells are found elsewhere in the body — e.g., within the sympathetic ganglia, and also strung out in clumps along the course of the abdominal aorta below the level of the adrenal glands (Vincent). These outlying masses of chromaffin tissue appear to contain epinephrin, or a substance with similar physio- logical actions, so that the formation of this compound seems to be a property common to chromaphil tissue, no matter what its situation may be. A remarkable fact, and one calculated to induce caution in assigning a physiological function to epinephrin, is that the so-called parotid gland of a Jamaican toad secretes it in a concentration not much short of 5 per cent. (Abel). Function of the Adrenal Cortex. — The function of the cortical cells is obscure, but there is evidence that they, and not the chromaffin cells of the medulla, are concerned in the production of the internal secretion, whatever its nature may be, the loss of which leads so speedily to death on removal of the adrenals. For example, the period of survival after this operation is practically unaffected by the continuous intravenous administration of adrenalin, although the loss of the medulla- might be supposed to be compensated in this way. Still more convincing is the fact already referred to that after removal of one adrenal and denervation of the other in cats, epinephrin ceas>es to be discharged in detectable amount, and yet the animal survives in good health. In Addison's disease adrenalin is likewise powerless. The weakness (asthenia) of the skeletal, and to some extent of the cardiac, musculature which is characteristic of Addison:s disease, is produced experimentally within a few hours by ligation of both adrenals, and all the evidence goes to show that it is the cortical and not the mediillary region which is related to this disease. When the adrenals are not completely extirpated, compensatory hypertrophy of the re- maining portions may occur, and the animal survive indefinitely. In such cases it has been found that the hypertrophy is confined to the cortex. In animals like the rabbit in which accessory adrenals are not un- common, when these are present complete removal of both adrenals does not cause death. Now, the accessory adrenals consist of cortical substance without medulla. Portions of the adrenal are readily grafted under the skin, but only the cortex survives and grows. 666 INTERNAL SECRETION— ENDOCRINE GLANDS The functional difference between cortex and medulla is easily under- stood when we reflect that the morphological history of the two tissues is quite different. The medulla is developed from cells which push their way into the gla id from the rudiments of the sympathetic ganglia at that level, and is therefore of ectodermic origin. The cortex is derived from the same mesodermic structure which gives rise to the kidneys and genital organs. Embryologically, the cortical cells are related to the interstitial cells of the ovaries and of the testes, and like these are characterized by their richness in lipins (or lipoids). There is some evidence of an inter- relation between the thyroid and all these groups of sex cells (interstitial cells of ovary and testes, cortex of adrenal). Removal of a large part of the adrenals in the rabbit causes slight though definite hypertrophy of the thyroid and lymphoid hyperplasia. This is also seen in Addison's disease in man. Pituitary Body or Hypophysis. — In the pituitary body two parts essentially different in origin and function may be distinguished: (1) The large anterior lobe, or pars anterior, consisting of epithelial cells, many of which are filled with granules of the type seen in glandular epithelium, and abundantly provided with bloodvessels; (2) the smaller posterior or nervous lobe, or pars nervosa, also called the infundibular portion, consisting chiefly of neuroglia, the whole connected with the floor of the third ventricle by a stalk called the infundibulum. A further subdivision of the epithelial portion is made into the anterior lobe proper and the pars intermedia or intermediate lobe, con- sisting of epithelial cells, less granular and less richly supplied with bloodvessels than those of the pars anterior. The pars intermedia forms an epithelial investment of the pars nervosa, almost completely surrounding it and throwing out offshoots of epithelial cells into its substance, which is also invaded by colloid material secreted by the cells of the intermediate lobe. The differences in the structure of the anterior and posterior lobes of the pituitary body correspond to a difference in their development. The anterior lobe is developed (in man in the fourth week of intra-uterine life) from an ectodermal pouch (Rathke's pouch), which is pushed up from the roof of the bucco- pharyngeal cavity towards the mid-brain. The posterior lobe is developed from an extension of the neural ectoderm, which grows back- wards as the infundibular process till it meets and blends with that portion of the buccal pouch which gives rise to the pars intermedia. The pars intermedia is separated from the anterior lobe proper by a cleft which represents what is left of the lumen of Rathke's pouch. In connection with the interpretation of the results of experiments on removal of the pituitary body, it is of consequence to remember that a residue of the same epithelium which develops into the anterior lobe appears always to get cut off in the vault of the pharynx, constituting the so-called pharyngeal hypophysis, and consisting of a cord of cells identical with those of the anterior lobe (Haberfeld). Embryonic ' rests ' of hypophyseal tissue are also often found in the dura of the sella turcica, in which the pituitary body lies, and in the body of the sphenoid bone. Cells of the intermediate lobe also run up the stalk of the infundibulum, and even stretch for a little distance along the floor of the third ventricle. Add to this the formidable nature of the ADRENALS 667 operations required for the extirpation of the hypophysis from its sheltered position within the skull, and it will not be wondered at that complete harmony has not been attained as to the consequences of its removal. The best evidence at present is to the following effect: When the pituitary body is completely removed, death speedily and invariably ensues, in dogs, on the average within twenty-four to forty-eight hours. Puppies often live as long as two or three weeks. The much longer periods of survival occasionally witnessed are due to failure to remove some small portion of the hypophyseal epithelium. On the day after the operation the animals may be able to walk about, to eat and drink, and may show an interest in their surroundings. The temperature, pulse, and respiration at this time may be normal. Soon, however, they become lethargic, then comatose, with characteristically incurved spine, slow respira- tion, with long-drawn inspiration, a feeble pulse, perfectly limp muscles, and often a subnormal temperature, and the appearance of sugar in the urine. This deep coma passes into death, with no perceptible transition, and without a struggle (Paulesco, Gushing). The ablation of a part of the cortical substance of the anterior (epithelial) lobe of the hypophysis is compatible with permanent survival, and gives rise to no symptom of disorder. The same is true when only the posterior lobe is removed. This does not seem to be followed by any recognizable symptoms. In some animals, however, kept tinder observation for long periods after partial removal of the anterior lobe, a marked tendency to accumulate fat has been noted, accompanied by hypoplasia of the generative organs in adults or the persistence of the infantile condition in immature animals. On the other hand, complete removal of the anterior lobe causes death, just as if the whole gland had been taken away. Of all the structures included in the pituitary body, the most important from the functional point of view appears to be the superficial layer of the anterior lobe. Mere separation of the stalk of the hypophysis may produce effects sometimes as serious as those of total removal of the gland, probably owing to the disturbance caused in the circulation. It is stated, indeed, by some observers that the vulnerable point is the base of the infundibulum, and that if this is not injured extirpation of the hypophysis is not incompatible with continued existence, and that in adult animals the resultant changes are only slight, although much more pronounced, especially as regards the disturb- ances in metabolism and development in young animals. It has been asserted that the pituitary undergoes (compensatory ?) hyper- trophy after thyroidectomy. Some observers have accordingly assumed a similarity of function for these organs. It has even been stated that the production of colloid material by the cells of the pars intermedia is increased, and that colloid accumulates in the 668 INTERNAL SECRETION— ENDOCRINE GLANDS nervous portion of the posterior lobe. But this colloid, whatever its function may be, is very different from that of the thyroid alveoli, for the (sheep's) pituitary contains no iodine after extir- pation of the thyroid any more than before (Simpson and Hunter). And in man pathological changes (tumours) in the pituitary body are associated, not with myxcedema, or other disease connected with changes in the thyroid, but frequently with another condition, called acromegaly, in which the bones of the limbs and face, especially the hands and feet and the lower jaw, become hypertrophied. Another condition often associated with tumours of the pituitary is gigantism — a condition occurring before the normal growth of the bones is completed, and resulting in a great increase in the length of the bones both in the limbs and the trunk. Action of Intravenous Injection of Extracts of the Pituitary. — The effects on the vascular system of intravenous injection of extracts of the pituitary gland are also very different from those caused by thyroid extracts. The posterior lobe, or infundibular body, including the pars intermedia, contains two active substances, one pressor and the other depressor. The former is soluble in salt solution, but insoluble in absolute alcohol and ether; while the latter is soluble in salt solution as well as in alcohol and ether. The pressor substance (obtained in fairly pure form in the preparation called pituitrin, and in still greater concentration in a preparation to which the name hypophysin has been given) causes a rise of blood-pressure, due partly to constriction of the arterioles and partly to an increase in the force of the heart-beat, both of which are brought about by direct action. This rise of pressure lasts for a considerable time, and is sometimes accompanied by a slowing of the heart. A second dose injected before the effect of the first has passed off is inactive ; and this distinguishes the pituitary from the suprarenal extract. Associated with the pressor effect is an increase in the flow of the urine. Whether this is due to a separate diuretic substance, as some maintain, has not been definitely settled. Indeed, the factors on which the diuretic action depends are complex, and sometimes in spite of an increase of general arterial pressure, and of the volume of the kidney, there is no increased flow of urine. In diabetes insipidus pituitrin has been found to produce pre- cisely the opposite effect — namely, diminution in the excessive urinary flow. The pressor substance, unlike adrenalin, directly stimulates smooth muscle fibres (especially the arteries, uterus, and spleen) irrespec- tive of their innervation ^Dale). Many other points of difference exist between the pressor principles of the infundibular body and the adrenal medulla. For example, pituitrin constricts the pupil and diminishes the flow of saliva from the submaxillary gland, whereas adrenalin dilates the pupil and causes an increased flow of saliva. Hypophysin and other preparations of the posterior lobe have been employed to stimulate the uterine contractions in obstetrical practice, but their use does not seem to be free from ADRENALS 669 danger. . When injected intramuscularly, regular1 and powerful con- tractions of the uterus are excited in two or three minutes at any stage in parturition. Another effect of extracts of the infundibular body is on the mammary gland. An increased flow of milk follows the injec- Fig. 215. — Action of Extract of Infundibular Lobe (Pars Nervosa of Ox Pituitary). P, Carotid pressure; K, kidney volume; U, drops of urine; S, signal showing point at which pituitary extract was injected; T, time- trace in zo-second intervals. T is also the zero of the blood-pressure (Hering). tion of pituitrin (Simpson). But it is in doubt whether this is due to increased secretion and not solely to stimulation of the smooth muscle of the organ with expulsion of milk already formed. The depressor substance produces a marked fall of blood-pressure, even when it is injected during the rise of pressure caused by an injection of the pressor Fig. 216. — Action of Extract of Hypophyseal Lobe of Pituitary on the Blood- Pressure (W. W. Hamburger). The signal line at the top shows the time and length of injection of the saline extract into the blood. Time-trace (at bottom) shows second intervals. The figure is to be read from left to right. substance. The anterior lobe, or hypophysis, also contains a depressor substance. Intravenous injection of a saline extract causes a distinct fall of blood-pressure, accompanied usually by acceleration and weaken- ing of the heart (Fig. 216). A second injection immediately following the first produces no change in the pressure (W. W. Hamburger) . 670 INTERNAL SECRETION— ENDOCRINE GLANDS An active substance which has been termed ' tethelin,'* because of its influence on growth, has been separated from the anterior lobe. On intravenous injection into rabbits, it causes a slight fall of blood pres- sure and rto diuresis. Its effects are therefore quite different from those of the posterior lobe. Mixed with the food f* /I"! /\ °^ m*ce' *t has Deen found to exert a charac- '.' rJ / '* teristic influence upon their growth, retard- ing it at certain stages, and accelerating it at other stages (Robertson) . It is not at present possible to deduce from such clinical and experimental observations as those described any co- herent theory of the function of the pituitary. That there is some connection between the normal action of the gland, and in particular of its Anterior lobe, and the normal growth and nutrition of the body, including the skeleton, is scarcely to be doubted. The fact that administra- tion of the dried gland substance to dogs causes an increased excretion of calcium on a diet rich in calcium is a further indication of its influence on the meta- bolism of bone (Malcolm). But so far is the precise nature of this influence, if it exists, from being fully understood, that authorities of repute are still divided on the question whether the symptoms of acromegaly and gigantism are due to atrophy or to hypertrophy of the active 3of Fig. 217. — Action of Infundi- bular Extract upon Virgin Uterus of Guinea- Pig. The same dose was applied three times in succession to an iso- lated segment, at the points marked by the arrows at the bottom of the curves. After each application the segment was washed with Ringer's solution at R. The tracings have been made to overlap. (Reduced to one- half.) (Dale.) elements of the gland, to loss of its internal secretion, or to its manufacture in excessive amount. There is evidence that the colloid secretion of the posterior lobe, probably formed by the epithelial cells of the pars intermedia, passes through the nervous portion to enter the infundibulum and the third ventricle of the brain, where it breaks down in the cerebro-spinal fluid (Hering). And it has been suggested that in virtue of the action of the hormones (p. 404) in this secretion on the vascular system in general, and on the renal cells and the renal circulation in particular, the posterior lobe constitutes a mechanism for the control of the secretion of urine. But this suggestion is still in the realm of hypothesis. Some support is given to it by the observation that continued irritation of the posterior lobe by a plug of the gutta-percha compound used by dentists for temporarily filling teeth introduced through the floer of the sella turcica caused permanent polyuria in dogs, analogous to the diabetes insipidus seen in man (Matthews). * From T£0/;Xu>s, growing, nourishing. 67i Pineal Gland. — Extracts of the pineal gland injected into the circula- tion have no effect other than that due to the inorganic constituents of the calcareous concretions or ' brain sand,' which are its character- istic feature. Since in early life the organ has a glandular structure which is later replaced by fibrous tissue, it has been supposed that it may exercise some function in connection with growth. But so far the physiology of the pineal body is practically a blank sheet, or, at best, a budget of contradictory statements from which nothing certain can be deduced. Thus in two of the most recent papers, each based on a large number of careful experiments, one author concludes that removal of the gland in male guinea-pigs is associated with hastened development of the sexual organs, and in females with a tendency to breed earlier than the normal controls (Horrax). The other observer finds that feeding young guinea-pigs with pineal tissue from young animals determines an earlier sexual maturity than normal, and in- creases the rate of growth of the body (McCord) . The alleged influence of the invasion of the gland in young children by pathological growths in accelerating the development of the skeleton and reproductive organs, which has been supposed to indicate that it normally exerts a restraining or regulative influence on this develop- ment, is at present purely fanciful. Kidney. — The experiments of Bradford, which seemed to indicate that the kidney, in addition to its function as an excretory organ, plays an important, and indeed indispensable, part in protein metabolism, possibly by forming something of the nature of an internal secretion, Fig. 218.— Effect of Bone-Marrow on Blood-Pressure. Intravenous Injection of Saline Extract Vagi Intact. The uppermost line is a signal trace showing the time and length of injection. Below this is the record of the respirat ^ry move- ments, and lowest the blood-pressure tracing. To be read from leU to right. have not been confirmed. He stated that, when the half or two-thirds of one kidney and the whole of the other have been removed from a dog by successive operations, death ensues, although the quantity both of water and urea excreted by the fragment of renal substance that remains is far above the normal. In spite of the increased elimination S^A thHatfsubs,tance ,was ^d to accumulate in the tissues.Towing that the destruction of protein was increased— a conclusion which seemed to derive support from the wasting of the animal. It has since been shown that an increased output of nitrogen is not of constant occurrence and only takes place under the same conditions aT^ starvation (p. 603). As a matter of fact, the animals waste and d£ 672 INTERNAL SECRETION— ENDOCRINE GLANDS within a few days or weeks largely because they refuse to eat. Polyuria (increase of urine beyond the normal) does not necessarily occur. It is well known that when only one kidney is extirpated the other hyper- trophies, and no ill-effects ensue. The statement that extracts of the kidney when injected into the veins of an animal cause a rise of arterial blood-pressure, essentially through direct action on the peripheral vaso-motor mechanism, is of considerable interest, for it may possibly have some bearing on the rise of pressure and consequent hypertrophy of the heart associated with certain renal diseases. But there is not as yet sufficient evidence that the hypothetical pressor substance, to which the name ' renin ' has been given, in any sense represents an internal secretion of the kidney. .^^ The pressor substance (so-called ^JpM||L ' urohypertensine ') which can be Jfr ^\ extracted by ether from normal ^Jf \ human urine (Abelous) is probably Ajyt^Mr \| only excreted by the kidney, and KP^^ ^\ ^.^ perhaps arises from the putrefac- \jM|N ^on °* proteins in the intestine. If For it has been shown that in the putrefaction of (horse-) meat bases U«^^ rijfrf , _ ••— — blood-pressure. The most active Fig. 219. — Injection *of Extract of Bone- of these is a body known as Marrow with the Vagi Cut. To be read ^-hydroxyphenylethylamine, from left to right. formed from tyrosin (Barger and Walpole). Whether the pressor (vaso-constrictor) substance which appears to be liberated from the platelets when blood is shed, and may therefore be presumed to be more slowly liberated from such platelets as normally break down in the circulating blood, has any relation to the pressor substance of urine is unknown. It is also quite uncertain whether, as has been stated by some observers, extracts of the kidney or blood from the renal vein stave off for a time the onset of the uraemic symptoms that follow removal of both kidneys or ameliorate them when they have already appeared. Spleen. — The spleen does not produce an internal secretion necessary to life, for it can be removed both in animals and in man, without the development of serious symptoms. Its blood- forming and blood-destroying functions (p. 22) are taken on by other structures (particularly the red bone-marrow). The most definite changes following splenectomy are transient anaemia, increased resistance to haemolytic agents, and an increase in the content of cholesterol in the blood (Pearce). These effects are more pronounced in the young. This is intelligible if the spleen normally plays a considerable part in the destruction of worn-out erythrocytes with liberation of haemoglobin, the source of the bile- pigment. The formation of the bile-pigment is said to be inter- fered with, and its amount reduced by more than 50 per cent. (Pugliese). The spleen can be auto-transplanted readily into the subcutaneous tissues of animals. There is a great difference in the behaviour of ADRENALS 673 such transplants according to the age of the animal (Marine and Manley). In young rabbits, the transplants grow rapidly if the spleen is removed at the time of transplantation, while in sexually mature rabbits they do not grow and often undergo gradual absorp- tion. Likewise, transplants which have grown rapidly in young animals decrease in size after adult life is reached, indicating that the function of the spleen is more necessary in young than in older animals ; or that its function is more easily and completely assumed by other tissues, probably the bone marrow, in adults. A further and very suggestive fact, is that a subcutaneous spleen graft which has ' taken,' but is not growing, in a young animal whose spleen has not been removed, can be made to grow by removal of the spleen. These results indicate that splenic insufficiency is a necessary con- dition of growth of a splenic transplant, just as thyroid insufficiency is a necessary condition of growth of a thyroid transplant. The stimulus to growth in the one case, as in the other, must be assumed to be a chemical stimulus transmitted through the blood, and not dependent upon the nervous system. The salivary glands may be extirpated without any sensible change being produced in the normal metabolism. It has been stated how- ever, that the secretion of the gastric juice is diminished. It has been supposed that this may be due to the absence of a hormone (p. 404) normally produced in the salivary glands. A temporary increase in the gastric secretion is said to be caused when extracts of the glands of normal dogs are injected into the veins or into the peritoneal cavity of dogs deprived of their salivary glands (Hemmeter). But later experiments contradict the theory of the existence of a hormone in the salivary glands, which stimulates the secretion of gastric juice. The average rate of secretion into a Pawlow gastric pouch (p. 403) was not diminished in dogs after extirpation of the glands (Swanson.) Extracts of nervous tissue (sciatic nerve, white matter of brain, and spinal cord, but especially grey matter of brain) cause, on injection into the veins, a decided fall of arterial blood-pressure, which soon passes off, and can be renewed by a fresh injection. The fall of pressure is due to direct action upon the bloodvessels of a depressor substance in the extracts, and not to the action of vaso-motor nerves. It can be obtained after section of the vagi. Extracts of muscular tissue also cause a distinct though transient fall of pressure, but not so great a fall as in the case of extracts of nervous tissue. Saline decoctions of other tissues (testis, kidney, spleen, pancreas, liver, mucous membrane of stomach and intestine, lung, and mammary gland) all produce a fall of blood-pressure (Osborne and Vincent). The same is true of bone marrow (Brown and Guthrie; Figs. 218, 219). It must be repeated that there is no evidence that these depressor substances are specific internal secretions in the same sense as epinephrin. 43 CHAPTER XII ANIMAL HEAT FROM the earliest ages it must have been noticed that the bodies of many animals, and particularly of men, are warmer than the air and than most objects around them. The ' vulgar opinion ' of Bacon's time, ' that fishes are the least warm internally, and birds the most,' if it does not imply a very extensive knowledge of animal temperature, at least shows that the fundamental distinction of warm and cold-blooded animals, which is to-day more accurately expressed as the distinction between animals of constant tempera- ture (homoiothermal) and animals of variable temperature (poi kilo- thermal), had been grasped, and was even popularly known. Since that time the accumulation of accurate numerical results, and the advance of physical and physiological doctrine, have given us definite ideas as to the relation of animal heat to the metabolic processes of the body. It is impossible to understand the present position of the subject without an elementary knowledge of the science of heat. For this the student is referred to a textbook of physics. All that can be done here is to preface the physiological portion of the subject by a few remarks on the physical methods and instruments employed: SECTION I. — THERMOMETRY AND CALORIMETRY. Temperature. — Two bodies are at the same temperature if, when placed in contact, no exchange of heat takes place between them. They are at different temperatures if, on the whole, heat passes from one to the other, and that body from which the heat passes is at the higher temperature. It is known by experiment that if two bodies of different temperature are placed in contact, heat will pass from one to the other till they come to have the same temperature. If, then, we have the means of finding out the temperature of any one body, we can arrive at the temperature of any other by placing the two in con- tact for a sufficiently long time, under the proviso that the quantity of heat necessary to bring the temperature of the first body, which may be called the ' measuring ' body, to equality with that of the second is so small as not to make a sensible difference in the latter. This is the principle on which thermo metric measurements depend. A mercurial thermometer consists of a quantity of mercury ordinarily contained in a thin glass bulb, the cavity of which is continued into a tube of very 674 THERMOMETRY AND CALORIMETRY 675 fine bore in the stem. Like most other substances, mercury expands when the temperature rises, and contracts when it sinks, and the amount of expansion or contraction is shown by the rise or fall of the mercurial column in the stem of the thermometer. The point at which the meniscus stands when the bulb is immersed in melting ice or ice-cold water is, on the centigrade scale, taken as zero ; the point at which it stands when the thermometer is surrounded by the steam rising from a vessel of boiling water is taken as 100 degrees. The intermediate portion of the stem is divided into degrees and fractions of degrees. When, now, we measure the temperature of any part of an animal with such a thermometer, we place the bulb in contact with the part until the mercury has ceased to rise or fall. We know then that the mercury has ceased to expand or contract, and therefore that its temperature is stationary, and presumably the same as that of the part. It is to be noted that we have gamed no information whatever as to the amount of heat in the body of the animal. We have only observed that the mercury of the thermometer when its temperature is the same as that of the given part expands to an extent marked by the division of the scale at which the column is stationary. And we know that if the mercury rises to the same point when the thermometer is applied to another part, the temperature of the latter is the same as that of the first part; if the mercury rises higher, the temperature is greater; if not so high, it is less. The thermometer, then, only informs us whether heat would flow from or into the part with which it is in contact if the part were placed in thermal connection with any other body of which the temperature is known. In other words, the temperature is a measure of the heat ' tension,' so to speak; and difference of tempera- ture between two bodies is analogous to difference of potential between the poles of a voltaic cell (p. 724), or to difference of level between the surface of a mill-pond and the race below the wheel. The temperature of an animal is measured in one of the natural cavities, as the rectum, vagina, mouth, or external ear, or in the axilla, or at any part of the skin. For the cavities a mercury thermometer is nearly always used ; the ordinary little maximum thermometer is most convenient for clinical purposes. The temperature of the skin may be measured by an ordinary mercury thermometer, the outer portion of the bulb of which is covered by some badly conducting material. An uncovered thermometer, heated nearly to the temperature expected, will also give results sufficiently accurate for most purposes, especially if the bulb is flat or in the form of a flat spiral, which can be easily applied to the surface. A theoretically better method, but more laborious in practice, is the use of a thermo-electric junction, or a resist- ance thermometer formed of a grating cut out of thin lead-paper or tin- foil (Fig. 220). This is especially useful for comparing the temperature of two portions of skin . The temperature of the solid tissues and liquids of the body may also be measured or compared by the insertion of mer- curial or resistance thermometers or thermo-electric junctions (p. 764). Calorimetry. — The quantity of heat given off by an animal is generally measured by the rise of temperature which it produces in a known mass of some standard substance. Sometimes, however, as in the ice- calorimeter of Lavoisier and Laplace, and the ether calorimeter of Rosenthal, a physical change of state — in the one case liquefaction of ice, in the other evaporation of ether — is taken as token and measure of heat received by the measuring substance, the number of units of heat corresponding to liquefaction of unit mass of ice or evaporation of unit mass of ether being known. The unit generally adopted in the measurement of heat is the quantity required to raise the temperature 676 ANIMAL HEAT of a kilogramme of water i° C., which is called a calorie,* or kilocalorie or large calorie. The thousandth part of this, the quantity needed to raise the temperature of a gramme of water by i°, is termed a small calorie or millicalorie or gramme-calorie. In the calorimeters which have been chiefly used in physiology either water or air has been taken as the measuring substance. The simplest form of water calorimeter is a box with double walls, the space between which is filled with a weighed quantity of water. The animal is placed inside the vessel, and the temperature of the water noted at the begin- ning and end of the experiment. Suppose that the quantity of water is 10 kilos, and that the temperature rises i° in thirty minutes, then the amount of heat lost by the animal is 10 calories in the half -hour, or 480 calories in the twenty -four hours; and if the rectal temperature is unchanged, this will also be the amount of heat produced. Here we assume (i) that all the heat lost by the animal has gone to heat the water and none to heat the metal of the calorimeter; (2) that none has been radiated away from the outer surface of the latter. The first assumption will seldom introduce any sensible error in a prolonged physiological experiment ; but it is very easy to determine by a separate observation the water-equivalent of the calorimeter — that is, the quantity of water whose temperature will be raised i ° by a quantity of heat which j ust suffices to raise the temperature of the metal by i° (p. 721). Then the water-equivalent is added to the quantity of water actu- ally present, and the sum is multiplied by the rise of temperature. If the tempera- ture of the room is constant, as will be approximately the case in a cellar, any error due to interchange of heat between Fig. 320. — Resistance Thermom- eter for measuring Temperature of Skin. G, grating of lead- paper, attached to a cover-slip, and mounted on a holder; W, W, wires to the Wheatstone's bridge. An increase of tem- perature causes an increase in the resistance of the lead. The balance of the bridge is thus dis- turbed. By experimental grad- uation the temperature value of the deflection, or of the change of resistance that balances it, is known (p. 725). the calorimeter and its surroundings may be eliminated by making the initial temperature of the water as much less than that of the air as the final temperature exceeds it. Then if the loss of heat by the animal is uniform, as much heat is gained during the first half of the experiment by the calorimeter from the air as is lost by it to the air during the last half. Or, without lowering the temperature of the water, the amount of heat lost by the calorimeter during an experiment may be previously determined by a special observation, and added to the quantity cal- culated from the observed rise of temperature. Or, finally, two similar calorimeters may be used, one containing the animal and the other a hydrogen flame, or a coil of wire traversed by a voltaic current, which is regulated so as to keep the temperature the same in the two calorim- eters. From the quantity of hydrogen burnt, or electricity passed, the heat-production of the animal can be calculated. In Atwater's great respiration calorimeter (Fig. 221) both the heat production and the respiratory exchange are measured. The heat pro- duced by the person in the calorimeter is carried away from it by a * The student should carefully note that when the term ' calorie ' is used without qualification, a large calorie, i.e., 1,000 gramme-calories, is meant. THERMOMETRY AND CALORIMETRY 677 stream of water flowing through the chamber in a series of tubes, the temperature within the calorimeter being kept constant by regulating the temperature and velocity of the entering stream of water. The quantity of the escaping water and the increase in its temperature are measured, and the heat-production can then be calculated. The apparatus consists of a chamber in which a human being can live for several days and nights. A stream of air is supplied, and the chemical changes produced in. this are investigated in the manner already described (p. 240). Fig. 22 1. —Respiration Calorimeter (Atwater). Interior of chamber. A corner 2 ANIMAL HEAT and the nitrogen excretion are increased. These phenomena mav last for several days (Ott, Richet, Aronsohn, and Sachs), and are due to stimulation of the portions of the brain in the immediate neighbour hoed of the injury. Electrical stimulation of this region has a similar effect. When the temperature has returned to normal, a fresh puncture may again cause a rise. As to the manner in which these centres are excited, there is some evidence that, in addition to any influence exerted on them by afferent nerves, they are capable of being directly affected by the temperature conditions of the blood passing through them, as well as by numerous drugs. Thus it is stated by Barbour and Wing that direct application of cold to the region of the corpus striatum. especially its caudate nucleus, in rabbits causes a rise in the rectal temperature, associated with shivering and consequent increase of heat-production in the contracting muscles, and with peripheral vaso-constriction and consequent diminution in the heat-loss. The application of warmth has the opposite effect : the peripheral bloodvessels dilate, the animal becomes quiet, and the rectal tern perature falls.* Some observers hold that the chief seat of the increased metab olism in the puncture fever is the skeletal muscles, others the liver. The question turns largely upon the success of the puncture ex- periment after the previous administration of curara on the one hand, and of strychnine on the other. For curara cuts out the motor innervation of the skeletal muscles, and strychnine convul- sions exhaust the store of hepatic glycogen. Certain investigators have found that after an adequate dose of curara no puncture fever can be obtained, and they locate the increased metabolism asso- ciated with the fever in the muscles. Others maintain that even after curara the puncture is followed by fever, but is not followed by fever if strychnine has first been given. They accordingly con- clude that the rapid combustion of the glycogen (or the dextrose derived from it) is the primary factor in the increased metabolism.' It may be pointed out, however, that neither experiment is a crucial test. For if strychnine reduces the liver glycogen, it also reduces the glycogen of the muscles. And if in the puncture fever the liver glycogen is transformed into dextrose more rapidly than usual, the dextrose is probably in great part used up in the muscles more rapidly than usual, eiss it would appear in the urine. The effect of strychnine on the puncture fever, then, is no proof that the muscles are not essentially concerned in it. On the other hand, the alleged absence of the fever after curara is not sufficient to show that the muscles are alone concerned. For curara causes a lowering of the body-temperature, which, if it be not overcompensated, may * Other observers, however, have failed to confirm these results (Cloetta and Waser). PEVER 703 mask the fever. The positive result of the puncture in curarized animals, which some observers say they have obtained, would, if animals, which some observers say they have obtained, would, if confirmed, be important evidence that the primary effect is not on the muscles, or, at least, not solely on them, but would not prove that it is on the liver. That the liver is concerned, however, is more directly indicated by the fact that during the puncture fever the liver continues to be what it is under normal conditions, the warmest organ in the body, warmer than the blood in the root of the aorta by about i° C. The most probable conclusion is that the increased production of heat in this form of experimental fever is due to an increased metabolism of carbo-hydrate (glycogen) both in the liver and in the muscles. Temperature Regulation in Hibernating Animals. — The behaviour of hibernating mammals, such as the marmot, dormouse, hedgehog, and bat, is of interest in connection with the temperature regulation. In the active waking state these animals are homoiothermal, but in profound winter sleep they are poikilothermal, the body-tempera- ture rising and falling with that of the air. The rectal temperature may be as low as 2° C. There is an intermediate state in which the animal is partially awake, though inactive, and its temperature is much below the normal, but considerably above that of its environ- ment. In this condition it has an imperfect thermotaxis, something like that of an ordinary mammal (including the human infant) in the period of immaturity, immediately after birth. When the hibernating mammal awakes the rise of temperature is enormous and abrupt. The temperature of a dormouse rose in an hour from 13' 5° C. to 357° C., and that of a bat in fifteen minutes from 17° C. to 34° C. (Pembrey). Fevers. — Fever is a pathological process generally caused by the poisonous products of bacteria, and characterized by a rise of temperature above the limit of the daily variation (p. 712). It is further associated with an increase in the rate of the heart and the respiratory movements, and a diminution in the alkalies and carbon dioxide of the blood. The total excretion of nitrogen is increased, at least in proportion to the amount of protein ingested, indicating an increase in the consumption of tissue-protein. The distribution of the nitrogen among the urinary constituents is altered. The ammonia (in the form of ammonium salts of organic acids), the uric acid, and to a smaller extent the creatinin (Leathes), are in- creased, while the urea is relatively decreased, even when its abso- lute amount is greater than normal. Creatin, which is not normally present in urine, unless the food contains it, may also appear in fever (Shaffer). It has been suggested that the proximate cause of fever is the action of bacterial poisons or of other substances on the ' heat centres,' and that antipyretics, or drugs which reduce the temperature in fever, do so by restoring the centres to their normal 704 ANIMAL HEAT state, by preventing the development of the poisons, aiding their elimination, or antagonizing their action. In favour of this view, it has been stated that when the basal ganglia are cut off, by section of the pons, from their lower nervous connections, fever is no longer produced by injection of cultures of bacteria which readily cause it in an intact animal, while antipyrin has no influence upon the temperature (Sawadowski). And while it is almost certain that some pyrogenic or fever-producing agents — cocaine, e.g. — act indirectly, through the brain or cord, it is quite possible that others affect directly the activity of the tissues in general, just as some antipyretics or fever-reducing agents, such as quinine, act imme- diately upon the heat-forming tissues, so as to diminish their metabolism, while others, like antipyrin, affect them through the nervous system. Quinine has no influence upon ' puncture ' fever in rabbits. A still more important action of antipyrin, and the group of antipyretics to which it belongs, is the increase in the heat- loss which they bring about by the dilatation of the bloodvessels of the skin. This effect is also produced through the nervous system. Fever is a condition so interesting from a physiological point of view, and of such importance in practical medicine, that it will be well to consider a little more closely the possible ways in which a rise of temperature may occur. It must not be forgotten that the febrile increase of temperature is always accompanied by other departures from the normal, and that all the fundamental febrile changes may even, in certain cases, be present without elevation, and even with diminution of temperature. But here we have only to do with the disturbance of the normal equilibrium between the loss and the production of heat; and it is evident that any of the five conditions illustrated in the diagram (Fig. 226) may give rise to an increase of temperature. It is not necessary to discuss whether cases of fever can actually be found to illustrate every one of these possibilities. It is probable that not infrequently dimin- ished loss and increased production may be both involved; and it ought to be remembered that the healthy standard with which the heat-production of a fever patient should be compared is not that of a man doing hard work on a full diet, but that of a healthy person in bed, and on the meagre fare of the sick-room. When this is kept in view, the comparatively low heat-production and respiratory exchange which have sometimes been found in fever cease to excite surprise. But in any case, no mere change in the absolute quantities of heat formed and lost is sufficient to explain the febrile rise of temperature; there must be a change in the relative proportion. That an increase in heat-production is not of itself enough to produce fever is proved by the fact that severe muscular work, which in- FEVER 7°5 creases the metabolism more than high fever, only causes in a healthy man a rise of about i° C. in the rectal temperature. When the work is over, the temperature comes rapidly back to normal. The essence of the change in fever is a derangement of the mechanism by which in the healthy body excess or defect of average metab- olism, or of average heat- loss, is at once compensated and the equilibrium of tem- perature maintained. This derangement only lasts as long as the temperature is rising. -When it becomes stationary at its maximum we have again adjustment, again equality of production and escape of heat; but the adjustment is now pitched for a higher scale of tempera- ture. A rough analogy, so far as one part of the process is concerned, may be found in the behaviour of the ordinary gas-regulator of a water-bath. It can be ' set ' for any temperature. That temperature, once reached, remains constant within nar- row limits of oscillation; but the regulator can be equally well adjusted for a higher or a lower temperature. It is, however, important to note that the equilibrium is more unstable in fever than in health, so that changes of external temperature more easily depress or increase the temperature of a fever patient than of a healthy man. Rosenthal has concluded from calorimetric observations that, in the first stage of fever, while the temperature is rising, there is always increased retention of heat. Maragliano actually found evidence, by means of the plethysmograph, that the cutaneous vessels are at this stage constricted, and that the constriction may even precede the rise of temperature. The blood- flow in the feet in cases of typhoid fever investigated by the calorimetric method (p. 122) was not found to exceed the normal flow, and was usually decidedly below the normal. Hyperexcitability of the vaso-con- strictor mechanism of the peripheral parts, especially of the skin, 45 Fig. 226. — Diagram to show tha Possible Relations between Heat- Production and Heat -Loss in Fever. 706 ANIMAL HEAT was present. All these observations lend support to the famous * retention ' theory of Traube. It has been suggested that the significance of the increased action of the cutaneous vaso-constrictor mechanism in typhoid fever is that the peripheral vaso-constriction is a compensatory arrangement which secures for the organs mainly suffering from the infective process an increased flow of blood to combat the infection. On this hypothesis the rise of temperature in so far as it depends upon diminished loss of heat is a secondary phenomenon inevitably following the redistribution of the blood, and unavoidable except by a corresponding diminution in the total metabolism. In the great majority of cases the production of heat is also increased, on the average by 20 to 30 per cent, of the normal production of a resting man. The increase may be much greater during the chill which ushers in so many infections on account of the muscular contractions in shivering. During the period of rising temperature the production of heat is not necessarily increased. At the height of the fever there is often, though apparently not always, an increase in the heat-production. After the crisis, while the fever is subsiding, the rate at which heat is lost rises sharply. As to the explanation of the increase of metabolism in fever, and especially of the increased metabolism of tissue-protein, various views have been held. Some have gone so far as to say that the increase is merely the consequence, not the cause, of the rise of temperature. But the rebutting evidence which has been brought against this view is strong and, indeed, overwhelming. It is perfectly true that, when the temperature of the body is artificially raised by preventing the free loss of heat for a sufficient time (so-called physiological fever), the destruction of protein is augmented. A fasting dog whose temperature was increased to 40° or 41° C. for twelve hours eliminated 37 per cent, more nitrogen than when the body-temperature was normal. But this increase in the protein metabolism could be entirely prevented by giving the animal a sufficient amount of carbo-hydrate. Similar results have been obtained in man. The carbon dioxide excretion and oxygen absorption are, of course, also markedly increased. But the increase in the nitrogen excretion is often much greater in fever than any increase which can be brought about by artificially raising the temperature of a healthy individual by means of hot baths. A typhoid patient was found to lose 10-8 grammes of nitrogen a day (corresponding to 318 grammes of muscle) during eight days of fever (F. Miiller). A portion of the loss of nitrogen on the routine fever regimen may be due to the fact that the ordinary typhoid patient is really on a semi-starvation diet, the heat-equivalent cf which is not much more than half his heat-production. Yet it has not been found possible to completely prevent the loss of FEVER 707 nitrogen by putting the fever patient on a diet rich in protein, or on a diet containing a moderate amount of protein with a large quantity of fat and carbo-hydrate, even when the total heat- value of the diet is much in excess of the 32 or 33 calories per kilo of body- weight which corresponds to the heat-production of a resting man. Another suggestive fact is that the excessive excretion of nitrogen does not run parallel with the rise of temperature in fever, but is often most marked after the crisis. During the stage of defer- vescence an enormous amount of urea is sometimes given off. In a case of typhus, in the mixed urine of the third and fourth days after the crisis, no less than 160 grammes of urea was found (Naunyn), or nearly three times the normal amount for a man on full diet. Again, when fever is caused by the injection of bacteria or their products, the increase in the carbon dioxide eliminated and oxygen consumed occurs even when the temperature is prevented from rising by cold baths. It seems perfectly clear, then, that the increase of metab- olism is, in many cases at least, a primary phenomenon of fever. Its course and incidence, falling as it does so largely upon the proteins, the steady loss of tissue nitrogen, and the inability of the tissues to recoup their losses from the protein of the food or to shield their own protein by burning more carbo-hydrate or fat, all suggest that the cells are poisoned by toxic products of the infective process. The poisoned bioplasm falls an easy prey to the hydro- lysing and oxidizing agents always present in the tissues. It breaks down more rapidly and builds itself up more slowly than normal bioplasm. This increased, and to some extent perverted, metab- olism, far from being occasioned by the febrile temperature, is quite probably the cause of the thermo-regulative upset which we call fever. For Mandel has shown — (i) that one of the purin bases (xanthin) causes fever in monkeys; (2) that the purin bases in the urine are increased both in infective fevers and the so-called aseptic or surgical fever — that is, in cases where the temperature rises after such injuries as extensive crushing of tissues without infection. There is a con- stant relation between the height of the fever and the quantity of purin bases excreted. The source of the purin bases in aseptic fever is presumably the autolysis of the injured tissue, from which they pass into the blood without being oxidized to uric acid. The xanthin fever can be prevented by salicylates, though not by antipyrin. It has been very generally admitted that the chief seat of excessive metabolism in fever is the muscles; but U. Mosso has stated that cocaine fever — the marked rise of temperature produced by injec- tion of cocaine — can be obtained in animals paralyzed by curara. This, even if true, would not support the conclusion that a ' nervous fever ' — that is to say, a fever due solely to increased metabolism in the nervous system — exists; for in a curarized animal a large 7o8 ANIMAL HEAT amount of ' active ' tissue (glands, heart, smooth muscle) still remains in physiological connection with the brain and cord. But, as a matter of fact, in an animal under a dose of curara sufficient to completely paralyze the skeletal muscle cocaine causes no appre- ciable rise of rectal temperature ; and this is strongly in favour of the view that the fever produced in the non-curarized animal is connected with excessive muscular metabolism. Significance of the Increased Temperature in Fever. — It remains to ask whether the rise of temperature is anything more than a superficial and, so to speak, an accidental circumstance. The question has already been raised in discussing the changes in the circulation in fever (p. 706). The orthodox view for many ages has undoubtedly been that the increase of temperature is in itself a serious part of the pathological process, a symptom to be fought with and, if possible, removed. And, indeed, it is not denied by anyone that the excessive rise of temperature seen in some cases of febrile disease (to 43° C., or even to 45°) is, apart from all other changes, a most imminent danger to life, unless, as is sometimes the case (in influenza, e.g., where a temperature of 44° has been observed), the high temperature lasts only a short time. Experimental heat paralysis, a condition in which all voluntary and reflex movements are abolished, is produced in frogs by raising the internal tempera- ture to about 34° C. On cooling, the animal recovers. A similar condition can be induced in mammals, but, of course, at a higher temperature. The central nervous system succumbs before the peripheral structures. The superior cervical ganglion in the cat 01 rabbit loses the power of transmitting nerve impulses at 50° C. But some evidence has been brought forward, mostly from the field of bacteriology, to support the idea that in infective processes the rise of temperature is of the nature of a protective mechanism, that the fever is, indeed, a consuming fire, but a fire -that wastes the body, to destroy the bacteria. The streptococcus of erysipelas, for ex- ample, does not develop at 39° to 40° C., and is killed at 39*5° to 41° C., and erysipelas infections in rabbits are less virulent if the body -temperature be artificially raised. Anthrax bacilli, kept at 42° to 43° C. for some time, are attenuated, and when injected into animals confer immunity to the disease. Heated for several days to 41° to 42° C., pneumococci render rabbits immune to pneumonia, and in rabbits in which ' puncture ' fever has been induced pneumo- coccus infections run a milder course. These bacteriological results are supported to a certain extent by clinical experience. For it has been observed that a cholera patient with distinct fever has a better chance of recovery than a case which shows no fever. But too much weight ought not to be given to isolated facts of this sort, and adverse evidence can be produced both from the laboratory TEMPERATURE TOPOGRAPHY 709 and the hospital. For although hens are immune to anthrax under ordinary conditions, but can be infected by inoculation when artificially cooled, frogs, equally immune at the temperature of the air, become susceptible when artificially heated. And it is impos- sible to deny that the use of cold baths in typhoid fever is sometimes of remarkable benefit. This benefit, however, while very unlikely to be connected with any directly unfavourable action of the reduced body-temperature on the growth of the bacilli, may perhaps be due, in some part at least, to an increase in the cutaneous vaso-con- striction which helps to send through the infected intestine a more copious stream of blood. SECTION IV. — DISTRIBUTION OF HEAT — -TEMPERATURE TOPOGRAPHY. The great foci of heat-formation — the muscles and glands — would, if heat were not constantly leaving them, in a short time become much warmer than the rest of the body ; while structures like the bones, skin, and adipose tissue, in which chemical change and heat- production are slow, would soon cool down to a temperature not much exceeding that of the air. The circulation of the blood insures that heat produced in any organ shall be carried away and speedily distributed over the whole body; while direct conduction also plays a considerable part in maintaining an approximately uniform temperature. The uniformity, however, is only approxi- mate. The temperature of the liver is several degrees higher than that of the skin, and the temperature of the brain several degrees higher than that of the cornea. The blood of the superficial veins is colder than that of the corresponding arteries. The crural vein, for example, carries colder blood than the crural artery, and the external jugular than the carotid. The heat produced in the deeper parts of the regions which they drain is more than counter- balanced by the heat lost in the more superficial parts. When loss of heat from the surface is sufficiently diminished by an artificial covering, or prevented by the protected situation of any organ with an active metabolism, the venous blood leaving it is warmer than the arterial blood coming to it. The temperature of the blood passing from the levator labii superioris muscle of the horse during mastication may be sensibly higher than that of the blood which feeds it ; the blood in the vena profunda femoris, and in the crural vein of a dog with the leg wrapped in cotton-wool, is warmer byo-i°to 0-3° than of the crural artery. The difference is due to the heat produced in the muscles, and it ought to be of this order of magnitude. The quantity of blood in a y-kilo dog is about £ kilo ; J of this, or ^ kilo, is in the skeletal muscles, and the average circulation-time through them may be taken as ten seconds. Six times in the minute, or 360 times in the hour, ^ kilo of blood passes through the muscles, and is heated on the average by 0-2°. "^60 "2 This represents a heat-production of about ^- x — , or 9 calories per hour. Now, the total heat-production of a y-kilo dog is about 7io ANIMAL HEAT 19 calories per hour, of which somewhat less than one-half is formed in the skeletal muscles. The blood of the inferior vena cava at the level of the kidneys may be 0-1° colder than that of the abdominal aorta, but is warmer than the blood of the superior cava. The right heart, therefore, receives two streams of blood at different temperatures, which mingle in its cavities. A controversy was long carried on as to the relative tem- perature of the blood of the two sides of the heart; but the researches of Heidenhain and Korner have shown that a thermometer passed into the right ventricle through the jugular vein stands, as a rule, slightly higher than a thermometer introduced through the carotid into the left ventricle. The method gives not so much the temperature of the blood in the two cavities as that of their walls. The thin-walled right ventricle is heated by conduction from the warm liver, from which it is only separated by the diaphragm, while the left ventricle loses heat to the cooler lungs. The difference of temperature is not caused by cooling of the blood in its passage through the pulmonary capillaries, for even when respiration is suspended, a difference of temperature between the two sides of the heart is found. Under ordinary circum- stances, the inspired air is already heated almost to body-temperature before it reaches the alveoli. But, while this is the case, a fall of less than j\j° in the temperature of the blood passing through the lungs would account for all the heat lost by the expired air. If half of the loss took place in the upper air-passages, less than ^5° would be suffi- cient. A slight difference of temperature in the blood of the two ven- tricles might be caused, even in the absence of respiration, by the heat developed in the cardiac 'muscle itself during contraction, a large pro- portion of which must be conveyed by the coronary veins into the right heart. The surface temperature varies between rather wide limits with the temperature of the environment. The temperature of cavities like the rectum, vagina, and mouth, and of secretions like the urine, approxi- mates to that of the blood in the great vessels or the heart, and under- goes only slight changes. An increase in the velocity of the blood causes the internal and surface temperatures to come nearer to each other, the former falling and the latter rising. When the loss of heat from a portion of the surface is prevented, the temperature of this portion approaches the internal temperature. For this reason a thermometer placed in the axilla approximately measures the internal temperature, and not that of the skin; and a thermometer in the groin of a rabbit, and completely covered by the flexed thigh, may stand as high as, or, it is said, even higher than, a thermometer in the rectum (Hale White). The temperature in the mouth is not a very reliable index of the deep temperature of the body, especially in cold weather or after exercise, as it is apt to be influenced by the inspired air. The mouth must, of course, be kept closed during the measurement. On the average its temperature is about the same as that of the axilla, and 0-4° C. below that of the rectum. The rectal temperature is 0-2° or 0-3° above that of the urine. In point of accuracy rectal observations are the best, and next to them determinations of the temperature of the stream of urine. The latter method, although subject to obvious limitations, is rapid and free from the danger of conveying infection to the person (Pembrey). The surface temperature is a rough index of the rate of heat-loss; the internal temperature, of the rate of heat-production. A normal skin temperature and a rising rectal temperature would probably in- dicate increased production of heat; an increased rectal temperature, TEMPERATURE TOPOGRAPHY 711 in conjunction with a diminished surface temperature, as in the cold stage of ague, might be due either to diminished heat-loss while the heat-production, remained normal, or to diminished heat-loss plus increased heat-production. The following tables illustrate the differences of temperature found in the body. It should be remembered that the numbers are not strictly comparable with each other; the temperature of the mammals in which direct observations have been made on the blood is not exactly the same as that of man, the temperature of the dog, for example, being a little (about i° C.) higher. Then in the same animal there is no very constant ratio between the temperature of the blood in two vessels or of the skin at two points. Even in the same vessel the temperature may vary with many circumstances, such as the velocity of the stream, and the state of activity of the organ from which it comes. Apart from physiological variations, experimental fallacies sometimes cause a want of constancy, especially in measurements of blood temperature. The insertion of a mercurial thermometer into a vessel is very likely to obstruct the passage of the blood; and if the blood lingers in a warm organ, it will be heated beyond the normal. In man the blood-temperature in the arteries at the wrist has been estimated indirectly by the calorimetric method of measuring the blood-flow in the hand (p. 122), probably with greater accuracy than would be attainable by the direct insertion of a thermometer, were this permissible. The temperature of the calorimeter is determined at which it neither imparts heat to the blood nor gains heat from the blood. On the assumption that the heat-production of the resting hand is negligible for this purpose,* the temperature so fixed will be that at which the blood enters the hand — i.e., the temperature of the arterial blood at the wrist. Blood. (Dog.) Right heart - - - - 38-8° C. Left ,, - - 38-6 Aorta - - 38-7 Superior vena cava - - 36-8 Inferior ,, ... 38-1 Crural vein- - ... 37-2! ,, artery - 38-0 Profunda femoris vein - 38-2 Portal vein - 38-39 ) Varies with activity Hepatic vein - - 38-4-39- 7 / of digestive organs. Arterial blood at wrist in man - 0-5 below rectal temperature. * Since, of course, some heat is produced in the hand even at rest, although doubtless less per unit of weight than in the resting body as a whole, the arterial blood temperature as thus determined must be somewhat too high. No error is caused by this in the calculation of the blood-flow in the hand (p. 122) ; for while the factor T — T' in the denominator is somewhat too great, thtt corresponding quantity of heat produced in the hand is included in H in the numerator. f The following numbers were obtained (in an anaesthetized dog whose rectal temperature had fallen 2° C.) for the temperature of the walls of the crural artery and vein, as measured by an electrical resistance thermometer. Leg of dog lightly wrapped in wool. ~\ Crural artery 34'95 ,, vein ------ 34-76 1 Rectum, 36*2 Leg more carefully wrapped up. [Air, i6'3 Crural artery ----- 34-70 7" ANIMAL HEAT Tissues. Brain - - - - - - 40° C Liver. - - 40-6-40*9 Subcutaneous tissue 2-1 lower than, that of subjacent muscles (man). Anterior chamber of eye - - 31-9) / uu--n Vitreous humour - - 36-i}(rabblt)- Cavities. (Man.) Axilla - 36-3-37-5° C. (97'3-99-5° F-)- Rectum - 36-37-8 Mouth - - 37-25 Vagina - - 37-5-38 Uterus - - 37-7-38-3 External auditory meal us - 37'3-37'8 Bladder (temperature of the escaping urine) - - 36-0-37-5 Respiratory Passages. (Horse.) Air fMiddle of nasal cavity - - 23-4° C. temperature, j ,, trachea - - 32-4 in inspiration. 4-5° C. I ,t ». - 34'4 in expiration. Natural Surfaces* Cheek (boy, immediately after running) - 36-25° f Anterior surface of forearm - 33'5-34'4 (Man) Posterior ,, ,, - 34-0 Room J Skin over biceps - 35-0 temperature ,'\ ,, ,, head of tibia - - 31-9 17-5° ,, immediately below xiphoid cartilage - 34-7 ^ ,, over sternum - - 33^2 On hair (boy) - - 30-0 Under hair over sagittal suture (boy) - 33'7-34'O Shaved skin of neck (rabbit) - 36-5 On hair ,, ,, - 31-5 ,, between eyes ,, - - - 30-7 Artificial Surfaces. (Surface of trousers over thigh - 23-7-28-7° temperature, coat over arm j_._o ,, waistcoat ----- 26-0 Normal Variations in the Temperature. — The internal tempera- ture, as has been already said, is not strictly constant. It varies with the time of day; with the taking of food; with age; to some extent with violent changes in the external temperature, such as those produced by hot or cold baths; and possibly with sex. On the average the range of variation in the temperature of the rectum or urine of a healthy man is from 36-0° C. (96-8° F.) to 37*8° C. (100-0° F.). In the monkey a very distinct and constant diurnal variation has been observed, and the range is much wider than in man (as much as 5*4° F.), the maximum falling between 6 and 8 p.m. and the minimum between 2 and 4 a.m. (Simpson). TEMPERATURE TOPOGRAPHY 713 The daily curve of temperature shows a minimum in the early morning, between two and six o'clock (36-3° C.), and a maximum in the evening, between five and eight o'clock (37*5° C.) ,(Fig. 227), The daily range in health may be taken as a little over i° C., 01 about 2° F. In fever it is generally greater, but the maximum and minimum fall at the same periods; and it is of scientific, and also of practical, interest that the early morning, when the temperature and pulse-rate are at their minimum, is often the time at which the flagging powers of the sick give way. From two to six o'clock in the morning the daily tide of life may be said to reach low-water mark. Even in a fasting man the diurnal temperature curve runs its course, but the variations are not so great. The ta- king of food of itself causes an increase of temperature, although in a healthy man this rarely amounts to more than half a degree. The rise of temperature is certainly due in part to the increased work of the alimentary canal, but it is in the main connected with the increase of metabolic activity which the entrance of the products of digestion into the blood brings about. The heat-production is especially increased by proteins. A dog weighing 15-3 kilos, the heat-production of which was 22-3 calo- ries during an hour previous to feeding, was given 1,200 grammes of meat at noon. The heat-production rose to 36 calories in the 2nd hour, and 42 calories in the 3rd. It remained above 40 calories per hour beyond the loth hour, and in the i4th hour it had only fallen to 37 calo- ries, to reach 25 calories in the 2ist hour. On the whole, the increase in heat-production ran parallel with, and was proportional to, the increase in the excretion of nitrogen (Williams, Richie and Lusk). The relatively unimportant share taken by the increased work of the gastro-intestinal tract in the augmentation of the metabolism is illus- trated by the fact that a high rate of heat-production was maintained till the 1 4th hour, even although by this time three-quarters of the nitrogen corresponding to the food protein had been eliminated in the urine, and the work of digestion and absorption must have been largely completed. The rise of temperature during digestion is gradual, the maximum being reached during the fourth hour, or even later. The cause of the daily variation of temperature has been much discussed. There is no doubt that several factors are concerned, among the most important being the variation in the amount of contraction of the skeletal muscles and the influence of food. Mus- Fig. zij. — Curve showing the Daily Variation of Body-Temperature. 425 MS cular exercise is capable of causing a considerable rise in the tem- perature of the rectum and urine, to 38*5° C. (101-3° F.) or even 38*9° C. (102° F.) without producing any feeling of distress. Other unknown influences seem also to be involved, as is shown by the fact that in persons who work at night and sleep during the day the curve of temperature, although greatly altered, is not reversed. Recent observations on this subject are those of Benedict. By means of a resistance thermometer in the rectum, readings were taken usually every four minutes. With such a thermometer no disturb- ance of the person's sleep is necessary to obtain a reading. He can sit with- out discomfort in any position, walk about the room (returning to the ob- server's table for the observations), and even ride a stationary bicycle. Even years of night- work do not elim- inate the tendency to a fall of tempera- ture at night, a minimum in the early morning, and a morning rise. As to the relation of age and sex to temperature, it is only necessary to remark that the mean temperature both of the young child and of the old man is somewhat higher than that of the vigorous adult; but a point of more importance is the rela- tive imperfection of the heat-regulation in infancy and age, and the greater effect of accidental circumstances on the mean temperature. Thus, old people and young children are specially liable to chills, and a fit of crying may be sufficient to send up the temperature of a baby. In infants an hour or two old the temperature may be as low as 34° C. (93*2° F.) or 33*0° C. (91-4° F.) even when they are fully clothed in a room at 15° C. (59° F.). It rises gradually during the first day or two, but shows marked variations. On the fifth day after birth, e.g., the rectal temperature ranged from 36*2° C. (97-16° F.) to 33-5° C. (92-3° F.) in a child weighing 5^ pounds (Babak). The temperature of women is generally a little higher than that of men, and is also somewhat more variable. A fall of temperature, rarely amounting to more than i° F., is associated with the menstrual period. After death the body cools at first rapidly, then more slowly (Fig. 2:8). But occasionally a post-mortem rise of temperature may take place. In certain acute diseases (like tetanus) associated with excessive muscular contraction this has been especially noticed; in bodies wasted by prolonged illness it does not occur. Nearly an 28 Fig. 228. — Curve of Cooling after Death : Guinea - Pig. Time marked along horizontal, and temperature along vertical axis. At a ether and chloroform given to kill animal; death, as indicated by stoppage of the heart, took place at b. The dotted line shows the course the curve would have taken if death had occurred at the moment the anaesthetics were given. Air of room i7-6°. PRACTICAL EXERCISES 715 hour after death, in a case of tetanus, the temperature was found to be 45-3°, while before death it was 44-7° (Wunderlich). In dogs a slight post-mortem rise may be demonstrated, especially when the body is wrapped up; but when an animal has been long under the influence of anaesthetics no indication whatever of the phenom- enon may be obtained. The explanation of post-mortem rise of temperature is to be found : (i) In the continued metabolism of the tissues for some time after the heart has ceased to beat, for the cell dies harder than the body. (2) In the diminished loss of heat, due to the stoppage of the circulation. (3) To a small extent in physical changes (rigor mortis, coagulation of blood) in which heat is set free. PRACTICAL EXERCISES ON CHAPTERS X., XI., AND XII. i. Glycogen* — (i) Preparation. — (a) Cut an oyster into two or three pieces, throw it into boiling water, and boil for a minute or two. Rub up in a mortar with clean sand, and again boil. Filter. Precipitate any proteins which have not been coagulated, by adding alternately a drop or two of hydrochloric acid and a few drops of potassio-mercuric iodide so long as a precipitate is produced. Only a small quantity of these reagents will be required, as the 'greater part of the proteins has been already coagulated by boiling. Filter if any precipitate has formed. The nitrate is opalescent . Precipitate the glyco gen from the nitrate (after concentration on the water-bath if it exceeds a few c.c. in bulk) by the addition of four or five times its volume of alcohol. Filter off the precipi- tate, wash it on the filter with alcohol, and dissolve it in a little water. To some of the solution add a drop or two of iodine ; a reddish-brown (port wine) colour is produced, which disappears on heating, returns on cooling, is removed by an alkali, restored by an acid. Add saliva to some of the glycogen solution, and put in a bath at 40° C. In a few minutes reducing sugar (maltose) will be found in it by Trommer's test (p. 10). Note that dextrin (erythrodextrin) gives the same colour with iodine as glycogen does. Dextrin is also precipitated by alcohol, but a greater proportion must be added to cause complete precipitation. Glycogen is completely precipitated by saturation with magnesium sulphate or ammonium sulphate, so that the filtrate no longer gives the reddish colour with iodine. A pure solution of erythrodextrin is not precipitated. On the addition of a drop or two of a solution of basic lead acetate to a solution of glycogen in distilled water, a pre- cipitate forms immediately. When the same reagent is added to a solution of dextrin in distilled water there is no immediate precipitate. Maltose is formed when dextrin is digested with saliva. (&) Cut another oyster into pieces, throw it into boiling water acidu- lated with dilute acetic acid, and boil for a few minutes. Rub up in a mortar with sand, boil again, and filter. Test a portion of the ni- trate with iodine for glycogen. Precipitate the rest with alcohol, filter, dissolve the precipitate in water, and test again for glycogen. On boiling some of the opalescent solution for a few minutes after the addition of a few drops of sulphuric acid the opalescence disappears, and * For the quantitative estimation of glycogen in organs, Pfluger's method is the best. The organ is minced and heated with strong (60 per cent.) potas- sium hydroxide. The glycogen is precipitated with alcohol, and then, after hydrolysis with hydrochloric acid, estimated as dextrose. 716 METABOLISM AND ANIMAL HEAT when the solution has been neutralized with sodium hydroxide it gives Trommer's test, owing to the hydrolysis of the glycogen into dextrose. (2) Deeply etherize a dog or rabbit five hours after a meal rich in carbo-hydrates — e.g., rice and potatoes in the case of the dog, carrots in the case of the rabbit. Fasten it on a holder. Clip off the hair over the abdomen in the middle line. Make a mesial incision through the skin and abdominal wall from the ensiform cartilage to the pubis. The liver will now be rapidly cut out (by the demonstrator) and divided into two portions, one of which will be (distributed among the class and) treated as in (a) or (b) ; the other will be kept for an hour at a temperature of 40° C., and then subjected to process (a) or (b). Little, if any, sugar and much glycogen will be found in the portion which was boiled immediately after excision. Abundance of sugar will be found in the portion kept at 40° C. ; it may or may not contain glycogen. 2. Catheterism. — In many physiological experiments it is necessary to obtain urine from the bladder by means of a catheter. It is possible to pass a fine rubber catheter into the bladder of a male dog. A larger one is easily passed in a male rabbit, and a still larger in a bitch, which is often used for experiments on metabolism. Even in the bitch the opening of the urethra lies entirely concealed within the vagina, much deeper than the cul-de-sac in the mucous membrane, into which the beginner usually tries to force the catheter. For a first attempt the animal should be etherized and fastened on a holder. The little or index finger of the left hand is passed into the vagina till the sym- physis pubis can be felt. A little below this is the opening of the urethra. With the right hand the point of a catheter of suitable calibre is directed along the finger, and after a little ' guess and trial ' it slips into the bladder, its entrance being announced by the escape of urine. A glass tube drawn out to a sufficiently small calibre and bent near the point is the easiest form of catheter to pass in a bitch. The point must, of course, be rounded in the flame. The insertion of the catheter is much facilitated by the use of a speculum. When the bitch is to be used in a long series of experiments an operation is sometimes performed first of all to render the urethral orifice more accessible. 3. Glycosuria. — (i) (a) Weigh a dog (female by preference) or rabbit. Fasten on a holder, and etherize. Insert a glass cannula into the femoral or saphena vein of the dog, or into the jugular of the rabbit (p. 214). Fill a burette with a 2 per cent, solution of dextrose in physiological salt solution, connect it with the cannula by means of an indiarubber tube, taking care that there are no air-bubbles in the tube, and slowly inject as much of the solution as will amount to £ or f- grm. of sugar per kilo of body- weight. Tie the vein, remove the cannula, and in half an hour evacuate the bladder by passing a catheter, by pressure on the abdomen, or, if both of these methods fail, by tapping the bladder with a trocar pushed through the linea alba (supra pubic puncture). In an hour again draw off the urine. Test both specimens for sugar. In this experiment the opportunity may also be taken to demon- strate that egg-albumin, when injected into the blood, is excreted by the kidneys, a filtered solution containing the albumin of one egg and sugar in the quantity mentioned being injected. The catheter may be inserted before the injection is begun, and the bladder evacuated. After the injection the urine that drops from the catheter may be collected in test-tubes, first every two minutes, and then, as soon as sugar is found, every ten minutes. Determine the interval between injection and the appearance of the first trace of sugar and albumin. If a sufficient amount of urine is obtained, the PRACTICAL EXERCISES 7*7 quantity of sugar in successive specimens may be estimated and com- pared. The rate of flow of the urine as measured by the number of drops falling from the catheter may also be estimated from time to time in order to determine whether diuresis is taking place. If a rabbit is used for this experiment, the sugar solution may be injected into the ear vein. The vein is caused to swell up by pressing on it with the finger and thumb, and the hypodermic needle is then inserted towards the heart. (b) Instead of collecting the urine by a catheter in the bladder, the abdomen of the dog may be opened, and a cannula tied into each ureter. The two cannulse are then connected by short rubber tubes with a glass Y-piece, on the stem of which a test-tube is tied for collecting the urine. Replace the test-tube by a fresh one from time to time. The urine already in the bladder is removed by pressure or by a trocar, and tested for sugar, since the anaesthetic itself may cause a certain amount of glycosuria. Test the samples of urine obtained from the ureters for sugar, and in those in which it is present estimate its amount. Note also any changes in the rate of secretion of urine, and any abnormal constituents, as albumin. (2) Phlorhizin Glycosuria. — Dissolve J grm. of phlorhizin in warm water, and inject it subcutaneously into a rabbit. Obtain a sample of the urine at the end of two hours, by pressure on the abdomen with the thumb or by passing a catheter, and test for sugar. If none is present, wait some time longer, and again test the urine. This experiment can also be performed without risk on man. One grm. of phlorhizin has been injected twice a day without disturbing the individual. Much sugar is found in the urine, but it disappears the day after the administration of phlorhizin is stopped. The phlorhizin may also be given by the mouth, but more is required, and it is not very easily absorbed, and often causes diarrhoea (v. Mering). (3) Alimentary Glycosuria. — The urine having been tested for sugar for two successive days, and none being found — (a) A large quantity of dextrose is to be taken in the form which is most agreeable to the student some hours after a meal. The urine of the next twenty -four hours is to be collected and measured. A sample of it is then to be tested for reducing sugar by Trommer's and the phenyl-hydrazine test. If any sugar is found, the reducing power of a definite quantity of the urine is to be determined by titration with Fehling's solution (p. 526). (b) Instead of dextrose use cane-sugar and proceed as in (a). But estimate the reducing power of the urine (a) before and (/3) after boiling with hydrochloric acid (p. 465). (c) A large meal of rice and arrowroot, sweetened with as much dex- trose as the observer can induce himself to swallow, is to be taken, and the urine treated as in (a). (d) A large number of sweet oranges may be eaten.* 4. Estimation of the Sugar in Blood — Method of Lewis and Benedict, somewhat modified. — The blood must be used immediately after being drawn. When only a single sample is required it can be obtained by puncturing a vein with a large-sized hypodermic needle. If a number of students have to be provided with blood an animal may be killed by decapitation. Long anesthetization is to be avoided, as this causes hyperglycaemia. To take blood from a' vein, place a few crystals of potassium oxalate in the tip of a 2 c.c. pipette, and having punctured the vein with a needle, draw up blood into the pipette to a little above the * These experiments may be distributed among the class so that each student does one. 718 METABOLISM AND ANIMAL HEAT mark. Allow the blood to run back just to the mark, and immediately discharge the contents of the pipette into a large test tube (about 50 c.c. capacity) containing 8 c.c. of distilled water. Shake until haemolysis is complete. Then add 15 c.c. of saturated aqueous solution of pure picric acid ; shake thoroughly to precipitate the proteins of the blood, and filter through a small filter paper. Into each of two long narrow test tubes (capacity about 25 c.c.), which have been marked accurately with a file at 10 c.c., measure 7 c.c. of the filtrate, add 2 c.c. of the saturated solution of picric acid and i c.c. of a 10 per cent, solution of anhydrous sodium carbonate. A duplicate determination wrill thus be made. The test tubes are placed in an autoclave (according to Pearce's modification of the method), and the pressure gradually brought up to 20 to 25 pounds to the square inch (2-5 kilogrammes to the square centi- metre), and kept at this level for twenty-five minutes. Picramic acid is formed by the reducing action of the sugar, and from the amount of picramic acid estimated by a colorimeter the amount of sugar is de- duced. Let the autoclave cool down till the pressure is zero, remove the tubes and allow them to cool to room temperature. Sufficient water is now added to each tube to bring the volume back to the 10 c.c. mark, and after shaking the contents are filtered through cotton into the (Duboscq) colorimeter bottles. The determination is made by comparison with a solution containing a known amount of pure dex- trose which has been carried through the same process, or with a picramic acid standard solution.* A solution of pure dextrose may be used as a standard instead of the picramic acid, as follows : Into a test tube place 4 c.c. of a solution, corresponding to 0-56 milligramme dextrose; add 5 c.c. of the saturated picric acid solution, and i c.c. of 10 per cent, solution of anhydrous sodium carbonate. Place in the autoclave and proceed as above described, bringing the final volume again to 10 c.c., and filtering through cotton into one of the colorimeter bottles. Calculation : Reading of standard mgm. dextrose in standard^ mgm. dextrose per Reading of unknown c.c. blood used J~ c.c. blood. The amount of blood taken for each of the duplicate determinations is •£§ x 2 c.c. =0-56 c.c. The standard=o-56 milligramme dextrose in 10 "c.c., so that the second fraction on the left side of the equation becomes unity, and the reading of the standard divided by the reading of the unknown gives at once the number of milligrammes of dextrose in i c.c. of blood. 5. Milk. — (i) Examine a drop of fresh cow's milk with the micro- scope. Note the fat globules of various sizes. (2) Determine the specific gravity of the milk with a hydrometer (lactometer). Then centrifugalize some of the milk to separate the cream, which rises to the top of the tubes. Remove the cream and determine the specific gravity of the skimmed milk. It will be found to have increased, since the fat is of lower specific gravity than the rest of the milk. Normal cow's milk has a specific gravity of 1,028 to 1,034, skimmed milk 1.033 to 1,037. * Picramic acid, 56 milligrammes, anhydrous sodium carbonate, 100 milligrammes, and distilled water to make up 1,000 c.c. Dissolve the sodium carbonate in about 50 c.c. of water, add the picramic acid and dissolve it with the aid of heat. When cooled to room temperature add enough water to make 1,000 c.c. Picramic acid obtained on the market varies in the amount of colour produced in the preparation of this solution. The standard should therefore be compared with a solution of pure dextrose which has been standard- ized by another method, and then treated by the Lewis and Benedict method. PRACTICAL EXERCISES 719 (3 ) Test the reaction of the milk to litmus-paper . It is slightly alkaline. (4) (a) Put 10 c.c. of milk in a test-tube, and nearly fill it up with water. Add strong acetic acid drop by drop. A precipitate of casein- ogen is thrown down which entangles the fat, and carries it down mechanically along with it. Filter off the precipitate. Keep the filtrate for (b). Wash the precipitate with water, scrape a portion of it off the filter, and add to it some 2 per cent, sodium carbonate solution. The caseinogen dissolves, while the fat remains in suspension. The solution gives the colour reactions for proteins (p. 8). (6) Test some of the filtrate (p. 10) for lactose. Add dilute sodium carbonate solution to another portion till it is only slightly acid. Boil, and lactalbumin is coagulated. Remove the lactalbumin by filtering, and test this filtrate for earthy (i.e., calcium and magnesium) phos- phates by adding a few drops of ammonia, which precipitates them as a slight cloud. (c) To 5 c.c. of milk add an equal volume of saturated ammonium sulphate solution. The caseinogen is precipitated, entangling the fat. Filter off. The filtrate may be used to test for lactalbumin by boiling. The addition of water to the precipitate of caseinogen (and fat) on the filter causes the caseinogen to dissolve, as it is soluble in weak salt solutions. Caseinogen can also be precipitated by saturating milk with sodium chloride or magnesium sulphate. (5) To 5 c-c- °* milk a(ld. a couple of drops of 20 per cent, sodium or potassium hydroxide, and then a few c.c. of ether. Shake up. The ether dissolves the fat, and the opacity of the milk diminishes. Take off the ether with a pipette, evaporate away most of it on a water-bath, and place a drop or two of the remainder on a filter-paper. A greasy stain is left, showing the presence of the fat of the milk, or butter. (6) Clotting of Milk. — (a) To a few c.c. of milk in a test-tube add a few drops of rennet. Place the tube in a bath at 40° C. In a short time a clot or curd is formed, consisting of casein, which is derived from the caseinogen. The fat is entangled in the clot. On standing some time the clot contracts, and exudes the whey. Boil some of the whey after slight acidulation with acetic acid; the lactalbumin and whey- protein are coagulated. Test another portion of whey for proteins by one of the general protein tests (p. 8) — e.g., the xanthoproteic. (b) Repeat (a) but use rennet which has been previously boiled. The milk is not curdled, because the ferment has been inactivated by boiling. (c) To 10 c.c. of milk add 3 c.c. of i per cent, potassium oxalate. Divide the oxalated milk into three portions — A, B, and C. To A add a few drops of rennet, to B i c.c. of 2 per cent, calcium chloride solu- tion and a little rennet, and to C i c.c. of 2 per cent, calcium chloride solution alone. Put the tubes at 40° C. Clotting will occur in B, but not in A or C. 6. Cheese. — (i) Rub up some finely-grated cheese in a mortar with 2 per cent, sodium carbonate solution. Filter. The filtrate contains casein, which can be precipitated by adding dilute acetic acid by drops to a portion of the filtrate. The precipitate is soluble in excess of the acid. With another portion of the filtrate perform some of the general protein tests (p. 8). (2) Shake up some finely-grated cheese in a dry test-tube with ether. Take off the ether with a pipette, and evaporate on a water-bath till only a few drops remain. With a glass rod put a drop of the ether on a piece of filter-paper. A greasy spot is left, showing that fat is present. 7. Flour. — (i) Mix some wheat-flour v/ith a little water into a stiff dough. Let it stand for a few minutes at body-temperature to facilitate the formation of gluten. Wrap a piece in cheese-cloth, forming a kind 720 METABOLISM AND ANIMAL HEAT of bag, and knead it with the fingers in a capsule of water. The starch grains come through the cheese-cloth. Pour the water into a beaker. It is opaque, and on standing the starch grains sink to the bottom, (a) Test for starch with the iodine test, and also examine microscopically. The grains are round, with a central hilum, and are smaller than those of potato starch (p. n). (b) Test for sugar by Trommer's test (p. 10). None is present unless the flour has been made from inferior grain in which some germination has taken place. (2) Go on kneading the dough till no more starch comes through. The sticky mass which remains in the bag is a protein called gluten, which is formed from certain globulins and other proteins in the flour on addition of water. Oatmeal, ground rice, and other grains poor in gluten-forming globulins do not form dough when mixed with water. Suspend some of the gluten in water in a test-tube, and apply to it the general protein colour tests (p. 8). 8. Bread. — (i) Rub up a small piece of the crumb of a stale loaf in a mortar with water. Strain through cheese-cloth. The fluid which passes through contains starch grains, (a) Filter it, and test a portion of the filtrate for dextrose by Trommer's test. A positive result is obtained. Test another portion with iodine for erythrodextrin. (b) Test a portion of the residue of the bread which has not passed through the cheese-cloth for protein by the general protein tests — e.g., the xanthroproteic or Millon's tests. (2) Repeat (i) using the crust of the bread. Both dextrose and erythrodextrin are present in the cold-water extract, but the dextrose is less plentiful than in the crumb, having been converted into caramel in the baking. The sugar and dextrin are formed from the starch of the flour by the ferments of the yeast employed to make the bread rise. 9. Variations in the Total Nitrogen (p. 521) and in the Quantity of Urea excreted, with Variations in the Amount of Proteins in the Food. — The student should put himself, or somebody else if he can, for two days on a diet poor in proteins, then (after an interval of forty-eight hours on his ordinary food) for two days on a diet rich in proteins. A suitable table of diets will be supplied. The urine should be collected on ths six days of the period of experiment, on the day before it begins, and on the day after it ends. Small samples of the mixed urine of the twenty-four hours for each of these eight days should be brought to the laboratory, and the quantity of urea determined by the hypobromite method. The volume of the urine passed in each interval of twenty- four hours being known, the total excretion of urea for the twenty-four hours can be calculated, and a curve plotted to show how it varies during the period of experiment.* If sufficient time is available, the experiment will be made still more instructive by determining the total nitrogen in each sample in addition to the urea. A curve showing the variation in the total nitrogen can then be plotted on the same paper as the urea curve, and a table calculated giving the percentage of the total nitrogen contained in the urea for each day of the experiment. 10. Action of Epinephrin (Adrenalin). — Several experiments to illus- trate this are given in the Practical Exercises following other chapters, but may equally well be performed here. (See Experiment 8, p. 66; Experiment 3, p. .453-.) * In 17 healthy students the average amount of urea excreted in twenty four hours on the ordinary diet was 29-51 grammes (minimum 19-35 grammes maximum 46-01 grammes) ; on a diet poor in protein, average 20-75 grammes (minimum 9-52 grammes, maximum 32-86 grammes) ; on a diet rich in protein, average 38*83 grammes (minimum 23-26 grammes, maximum 67-82 grammes). PRACTICAL EXERCISES 721 ii.* Measurement of the Quantity of Heat given off in Respiration. — This may be done approximately as follows: Put in the inner copper vessel, A, of the calorimeter shown in Fig. 224 (p. 682) a measured quantity of water sufficient to completely cover the series of brass discs. Place A in the wide outer cylinder, the bottom of which it is prevented from touching by pieces of cork. The outer cylinder hinders loss of heat to the air. Suspend a thermometer in the water through one of the holes in the lid. In the other hole place a glass rod to serve as a stirrer, Read off the temperature of the water. Put the glass tube connected with the apparatus in the mouth, and breathe out through it as regu- larly and normally as possible, closing the opening of the tube with the tongue after each expiration and breathing in through the nose. Continue this for five or ten minutes, taking care to stir the water fre- quently. Then read off the temperature again. If W be the quantity of water in c.c., and t the observed rise of temperature in degrees Centi- grade, Wt equals the quantity of heat, expressed in small calories (p. 675), given off by the respiratory tract in the time of the experiment, on the assumptions (i) that all the heat has been absorbed by the water, (2) that none of it has been lost by radiation and conduction from the calorimeter to the surrounding air. Calculate the loss in twenty -four hours on this basis ; then repeat the experiment, breathing as rapidly and deeply as possible, so as to increase the amount of ventilation. The quantity of heat given off will be found to be increased, f In an experiment of short duration (2) is approximately fulfilled. As to (i), it must be noted that in the first place the metal of the calorimeter is heated as well as the water, and the water-equivalent of the apparatus must be added to the weight of the water (p. 676). The water-equivalent is determined by putting a definite weight of water at air temperature T into the calorimeter, and then allowing a quantity of hot water at known temperature T' to run into it, stirring well, and noting the temperature of the water when it has ceased to rise. Call this temperature T". Enough hot water should be added to raise the temperature of the calorimeter about 2° C. The quantity run in is obtained by weighing the calorimeter before and after the hot water has been added. Suppose it is m. Let the mass of the cold water in the calorimeter at first be M, and let M'=the mass of water which would be raised i° C. in temperature by a quantity of heat suffi- cient to increase the temperature of all the metal, etc., of the calorimeter by i° — in other words, the water-equivalent of the calorimeter. The rnass m of hot water has lost heat to the amount of m (T' — T"), and this has gone to raise the temperature of a rnass of water M, and metal equivalent to a mass of water M', by (T" — T) degrees. .-. m (T'— T") = M(T"— T) + M'(T"— T). Everything in this equation except M' is known, and .•'. M', the water-equivalent of the calorimeter, can be deduced, and must be added in all exact experiments to the mass of water contained in it. Secondly, all the excess of heat in the expired over that in the inspired air is not given off to the calorimeter, for the air passes out of it at a slightly higher temperature than that of the atmosphere. At the beginning of the experiment this excess of temperature is zero. If at the end it is i° C., the mean excess 13 0-5° C. Now, when respiration * This experiment is given as an example of a simple calorimetric measure- ment, which can be easily performed with sufficient accuracy by students, and involves the essential principles of such determinations. f The average heat-loss by the lungs for 51 men (calculated for the 24 hours) was 312,000 small calories for normal, 919,000 for the fastest, and 195,000 for the slowest breathing. 722 METABOLISM AND ANIMAL HEAT is carried on in a room at a temperature of 10° C., the expired air has its temperature increased by nearly 30° C. About jfa of the heat given off by the respiratory tract in raising the temperature of the air of respiration would accordingly be lost in such an experiment. But since the portion of the heat lost by the lungs which goes to heat the expired air is only £ of the whole heat lost in respiration (p. 682), the error would only amount to 5^ of the whole, and this is negligible. Thirdly, the air leaves the calorimeter saturated with watery vapour at, say.. 10-5°, while the inspired air is not saturated for 10° C. Now, the quantity of heat rendered latent in the evaporation of water suffi- cient to saturate a given quantity of air at 40° C. (the expired air is saturated for body-temperature) is six times that required to saturate the same quantity of air at 10°. If, then, the inspired air is half saturated, the error under this head is -^ , or 8£ per cent. If the inspired air is three-quarters saturated, the error is ^V or about 4 per cent. If the air is fully saturated before inspiration, as is the case when it is drawn in through a water-valve (Fig. 229) by a tube fixed in one nostril, the only error is that due to the slight excess of temperature of the air leaving the calorimeter over that of the inspired air. The latent heat of the aqueous vapour in saturated air at 10-5° C. is about more than the latent heat of the aqueous vapour in the same mass of saturated air at 10° C., or about T^0 of the latent heat in saturated air at 40°. The error in this case would therefore be under i percent. The tubes must be wide and the bottle large. 12. In the observations on the blood-flow in the hands (Experiment 31, p. 219) data on the quantity of heat given off by the hands when immersed in water at a given temperature have already been obtained. Additional data should be got by putting the hand into the calorimeter without previous immersion in the bath, and comparing the heat given off during the period when the hand is acquiring the temperature of the calorimeter with that (riven off when the steady state has been reached. Different calorimeter tmi- peratuies should be employed. It will be found that as the calorimeter temperature is diminished the quantity of heat given off may be increased although the blood-flow is diminished, each gramme of blood passing through the hand giving off more heat the lower the calorimeter temperature. The quantity of heat lost by the hand, at a given temperature of the calorimeter, per square centimetre of skin surface can be cal- culated. If no special instrument for measuring the area of irregular surfaces is available, the surface of the hand can be arrived at ap- proximately by covering it with strips of gummed paper of known breadth, and noting the length used to cover the whole hand up to the lower level of the styloid process of the ulna. Or an old thin glove which fits the hand can be cut off at this level and weighed. As large a piece as possible of regular shape is then cut from the glove, weighed, and its area deduced by measuring it with a rule. The area of the whole glove, on the assumption that it is of uniform thickness, is thus known. Or, without cutting the glove, it may be laid flat on a piece of paper, an outline of it traced, and the paper cut out. The weight of the paper cut out is compared with that of a piece of paper of known area, and its area deduced. Obviously this is approximately equal to half the surface of the hand. Fig. 229. — Bottle arranged for Water-Valve. CHAPTER XIII THE PHYSIOLOGY OF THE CONTRACTILE TISSUES SECTION I. — PRELIMINARY OBSERVATIONS — PHYSICAL AND TECHNICAL DATA. IN all the great functions of the body muscular movements play an essential part. The circulation and the respiration, the two functions most immediately essential to life, are kept up by the contraction and relaxation of muscles. The movements of the digestive canal, the regulation of the blood-supply to its glands and to all parts of the body, and that immense class of movements which we call voluntary, are all dependent upon muscular action, which, again, is indebted for its initiation, continuance, or control, to impulses passing along the nerves from the nerve-centres. Hitherto we have not gone below the surface fact, that muscular fibres have the power of contracting, either automatically, or in response to suitable stimuli. In this chapter and the two next we shall consider in detail the general properties of muscle, nerve, and the other excitable 'tissues. Lying deeper than the peculiarities of individual muscles, muscular tissue has certain common properties — physical, chemical, and physiological. The biceps muscle flexes the arm upon the elbow, and the triceps extends it. The external rectus rotates the eyeball outwards. The intercostal muscles elevate the ribs. The sphincter ani seals up by a ring-like contraction the lower end of the alimentary canal. These actions are very different, but the muscles that carry them out are at bottom very similar. And it cannot be doubted that the functional differences are due entirely, or almost entirely, to differences of anatomical connection, on the one hand with bones and tendons, on the other with the nerve-cells of the spinal cord and brain , The common properties in which all the skeletal muscles agree are the subject-matter of the general physiology of striated muscle. The cardiac muscle differs more, both in structure and in function, from the skeletal muscles than these do among themselves; the smooth muscle of the intestines and bloodvessels still more. But every muscular fibre, striped or unstriped, resembles every other 723 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES muscular fibre more than it does a nerve- fibre or a gland-cell or an epithelial scale. The properties common to all muscle make up the general physiology of muscular tissue. A nerve-fibre is at first sight very different from a muscular fibre. It has diverged more widely from the primitive type of undiffer- entiated protoplasm, but.it retains, in common with the muscle- fibre, susceptibility to stimulation, or excitability, the capacity for growth, and to a limited extent the capacity for reproduction ; and while it has lost the power of contraction or contractility, it has developed in a higher degree than any other tissue the power of conducting the excited state. This inheritance of primitive pro- perties, retained aliko by both tissues, is the basis of the general physiology of muscle and nerve. The electrical organ of Torpedo or Malapterurus is inter- mediate in some respects between muscle and nerve, and has properties common to both. In the gland-cell the chemical powers of native protoplasm have been specialized and developed. Con- tractility has been, in general, entirely lost ; but excitability remains. The idea that certain common en- dowments find expression in the action of muscle, nerve, electrical organ, gland, etc., in the midst of all their apparent differences, is the basis of the general physiology of the excitable tissues. It is impossible to understand the general physiology of muscle and nerve without some acquaintance with electricity. It would be out of place to give even a complete sketch of this preliminary but essential knowledge here; and the student is expressly warned that in this book the elementary facts and principles of physics are assumed to be part of his mental outfit. But in describing some of the electrical apparatus most commonly used in the study of this portion of our subject, and which are employed in the Practical Exercises, it may be useful to recall some of the physical facts involved. Batteries. — The Daniell cell is perhaps better suited for physiological work than any other voltaic element, although for special purposes Grove, Leclanche, bichromate of potassium or dry batteries may be employed (p. 197). Storage batteries or current from the street supply may also be used. Inside the Daniell cell the current (the positive electricity) passes from zinc to copper; outside, from copper to zinc. The copper is called the positive, the zinc the negative, pole. When the current is passed through a tissue, the electrode by which it enters is termed the anode, and that by which it leaves the tissue the kathode. The anode is, therefore, the electrode connected with the copper of the Daniell's cell; the kathode is connected with the zinc. Potential — Current Strength — Resistance. — We do not know what in reality electricity is, but we do know that when a current flows along Fig. 230. — Daniell Cell. A, outer vessel; B, copper; C, porous pot; D, zinc rod; D is supposed to be raised a little so as to be seen. PRELIMINARY DATA 725 a wire energy is expended, just as energy is expended when water flows from a higher to a lower level. Many of the phenomena of current electricity can, in fact, be illustrated by the laws of flow of an incom- pressible liquid. The difference of level, in virtue of which the flow of liquid is maintained, corresponds to the difference of electrical level, or potential, in virtue of which an electrical current is kept up. The positive pole of a voltaic cell is at a higher potential than the negative. When they are connected by a conductor, a flow of electricity takes place, which, if the difference of level or potential were not constantly restored, would soon equalize it, and the current would cease; just as the flow of water from a reservoir would ultimately stop if it was not replenished. If the reservoir was small, and the discharging-pipe large, the flow would only last a short time ; but if water was constantly being pumped up into it, the flow would go on indefinitely. This is prac- tically the case in the Daniell cell. Zinc is constantly being dissolved, and the chemical energy which thus disappears goes to maintain a constant difference of potential between the poles. Electricity, so to speak, is continually running down from the place of higher to the place of lower potential, but the cistern is always kept full. The difference of electrical potential between two points is called the electromotive force; and from its analogy with difference of pressure in a liquid, it is easy to understand that the intensity or strength of the current — that is, the rate of flow of the electricity between two points of a conductor — does not depend upon the electromotive force alone, any more than the rate of discharge of water from the end of a long- pipe depends alonfe on the difference of level between it and the reser- voir. In both cases the resistance to the flow must also be taken account of. With a given difference of level, more water will pass per second through a wide than through a narrow pipe, for the resistance due to friction is greater in the latter. In the case of an electrical current, a wire connecting the two poles of a Daniell's cell will represent the pipe. A thick short wire has less resistance than a thin long wire; and for a given difference of potential, of electric level, a stronger current will flow along the former. But for a wire of given dimensions, the in- tensity of the current will vary with the electromotive force. The relation between electromotive force, strength of current, and resistance 17 were experimentally determined by Ohm, and the formula C= ~ » which expresses it, is called Ohm's Law. It states that the current varies directly as the electromotive force, and inversely as the resist- ance. For the measurement of electrical quantities a system of units is necessary. The common unit of resistance is the ohm, of current the ampdre, of electromotive force the volt. The electromotive force of a Daniell's cell is about a volt. An electromotive force of a volt," acting through a resistance of an ohm, yields a current of one ampere. But the current produced by a Daniell's cell, with its poles connected by a wire of i ohm resistance, would be less than an ampere, because the internal resistance of the cell itself — that is, the resistance of the liquids between the zinc and the copper — must be added to the external resistance in order to get the total resistance, which is the quantity represented by R in Ohm's Law. Measurement of Resistance. — To find the resistance of a conductor, we compare it with known resistances, as a grocer finds the weight of a packet of tea by comparing it with known weights. The Wheatstone's bridge method of measuring resistance depends on the fact that if four resistances, AB, AD, BC, CD, are connected, as in Fig. 231, with each 726 THE PHYSIOLOGY OF 711 L CONTRACTILE TISSUES other, and with a galvanometer. G, and a battery, F, no current will A T$ 'R'"* flow through the galvanometer \\lien YTS=QFJ- In making the measurement, a resistance box, containing a large number of coils of wire of different resistances, is used. The resistances corresponding to AB and AD may be made equal, or may stand to each other in a ratio of i : 10, i : 100, etc. Then, the unknown resistance being CD, BC is adjusted by taking plugs out of the box till, on closing the current, there is either no deflection, or the deflection is as small as it is possible to make it with the given arrangement. Galvanometers. — A galvanometer is an instrument used to detect a current, to determine its direction, and to measure its intensity. Since, by Ohm's law, electromotive force, resistance, and current strength are connected together, any one of them may be measured by the galvanometer. A galvanometer of Fig. 231. Wheat- the kind ordinarily used in physiology consists essen- stone s Bridge, tially of a small magnet suspended in the axis of a coil of wire, and free to rotate under the influence of a current passing through the coil. The most sensitive instruments possess a small mirror, to which the magnet is rigidly attached. A ray of light is allowed to fall on the mirror, from which it is reflected on to a scale ; and the rotation oi the mir- ror is magnified and measured by the ex- cursion of the spot of light on the scale. The method of read- ing by a telescope can be applied to any mirror galvanometer, B and is often extreme- ly convenient in physiological work. Sometimes a small scale is fastened on the mirror itself, and o'bserved directly through a low-power microscope. Fig. 232. — Diagram of String Galvanometer. The string or fibre CC is stretched be- tween the poles of a powerful electromagnet. When a current passes down the string it is deflected in the direction of the larg» arrow a — i.e., at right angles to the magnetic field NS. When the current is reversed, the string moves in the opposite direction. The movements of the string can be observed by a micro- scope, A (objective E), passing through a hole bored through the centre of the magnet poles. For obtaining records a source of light is placed at B and con- centrated on the fibre by a condenser, F, and the movements of the shadow are recorded by photography. In the d'Arsonval galvanometer the current passes through a small coil of fine wire suspended in the field of a strong magnet. When the current passes the coil is deflected, carrying with it a small mirror attached to the suspending filament. A great advantage of this galvano- meter in many situations is that it is unaffected by neighbouring currents. PRELIMINARY DATA 727 The string galvanometer of Einthoven has peculiar merits for certain physiological purposes. It consists of a silvered quartz- or glass-fibre stretched in a very strong magnetic field. When traversed by a current the fibre is deflected, and by means of a beam of light the deflection is greatly magnified (Fig. 232). A rheocord is an instrument by means of which a current may be divided, and a definite portion of it sent through a tissue (Fig. 233). A compensator is simply a rheo- cord from which a branch of a current is led off, to balance or ' compen- sate ' any electrical difference in a tissue, like that which gives rise to the current of rest of a muscle, for example (Fig. 234). An electrometer is an instrument for measuring electromotive force — Fig. 233 — Diagram of Rheocord (after Du Bois-Reymond's Model). Fig. 234. — Compensator. Description of Fig. 233 : 1 to VII are pieces of brass connected with the wires a to f in such a way that, by taking out any of the brass plugs i to 5, a greater or less resistance may be interposed between the binding-screws A and B. The two wires a are connected by a slider s, filled with mercury or otherwise making contact between the wires. The current from the battery B divides at A and B, part of it passing through the rheocord, part through N, the nerve, muscle, or other conductor which forms the alternative circuit. When a sufficient resistance R is interposed in the chief circuit to make the total strength of the current independent of changes in the resistance of the rheocord, the strength of the current passing through N will vary inversely as the resistance of the rheocord. When all the plugs are in, and the slider close up to A, there is practically no resistance in the rheocord, and all the current passes across the brass pieces and plugs to B, and thence back to the battery. As s is moved father away from A, the resistance of the rheocoid is increased more and more, and the intensity of the current passing through N becomes greater and greater. The scale S shows the length of wire interposed for any position of s, and this gives a rough measure of the fraction of the current passing through N. When plug i or 2 is taken out, a resistance equal to that of the two wires a is interposed; plug 3, twice that of a ; plug 4, five times; plug 5, ten times. Description of Fig. 234': W is a wire stretched alongside a scale S. A battery B is connected to the binding-screws at the ends of the wire. A pair of unpolarizable electrodes are connected, one with a slider moving on a wire, the other through a galvanometer with one of the terminal binding-screws. In the figure a nerve is shown on the electrodes, one of which is in contact with an uninjured portion, the other with an injured part. The slider is moved until the twig of the compensating current just balances the demarcation current of the nerve and the galvanometer shows no deflection.. 728 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES that is, differences of electric potential. Lippmann's capillary elec- trometer has been much employed in physiology. A simple form, suitable for students working in a class where a considerable number of copies of the instrument is needed, can be conveniently made as follows : A glass tube is drawn out to a capillary at one end and filled with mer- cury. The tube is inserted into a small glass bottle,* and fastened in its neck by a cork or a plug of sealing-wax which does not quite fill the opening, so that the interior of the bottle is still in communication with the external air. The upper end of the tube is connected by a short piece of rubber tubing with a glass T-tube as in Fig. 235. The bottle is partially filled with 5 to 10 per cent, sulphuric acid, under which the capillary dips. By means of a small reservoir made from a piece of glass tubing filled with mercury, and connected with the stem of the T-tube, a little mercury is forced through the capillary so as to expel the air in it. When the pressure is lowered again, sulphuric acid is drawn up, and now lies in the capillary in contact with the meniscus of the mercury. A platinum wire fused through the tube, or simply inserted through its upper end, dips into the mercury. Another, passing through the cork, or, better, fused through the bottom of the bottle, makes contact with the sulphuric acid through some mercury. The bottle is fastened on the stage of a microscope, the capillary brought into focus, and the meniscus adjusted by raising or lowering the reser- voir. When the platinum wires are connected with points at different potential, a current begins to pass through the instrument, and the meniscus of the mercury in the capillary tube, where the current density is the greatest, becomes polarized by the ions separated from the sulphuric acid at the surface of contact between the acid and the mer- cury, so that the meniscus is no longer in equilibrium in the tube. The surface tension (p. 429) is diminished when the direction of the current is from mercury to acid (mercury at a higher potential than acid), and is no longer able to counterbalance the hydrostatic pressure of the mercury. The meniscus therefore moves down in the tube. With the opposite direction of current (mercury at a lower potential than acid) the surface tension is increased, and the meniscus moves up. The polarization develops itself almost instantaneously, and thus an electromotive force is at once established in the opposite direction to that between the points connected with the electrometer, and equal to it so long as the external electromotive force is not sufficiently great to cause continuous electrolysis of the acid — that is, so long as it is below about 2 volts. The external current is therefore at once compensated, and after the first moment no current passes through the instrument, which is accordingly not a measurer of current, but of electromotive force. Induced Currents. — When a coil of wire in which a current is flowing is brought up suddenly to another coil, a momentary current is developed in the stationary coil in the opposite direction to that in the moving coil. Similarly, if instead of one of the coils being moved a current is sent through it, while the other coil remains at rest in its neighbour- * A parallel-sided bottle is best, as it gives the clearest image of the menis- cus. But it is easiest to make a cylindrical bottle from a piece of wide glass tubing, and to insert a platinum wire into it before closing it at the bottom in the blow-pipe flame. The tube can then be firmly fastened with sealing-wax in a depression in a piece of wood, the wire being brought out through a hole in the wood. Once the instrument is arranged, there is little chance of the capillary getting broken, and there is very little evaporation of the acid. PRELIMINARY DATA 729 hood, a transient oppositely-directed current is set up in the latter. When the current in the first coil is broken, a current in the same direction is induced in the other coil. Fig. 236. Fig. 235. — A Simple Capillary Electrometer. B bottle containing sulphuric acid; Hg, mercury; E,E', platinum wires. £ dips into the mercury in the vertical tube, and E' is fused through the bottom of B, so as to make contact with the mercury in B, the other end of it passing out through a small hole in the wooden platform F, on which B rests. F is fastened to the stage of the microscope S by a pin, G, passing through one of the clip-holes, and to the wooden upright D by the pin H. D fits tightly over the microscope stage, but can be moved laterally a little so as to bring the capillary into the middle of the field. /, stem of glass T-tube passing through a hole in D. L, rubber tube connecting the capillary point with the vertical portion of the T-tube. A is a reservoir containing mercury connected by the rubber tube M to I. A can be raised or lowered by sliding it in the clips K. C, magnified portion of the capillary tube showing the meniscus. Fig. 236. — Capillary Electrometer (after Frey). Du Bois-Reymond's Sledge Inductorium (Fig, 237). — This consists of two coils, the primary and the secondary, the former having a com- paratively small number of turns of fairly thick copper wire, the latter a large number of turns of thin wire. The object of this is that the resistance of the primary, which is connected with one or more voltaic 730 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES cells, may not cut down the current too much; while the currents induced in the secondary, having a high electromotive force, can readily pass through a high resistance, and are directly proportional in intensity to the number of turns of the wire. By means of various binding-screws and the electro -magnetic inter- rupter or Neef 's hammer, shown in the figure and explained below it, the current can be made once in the primary or broken once, or a con- stant alternation of make and break can be kept up. We can thus get a single make or break shock in the secondary, or a series of shocks, sometimes called an interrupted or faradic current. Such a series of stimuli can also be got by making and breaking a voltaic current at any given rate. A ' self -induced ' current can also be obtained from a single coil ; for instance, from the primary coil alone of the induction apparatus. The Fig. 237. — Du Bois-Reymond's Inductorium. B, primary, B', secondary, coil. H, guides in which B' slides, with scale; D, electro-magnet; E, vibrating spring; i, wire connecting wire of D to end of primary; v, screw with platinum point, connected with other end of primary ; A, A', binding-screws, to which are attached the wires from battery. A' is connected with the wire of the electro-magnet D; and through it and t with the primary. reason of this is, that when a current begins to flow through any turn of a coil of wire it induces in all the other turns a current in the opposite direction, and, when it ceases to flow, a current in the same direction as itself. The former current, ' the make extra shock,' being in the opposite direction to the inducing current, is retarded in its develop- ment, and reaches its maximum more slowly than ' the break extra shock.' But, as we shall see, the suddenness with which an electrical change is brought about is one of the most important factors in elec- trical stimulation, and therefore the break extra shock is a much more powerful stimulus than the make. Owing to these self-induced cur- rents, the stimulating power of a voltaic stream may be much in- creased by putting into the circuit a coil of wire of not too great reristance. The self-induction of the primary also affects the stimulating power of the currents induced in the secondary; the shock induced in the secondary by break of the primary current is a stronger stimulus than that caused at make of the primary. The reason is that with a given distance of primary and secondary, and a given intensity of the voltaic current in the primary, the abruptness with which the induced current PRELIMINARY &ATA 73» in the secondary is developed depends upon the rapidity with which the primary current reaches its maximum at closing, or its minimum (zero) at opening. Now, the make extra current retards the development of the primary current, while in the opened circuit of the primary coil the current intensity falls at once to zero. The inequality between the make and break shocks of the secondary coil can be greatly reduced by means of Helmholtz's wire. Connect one pole of the battery with v (Fig. 237), and the other with A'. Join A and A' by a short, thick wire. With this arrangement the primary cir- cuit is never opened, but the current is alternately allowed to flow through the primary, and short-circuited when the spring touches v. The ' make ' now corresponds to the sudden increase of intensity of the current in the primary when the short-circuit is removed, and the ' break ' to its sudden decrease when the short-circuit is established. In both cases self-induced currents are developed, and therefore both shocks are weakened. But the opening stimulus is now slightly the weaker of the two, because the opening extra shock has to pass through a smaller resistance (the short-circuit) than the closing extra shock (which passes by the battery), and therefore opposes the decline of current intensity on short-circuiting more than the closing shock opposes the increase of current intensity on long- circuiting through the primary. By means of wires con- nected with the terminals of the secondary coil, and leading to electrodes, a nerve or muscle may be stimulated . It is usual to connect the wires to a short - circuiting key (Fig. 240), by opening which the induced current is thrown lated. For some purposes Fig. 238. — Unpolarizable Electrodes. A. hook- shaped; B, U -tubes; C, straight; D, clay in contact with tissue; S, saturated zinc sulphate solution ; Z, amalgamated zinc wire. into the tissue to be stimu- the electrodes may be of platinum; but all metals in contact with moist tissues become polarized when currents pass through them — that is, have decomposition products of the electrolysis of the, tissues deposited on them. And as any slight chemical difference, or even perhaps a difference of physical state, be- tween the two electrodes will cause them and the tissues to form a battery evolving a continuous current, it is often desirable to use un- polarizable electrodes. Unpolarizable Electrodes. — Some convenient forms of these are represented in Fig. 238. A piece of amalgamated zinc wire dips into saturated zinc sulphate solution contained in the upper part of a glass tube. The lower end of the tube may be straight, but drawn out so as to terminate in a not very large opening, or it may be bent into a hook, in the bend of which a hole is made. Before the tube is filled with the zinc sulphate solution, the lower part of it is plugged with china clay made up with physiological salt solution. The clay just projects through the opening, and thus comes in contact with the tissue. When these electrodes are properly set up, there is very little polarization for several hours, but for long experiments, U-shaped tubes, filled with saturated zinc sulphate solution, are better. The amalgamated zinc dips into one limb, and a small glass tube filled with clay, on which the tissue is laid, into the other. 732 Pohl's Commutator or Reverser (Fig. 239) consists of a block of paraf- fin or wood with six mercury cups, each in connection with a binding- screw (not shown in the figure). Cups i and 6 and 2 and 5 are connected by copper wires, which cross each other without touching. The bridge consists of a glass or vulcanite cross-piece a, to which are attached two wires bent into semicircles, each connected with a straight wire dip- ping into the cups 3 and 4 respectively. With the bridge in the posi- tion shown in the figure, a current coming in at 4 would pass out by the wire connected with i, and back again by that connected with 2, in the direction shown by the arrows. When the bridge is rocked to the other side so that the bent wires dip into 5 and 6, the direction of the current is reversed. The cross-wires may be taken out altogether, and the commutator used to send a current at will through either of two circuits, one connected with i and 2, and the other with 5 and 6. Du Bois -Reymond's Short- circuiting Key. — A cheap and convenient form is shown in Fig. 240. Time - Markers — Electric Sig- nal.— It is of importance to know the time relations of many physiological phenomena which are graphically recorded ; for Fig. 239.— Pohl's Commutator. Fig. 240. — Short-Circuiting Key. example, the contraction of a skeletal muscle or the beat of a heart. For this purpose a tracing showing the speed, of the travelling sur- face in a given time is often taken simultaneously with the record of the movement under investigation. For a slowly-moving surface it is sufficient to mark intervals of one or two seconds, and this is very readily done by connecting an electro-magnetic marker (such as the electric signal of Deprez) with a circuit which is closed and broken by the seconds pendulum of an ordinary clock or a metronome (Fig. 88, p. 195). Special clocks have also been constructed which permit of the time intervals being varied. For shorter intervals a tuning-fork is used, which makes and breaks a circuit including an electromagnetic marker, or writes on the drum directly by means of a writing-point attached to one of the prongs. Amoeboid movement (p. 16) is the most primitive, the least elaborated form of contraction. The maximum velocity of the movement has been reckoned at o-ooS millimetre a second. Stimu- CILIA 733 lation with the constant current or induction shocks causes the whole of the pseudopodia to be drawn in. This illustrates a universal property of protoplasm, excitability, or the power of re sponding to certain influences, or stimuli, by manifestations of the peculiar kind which we distinguish as vital or physiological. Many unicellular organisms and the chief varieties of the white blood- corpuscles possess the power of amoeboid movement ; and we have already dwelt upon some of the important functions fulfilled by such movement in the higher animals and in man. A great dis- tinction between this kind of contraction and that of a muscular fibre is that it takes place in any direction. Cilia. — Cilia possess a higher and more specialized grade of contractility. They are very widely distributed in the animal kingdom; and analogous structures are also found in many low plants, such as the motile bacteria. In the human subject ciliated epithelium usually consists of several layers of cells, the most superficial of which are pear-shaped, the broad end being next the surface, and covered with extremely fine processes, or cilia, about 8 fj, in length, which are planted on a clear band. It lines the respiratory passages, the middle ear and Eustachian tube, the Fallopian tubes, the uterus above the middle of the cervix, the epididymis, where the cilia are extremely long, and the central cavity of the brain and spinal cord. Ciliary motion can be readily studied by placing a scraping from the palate of a frog or a small portion of a gill of a fresh-water mussel under the microscope in a drop of physiological salt solution. The motion of the cilia is at first so rapid that it is impossible to make out much, except that a stream of liquid, recognized by the solid particles in it, is seen to be driven by them in a constant direction along the ciliated edge. When the motion has become less quick, which it soon does if the tissue is deprived of oxygen, it is seen to consist in a swift bending of the cilia in the direction of the stream, followed by a slower recoil to the original position, which is not at right angles to the surface, but sloping streamwards. All the cilia on a tract of cells do not move at the same time ; the motion spreads from cell to cell in a regular wave. The energy of ciliary motion may be considerable, although far inferior to that of muscular con- traction. The work which cilia are capable of performing can be calculated by removing the membrane, fixing it on a plate of glass, cilia outwards, putting weights on the glass plate, and allowing the cilia, like an immense number of feet, to carry it up an inclined plane. Bowditeh found in this way that the cilia on a square centimetre of mucous membrane did nearly 7 gramme-millimetres of work per minute (equal to the raising of 7 grammes to a height of a milli- metre). 734 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Since the cilia in the respiratory tract all lash upwards, they must play an important part in carrying up foreign particles taken in with the air, and the mucus in which they are entangled, as well as patho- logical products. Engelmann found that the energy of ciliary motion in- creases as the temperature is raised up to about 40° C., after which it Fig. 241— Ciliated Cell (M. Heidenhain). From a ' liver duct ' of the garden snail x 2,500. Fig. 242. — Ciliated Cell (Schneider). From a flatworm (Planocera folium). i, space between two adjoining ciliated cells; 2, basal bodies; 4, inner granule; 5, 'cilia roots'; 6, boundary layer. diminishes quickly. Over-heating causes cilia to come to rest, but if the temperature has not been too high, and has not acted too long, they recover on cooling, thus exhibiting the phenomena of heat standstill which we have already studied in the heart. It is not well understood in what way the contraction of the cilia depends upon their connection with the body of the ciliated cell. Very few cases occur in which cilia have the power of independent motion when severed from the cell-body. It has been observed in certain low forms of animals that cilia which have been broken off from the cell are still able to contract when a small portion of the substance of the cell-body at the point where the cilium is attached to the cell, the so-called basal piece, or basal body (Fig. 242), has come off along with them. In other forms isolated cilia can contract in the absence of anything corresponding to the basal piece. It cannot, therefore, be said that continuity with the basal piece is absolutely necessary. Nor is it known what significance for the ciliary movements is possessed by the long fibrillse, called the ' roots of the cilia,' which in some animals PHYSICAL PROPERTIES OF MUSCLE 735 run down through the cell from the basal bodies (Figs. 241, 242). In some worms and molluscs ciliated cells are supplied with nerve-fibres, but this has not been demonstrated for the higher animals. SECTION II. — PHYSICAL PROPERTIES AND STIMULATION OF MUSCLE. Since most of our knowledge of the general physiology of muscle has been gained from striped muscle, in what follows we always refer to ordinary skeletal muscle, unless it is otherwise stated. The sartorius and the gastrocnemius are the classical objects for experiments on striated muscle. For smooth muscle the adductor muscle of Anodon, the fresh-water mussel, a ring cut from the middle portion of the frog's stomach, the rabbit's ureter and uterus, and the cat's bladder, have been most used. Physical Properties of Muscle — Elasticity. — All bodies may have their shape or volume altered by the application of force. Some require a large force, others a small force, to produce a sensible amount of dis- tortion. The elasticity of a body is the property in virtue of which it tends to recover its original form or bulk when these have been altered. Liquids and gases have only elasticity of volume; solids have also elasticity of form. Most solids recover perfectly, or almost perfectly, from a slight deformation. The limits of distortion within which this occurs are called the limits of elasticity, and they vary greatly for different substances. Living muscle has very wide limits of elasticity; it may be deformed — stretched, for example — to a very considerable extent, and yet recover its original length when the stretching force ceases to act. The extensibility of a body is measured by the ratio of the increase of length, produced by unit stretching force per unit of area of the cross-section, to the original length of a uniform rod of the substance. Is If e is the extensibility, e= =-=> where / is the increase of length, L the original length, 5 the cross-section, and F the stretching force. Suppose we wish to compare the extensibility of two substances. Let A and B be strips or rods of the substances, the length of A being 500 mm., that of B 1,000 mm.; the cross-section of A, 100 sq. mm., of B, 200 sq. mm. Let the elongation produced by a weight of i kilo be 10 mm. in each, then the extensibility of A is — — = 2 ; and that 500 x i 10 x 200 of B is - — = 2; that is, the substances are equally extensible. . 1,000 x i Young's modulus of elasticity, or the coefficient of elasticity, is the quotient of the deforming force acting on unit area of the given body by the deformation produced (within the limits of elasticity). In the "pj T f? above example it is — -*-?-• that is, -=—, the reciprocal of the extensi- bility e. For steel the coefficient of elasticity is very large, for muscle small. Or, as we may otherwise express it, living muscle within its limits of elasticity is very extensible ; a small force per unit area of cross-section of a prism of it will produce a comparatively great elonga- tion. The extensibility, however, diminishes continually with the elongation, so that equal increments of stretching force produce always 736 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES less and less extension. If, for instance, the sartorious or semi-mem- branosus of a frog be connected with a lever writing on a blackened surface, and weights increasing by equal amounts be successively attached to it, the recording surface being allowed to move the same distance after the addition of each weight, a series of vertical lines, representing the amount of each elongation, will be traced. When the lower ends of all the vertical lines are joined, a smooth curve with the concavity upwards is obtained (Fig. 243). This is a property common to living and dead muscle and to other animal structures, such as arteries. Marey's method, in which the weight is continuously in- creased from zero and then continuously decreased to zero again by the flow of mercury into and out of a vessel attached to the muscle, gives directly the curve of extensibility. The elongation of a steel rod or other inorganic solid is proportional within limits to the extending force per unit of cross-section ; and a curve plotted with the weights for abscissae and the amounts of elongation for ordinates would be a straight line. But this is not a fundamental dis- tinction between animal tissues, and the materials of unorganized nature, as some writers seem to suppose. For when the slow after-elongation which follows the first rapid increase in length in the loaded, 243 — Curves of Extensi- excised muscle is waited for, the curve of bility. M, of muscle; S, of an extensibility comes out a straight line ordinary inorganic solid. (Wundt), and within limits this is also the case for human muscles in the intact body. And although a steel rod much more quickly reaches its maximum elongation for a given weight when loaded, and its original length when the weight is removed, than does a muscle, time is required in both cases, and the difference is one of degree rather than of kind. When muscle (striated or smooth) is not stretched beyond the limit of physiological relaxation, the amount of stretching is proportional to the weight, and the same is true of all the simple tissues of the body (Haycraft). Dead muscle is less extensible than living, and its limits of elasticity are much narrower. In the state of contraction the extensibility is increased in excised frog's muscle. When fatigue comes on after many excitations, the after-elongation becomes more pronounced, but the return after unloading is very incomplete. Donders and Van Mans- veldt have found that contraction causes little difference in the muscles of a living man, although fatigue increases the extensibility. The great extensibility and elasticity of muscle must play a con- siderable part in determining the calibre of the vessels, and in lessening the shocks and strains which the heart and the vascular system in general are called upon to bear, and must contribute much to the smoothness with which the movements of the skeleton are carried out, and immensely reduce the risk of injury to the bones as well as to the muscles themselves, the tendons and the other soft tissues. And not only is smoothness gained, but economy also; for a portion of the energy ol a sudden contraction, which, if the muscles were less ex- tensible and elastic, might be wasted as heat in the jarring of bone against bone at the joints, is stored up in the stretched muscle and again given out in its elastic recoil. The skeletal muscles, too, are even at rest kept slightly on the stretch, braced up, as it were, and STIMULATION OF MUSCLE 737 ready to act at a moment's notice without taking in slack. This is shown by the fact that a transverse wound in a muscle ' gapes,' the fibres being retracted, in virtue of their elasticity, towards the fixed points of origin and insertion. Smooth muscle, as we meet it in the hollow viscera, is highly distensible and elastic, as is suited to organs whose capacity is continually varying within wide limits (Fig. 244). In the further study of muscle it is necessary first of all to consider the means we have of calling forth a contraction — in other words, the various kinds of stimuli. Stimulation of Muscle. — A muscle may be excited or stimulated either directly or through its motor nerve. It is usual to classify stimuli as electrical, mechanical, chemical, or thermal. Electrical stimuli are by far the most commonly employed, and will be dis- cussed in detail. A prick, a cut, or a blow are examples of mechani- Fig. 244. — Extensibility of Smooth Muscle (Griitzner). The upper group of four cells (i to 4) is from a hollow organ, whose walls are contracted, and its lumen almost abolished; the under group represents the same fibres when the organ is full. The fibres are longer and somewhat darker. They are also displaced somewhat along each other. cal stimuli. The action of a fairly strong solution of common salt or of a dilute solution of a mineral acid is usually described as chemical stimulation. But in considering the excitation of nerve (p. 783) we shall see that physical changes are often mixed up with so-called chemical stimulation. The contraction caused is not a single brief twitch, as is the case with a not too severe mechanical excitation, but a sustained contraction or a tetanus. Sudden cooling or heating acts as a stimulus for muscle, but thermal stimulation is somewhat uncertain. It is not quite settled whether the contrac- tion which can be obtained from a muscle when it is subjected to brief local heating — to a ' thermic shock/ as some writers prefer to say (e.g., by the momentary glow of a platinum wire below but not touching it) — is an ordinary muscular contraction, or a physical, although transient, contracture analogous to that caused by certain drugs (Waller). Smooth, like striped, muscle is susceptible to electrical, mechanical, thermal, and chemical stimulation. In addition, in certain situations it can be excited by light (photic stimulation), as in the case of the excised iris of fish and amphibia. In all artificial stimulation there is an element of sudden or abrupt change, of shock, in other words; but we cannot tell in what the ' natural ' or ' physiological ' stimulus to muscular contraction in 47 738 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES the intact body really consists, nor how it differs from artificial stimuli. All we know is that there must be a wide difference, and that our methods of excitation must be very crude and inexact imitations of the natural process. Direct Excitability of Muscle. — The famous controversy on the existence of independent ' muscular irritability ' has long been forgotten, and has no further interest except for the antiquaries of science, if such exist. The direct excitability of muscle in the modern sense is not quite the same as the ' muscular irritability/ the dis- cussion of which occupied Haller and his contemporaries. What the modern physiologists have been called upon to decide is whether muscular fibres can be caused to contract except by an excitation that reaches them through their nerves. In this sense there can exist no doubt that muscle is directly excitable, and some of the proofs are as follows: (i) The ends of the frog's sartorius contain no nerves, yet they respond to direct stimulation. (2) Certain chemical stimuli — ammonia, for instance — excite muscle but not nerve. (3) When the motor nerves of a limb are cut they degenerate, and after a certain time stimulation of the nerve-trunk causes no muscular contraction, while the muscles, although atrophied, can be made to contract by direct stimulation. (4) Finally, there is the cele- brated curara experiment of Claude Bernard, which is described in a somewhat modified form in the Practical Exercises, p. 811. A ligature is tied firmly round one thigh of a frog, omitting the sciatic nerve; then curara is injected, and in a short time the skeletal muscles are paralyzed. That the seat of the paralysis is not the contractile substance of the muscles itself is shown by their vigorous response to direct stimulation. The ' block ' is not in the nerve- trunk, nor above it in the central nervous system, for the ligated leg is often drawn up — that is, its muscles are contracted — although the poison has circulated freely in the sacral plexus and the spinal cord. Further, if the nerve of the ligated leg be prepared as high up above the ligature as possible, where the curara must undoubtedly have reached it (just above the ligature the nerve has been isolated and the circulation in it more or less interrupted), stimulation of it will cause contraction of the muscles of the limb ; while excita- tion of the other sciatic is ineffective. It can be also shown, by means of the negative variation or current of action (p. 824), that a nerve-trunk on which curara has acted remains excitable, and capable of conducting the nerve- impulse. The conclusion, therefore, is that the curara paralyzes neither nerve-fibre nor the contractile substance of the muscular fibre, but some link between the two. If the assumption be made that the efferent medullated nerve-fibres within the muscle, since STIMULATION OF MUSCLE 739 they are anatomically similar to those in the nerve-trunk till near their terminations, are similarly affected by curara — and it is a justifiable assumption — the seat of the curara paralysis must either be the nerve-ending or some mechanism, physiological if not anatomical, interposed between the nerve-ending and the con- tractile substance. Now, Langley has shown that the contractions caused by the local application of dilute nicotine solution to points of the skeletal muscles of the frog, both in normal muscles and in muscles whose motor nerves and nerve-endings have degenerated after section of the nerves, are prevented by curara. He there- Fig- 245. — -Frog's Motor Nerve-Ending (Wilson). A, B, C, three muscle-fibres. The medullated nerve a loses its medullary sheath and breaks up on B at i. It gives off at 2 a large non-medullated branch, which also breaks up on B. The nerve- endings send ultraterminal fibrillae to A, B, and C, some of which were seen to end in small knobs. A separate non-medullated nerve, n, is shown, which forms a small plexus on B, one fibre of which penetrates to a lower plane than the other, and ends by forming a knob under the sarcolemma. fore concludes that, since nicotine produces its effects by a direct action on muscle, and not by an action on nerve-endings or on any special structure (such as the protoplasmic mass or ' sole ' at the nerve-ending in many animals) interposed between the nerve and the muscle, no such special structure existing in the frog (Fig. 245), curara must also act directly on the muscle. But obviously curara does not paralyze the general contractile substance of the muscle, else the curarized muscle would not contract on direct stimulation. Langley accordingly assumes that, in addition to the contractile or ' general ' substance, ' receptive ' substances exist in the fibre, through which the excitation is tx^ansferred to the con- tractile substance when the motor nerve is stimulated. He pictures these receptive substances as ' side-chains ' of the contractile mole- cule, in accordance with Ehrlich's theory of immunity (p. 31), and distinguishes those in the neighbourhood of the nerve-ending 740 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES from those present throughout the muscle fibre. Both the slow local tonic contraction an4 the quick, brief conducted contractions or twitches set up in a muscle fibre by nicotine, but especially the latter, are much more easily elicited in that part of it which lies under the nerve-ending than elsewhere. Indeed, the position of the nerve-endings in the superficial fibres of a muscle can be ascer- tained by observing the points which respond most readily to nico- tine. Nicotine and curara, etc., are supposed to combine with the receptive substance, which is then in both cases rendered incapable of being affected by nerve impulses. In the case of nicotine an additional action results from the combination with the receptive substance — -viz., the change in the contractile substance which leads to contraction. Curara paralyzes the transmission of the excitation from the motor nerves to smooth muscle — the muscles of the Fig. 246. — Tonic Contraction of Muscle during Passage of Constant Current. Two sartorius muscles of frog connected by pelvic attachments. Current from 12 small Daniell cells in series passed through their whole length. Current closed at m, opened at ft. Time trace, two-second intervals. bronchi, for instance — with much greater difficulty than to ordinary skeletal muscle, and the same is true of the inhibitory nerves of the heart. The action of curara gives us the means of stimulating muscle directly; when electrical currents are sent through a non-curarized muscle, there is in general a mixture of direct and indirect stimula- tion, for the nerve-fibres within the muscle are also excited. Induced currents stimulate nerve more readily than muscle. Voltaic currents may excite a muscle whose nerves have degenerated, while induced currents are entirely without effect. For direct stimulation, a curarized frog's sartorius or semi-mem- branosus is generally used on account of their long parallel fibres. For indirect excitation, a muscle-nerve preparation, composed of a frog's gastrocnemius with the sciatic nerve attached to it, is commonly em- ployed, as it is easy to isolate the muscle without hurting its nerve. STIMULATION OF MUSCLE 741 Stimulation by the Voltaic Current. — While the current continues to pass through a nerve without any sudden or great change in its in- tensity, there is no stimulation, and the muscle connected with the nerve remains at rest. The same is true of striated muscle when a weak current is passed directly through it. But in muscle the con- stancy of the rule is more and more frequently broken by exceptional results as the current is strengthened, a state of permanent contrac- tion being very apt to show itself during the whole time of flow (Wundt) (Fig. 246). Above a certain intensity of current a greater or less degree of permanent contraction is invariably produced. This is some- times called the ' closing tetanus.' It is, however, not a true tetanus, but a tonic contraction, which is strongest in the neighbourhood of the kathode, and does not spread far from it. A similar condition, the so-called galvanotonus, is normally seen in human muscles when they or their motor nerves are traversed by a stream of considerable inten- sity. Under certain con- ditions, too — e.g., when a strong current is allowed to flow for a comparatively long time through a muscle — the muscle remains contrac- ted after the opening of the current (so-called ' opening or Hitter's tet- anus '). Smooth muscle is excited to contraction even when a voltaic cur- rent is very gradually passed into it and slow- ly increased, and again when it is caused very gradually to disappear. But striped muscle is not stimulated under Fig. 2471 — Tonic Contraction during and after Flow of Voltaic Current. Curve from frog's gastroc- nemius. At M constant current closed, at B broken. Contracture continues after opening of current. Time trace, two-second intervals. these conditions. For nerve, and with these qualifications for muscle, too, the law holds that the voltaic current stimulates at make and at break, but not during its passage. Or, generalizing this a little, since it has been shown that a sudden increase or decrease in the strength of a current already flowing also acts as a stimulus, we may say that the voltaic current stimulates only when its intensity is suddenly and sufficiently increased or diminished, but not while it remains constant.* When a strong current is closed through a muscle there is an im- mediate sharp contraction (initial contraction). The muscle then promptly relaxes, but incompletely. When the current is opened, there is another contraction (Fig. 247). The force of the initial con- traction, as measured by the resistance necessary to prevent it, is greater than that of the tonic contraction which follows it. A second law of great theoretical importance is that of polar stimula- tion. At make the stimulation occurs only at the kathode ; at break only at the anode. This is true both for muscle and nerve, but it is most * This law of Du Bois-Reymond has been questioned by Hoorweg and others. It seems to need modification, but the subject cannot be discussed here. 743 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES directly and simply demonstrated on muscle. A long parallel-fibred curarized muscle is supported about its middle; the two ends, which hang down, are connected with levers writing on a revolving drum, and a current is sent longitudinally through the muscle. It is not difficult to see from the tracings that at make the lever attached to the kathodic end moves first, and that the other lever only moves when the contrac- tion started at the kathode has had time to reach it in its progress along the muscle. Similarly, at break the lever connected with the anodic end moves first. The law of polar excitation holds both for striated and for smooth muscle. Not only is there no excitation of unstriped muscle at the anode on closure of the current, but a previ- ously existing contraction disappears. For skeletal muscle the make is stronger than the break contraction. It has not been proved that this is the case for smooth muscle. SECTION III. — PHYSICAL AND MECHANICAL PHENOMENA OF THE MUSCULAR CONTRACTION. When a muscle contracts, its two points of attachment, or, if it be isolated, its two ends, come nearer to each other; and in exact proportion to this shortening is the increase in the average cross- section. The contraction is essentially a change of form, not a change of volume. The most delicate observations fail to detect the smallest alteration in bulk (Ewald). Living fibres kept con- tracted by successive stimuli can be examined under the microscope ; or fibres may be ' fixed ' by reagents like osmic acid, and sometimes a very good opportunity of studying the microscopic changes in contraction is given by a group of fibres in which the ' fixing ' reagent has caught a wave of contraction, and, so to speak, pinned it down. It is then seen that the process of contraction in the fibre is a miniature of that in the anatomical muscle. The individual fibres shorten and thicken, and the sum-total of this shortening and thickening is the muscular contraction which we see with the naked eye. The phenomena of the muscular contraction may be classified thus: (i) Optical, (2) Mechanical, (3) Thermal, (4) Chemical, (5) Sonorous, (6) Electrical. (5) will be treated under ' Voluntary Contraction ' ; (6) in Chapter XV. (i) Optical Phenomena — Microscopic Structure of Striped Muscle. — The structure of striped muscle has long been the enigma of histology ; and the labours of many distinguished men have not sufficed to make it clear. On the contrary', as investigations have multiplied, new theories, new interpretations of what is to be seen, have multiplied in proportion, and a resolute brevity has become the chief duty of a writer on elementary physiology in regard to the whole question. The muscle-fibre, the unit out of which the anatomical muscle is built up, is surrounded by a structureless membrane, the sarcolemma. The length and breadth of a fibre vary greatly in different situations. The maximum length is about 4 cm. ; the breadth may be as much as 70 /* and as little as 10 ft. When we come to analyze the muscle- fibre and to determine out of what units it is built up, the difficulty begins. The fibre shows alternate dim and clear transverse stripes, and OPTICAL PHENOMENA OF MUSCULAR CONTRACTION 743 can actually be split up into discs by certain reagents. It also shows a longitudinal striation, and can be separated into fibrils. Some have supposed that the discs are the real structural units which, piled end to end, make up the fibre. The fibrils they consider artificial. This view is erroneous. It seems certain that the fibres are built up from fibrils ranged side by side, and that the discs are artificial. The con- tents of the muscle-fibre appear to consist of two functionally different substances, a contractile substance, and an interstitial, perhaps nutri- tive, non-contractile material of more fluid nature. The contractile substance is arranged as longitudinal fibrils embedded in interfibrillar matter (sarcoplasm). In a muscle impregnated with chloride of gold the interfibrillar matter appears as a network. Schafer has described the contractile elements of the muscle-fibre (Figs. 248, 249) as fine columns (sarcostyles), divided into segments (sarcomeres) by thin transverse discs (Krause's membranes), occu- pying the position of the middle of each light stripe. Each sarcomere contains a sarcous element la. por- , tion of the dark stripe) with a clear substance at its ends, filling up the Fig. 248. — Living Muscle of Water- Beetle (highly magnified) (Schafer). s, sarcolemma ; a, dim stripe ; b, bright stripe; c, row of dots in bright stripe, which appear to be the enlarged ends of rod-shaped particles, d, but in reality represent expansions of the interstitial sub- stance (sarcoplasm). Fig. 249. — Portion of Leg Muscle of Insect, treated with Dilute Acetic Acid (Schafer). S, sarco- lemma; D, dot -like en- largement of sarcoplasm; K, Krause's membrane. The sarcous elements have been swollen and dissolved by the acid. space between the sarcous element and Krause's membrane, and con- stituting a portion of the light stripe. The sarcous element is itself double, and if the fibre be stretched, the two portions separate at a line which runs transversely across the middle of the dim stripe (Hensen's line). Schafer considers that the appearance of longitudinal fibrillation in the sarcous elements is due to the presence in them of fine longi- tudinal canals or pores. The Krause's membrane of the individual fibrils is scarcely ever visible in an intact mammalian fibre, and the apparent line in the clear stripe of an intact fibre is an optical appearance due to interference of light. Kuhne, who was fortunate enough to find one day a small nematode worm moving in the interior of a fibre, saw it pass along the fibre with perfect freedom, ignoring Krause's membrane. Possibly, however, it was moving in the sarcoplasm, the fibrils being simply pushed aside. 744 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Changes during Contraction — Theories of Contraction. — In contractions., according to Schafer, the clear substance between Krause's membrane and the sarcous element passes into the canals, which are open towards Krause's membrane, but closed towards Hensen's line. The sarcous element therefore swells up, and the sarcomere is shortened. In the extended muscle the clear substance leaves the pores of the sarcous element, and accumulates in the space between it and Krause's mem- brane. The sarcomere is thus lengthened and narrowed. While the existence of Schafer's pores is not admitted by all observers, there is a pretty general agreement that the sarcomere, like the cytoplasm of an amoeboid cell, does consist of two substances, one of which (the hyalo- plasm of the cell, the clear material of the sarcomere) interpenetrates the other (spongioplasm of the cell, substance of the sarcous element) ; and that in relaxation the clear fluid passes from the sarcous element to the ends of the sarcomeres, whereas in contraction it passes in the reverse direction into the sarcous elements. Whether the fluid passes into and out of the meshes of an actual network, or along actual physical pores in the sarcous element, or whether it is transferred by some process like molecular imbibition (p. 426), need not be discussed here, since it is not definitely known. The fundamental question by what process the transference is determined when,, the muscle is excited also remains unsettled. So far as is known at present, it is probable that the mechanical energy of the contracting muscle must be derived from the transformation of chemical energy into one of three forms : energy associated with osmotic processes, energy associated with imbibition, and energy associated with changes of surface tension. It is not diffi- cult to see that a sudden increase in the osmotic concentration in the sarcous element, due to the breaking up of large molecules or colloid aggregates into small molecules, or the liberation of electrolytes from the colloids, might lead to the rapid passage of water into it from the bright bands. A sudden change of permeability of the sarcous elements for dissolved substances in the clear fluid would have a similar effect. The same is true of a change in their power of imbibi- tion. But, according to Bernstein, it is scarcely to be supposed that the extraordinarily rapid movement of water molecules which must occur in contraction can be accounted for either by osmosis or by im- bibition. A more plausible theory is that the surface tension — say between the substance of the sarcous element and the clear fluid — is altered. That the shortening of the muscle in fatigue (p. 749) and rigor (p. 774), as well as its shortening in normal contraction, is due in some way to the liberation of metabolic products, especially lactic acid, is a theory of some standing, and fresh evidence in its favour has been recently supplied. Thus it has been pointed out that the course of heat production in the active muscle, and its relation to the time of the mechanical response, and the development and time relations of the electrical change which precedes that response, can be very naturally explained on the supposition that the liberation of lactic acid on or near some surface in the contractile substance is an essential factor in the contraction (Mines, etc.). It is known that in the presence of acid on the surface of certain colloid structures shortening occurs (Fischer and Strietman). The substance of the sarcous element which forms the dark stripe is doubly refracting, and therefore rotates the plane of polarization, but the clear substance of the light stripe is singly refracting. When an uncontracted fibre is viewed with crossed nicols, the dim stripe accordingly appears bright in the otherwise dark field. In the con- tracted fibre the doubly refractive material remains in the stripe which MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 745 is dim in ordinary light. There is no transference of it, but, according to most writers, the bands which are dim in ordinary light increase in size by the transference of liquid from the isotropous band. Diffraction Spectrum of Muscle. — When a beam of white light passes through a striped muscle, it is broken up into its constituent colours, and a series of diffraction spectra are produced, just as happens when the light passes through a diffraction grating (a piece of glass on which are ruled a number of fine parallel equidistant lines). The nearer the lines are to each other, the greater is the displacement of a ray of light of any given wave-length. It has accordingJy been found that when ?, muscular fibre contracts, the amount of displacement of the diffraction spectra increases. At the same time the whole fibre becomes more transparent. (2) Mechanical Phenomena. — The muscular contraction may be graphically recorded by connecting a muscle with a lever which is moved either by its shortening or by its thickening. The lever writes on a blackened surface, which must travel a.t a uniform rate if the form and time-relations of the muscle curve are to be studied, but may be at rest if only the height of the con- traction is to be recorded. The whole arrangement for taking a muscle- tracing is called a myograph (Fig. 287, p. 812). The duration of a ' twitch ' or single contraction (in- cluding the relaxation) of a frog's muscle is usually given as about one-tenth of a second, but it may vary considerably with temperature, fatigue, and other circumstances. It is measured by the vibrations of a tuning-fork written immediately below or above the muscle curve. When the muscle is only slightly weighted, it but very gradually reaches its original length after con- traction, a period of rapid relaxation being followed by a period of ' resi- dual contraction,' during which the descent of the lever towards the base-line becomes slower and slower, or stops altogether some distance above it. The duration of the con- traction of smooth muscle evoked by a single momentary stimulus is much greater than that of striped muscle (two to seven seconds for the rabbit's ureter; five to fifteen seconds for the cat's nictitating mem- brane; one to two minutes for the frog's stomach). Latent Period. — If the time of stimulation is marked on the tracing, it is found that the contraction does not begin simultaneously with it, but only after a certain interval, which is called the latent period. This can be measured by means of the spring myograph (Fig. 251) or of the pendulum myograph, a pendulum which in its swing carries a smoked plate against the writing-point of a lever connected with a muscle. The carrier of the recording plate opens, at a definite point in its passage, a key in the primary coil of an induction machine, and so causes a shock to be sent through the muscle or nerve, which is con- nected with the secondary. The precise point at which the stimulus is thrown in can be marked on the tracing by carefully bringing the Fig. 250. — Living Muscular Fibre (from Geolrupes stercorarius). i, in or- dinary; 2, in polarized light. (Van Gehuchten.) In living muscle (at least in fibres which are not extended) in contrast to dead muscle after treat- ment with reagents, the doubly re- fracting or anisotropous substance is present in the greater part of the fibre ; and with crossed nicols the position of the singly refracting or isotropous material is indicated only by narrow transverse black lines or rows of dark dots. 745 plate to the position in which the key is just opened, and allowing the lever to trace here a vertical line (or, rather, an arc of a circle). The portion of the time-tracing between this line and a parallel line drawn through the point at which the contraction begins gives the latent period. Helmholtz measured the length of the latent period by means of the principle of Pouillet, that the deflection of a magnet by a current of given strength and of very short duration is proportional to the time during which the current acts on the magnet. He arranged that at the moment of stimulation of the muscle a current should be sent through a galvanometer, and should be broken by the contraction of the muscle the moment it began. In this way he obtained the value of Y^JJ second for the latent period of frog's muscle. The tendency of ILL Fig. 251. — Spring Myograph. A, B, iron uprights, between which are stretched the guide-wires on which the travelling plate a runs; k, pieces of cork on the guides to gradually check the plate at the end of its excursion, and prevent jarring; b, spring, the release of which shoots the plate along; A, trigger-key, which is opened by the pin d on the frame of the plate. later observations has been to make the latent period shorter. Burdon Sanderson found that the change of form begins in unweighted or very slightly weighted muscle with direct stimulation in y^y second after, and the electrical change (p. 824) simultaneously with, the excitation. It is known that the apparent latent period depends upon the resistance which the muscle has to overcome in beginning its contraction. The maximum shortening, or ' height of the lift,' depends upon the length of the muscle, the direction of the fibres, the strength of the stimulus, the excitability of the tissue, and the load it has to raise. In a long muscle, other things being equal, the absolute shortening, and therefore the maximum height of the curve, will be greater than in a short muscle; in a muscle with fibres parallel to its length — the sartorius, for instance — it will be greater than in a muscle like the gastrocnemius, with the fibres directed at various angles to the long axis. For stimuli less than maximal, the absolute contraction increases MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 7*7 with the strength of stimulation, and a given stimulus will cause a greater contraction in a muscle with a given excitability than in a muscle which is less excitable. Under ordinary experimental condi- tions ,vt least, weak stimuli cause a smaller contraction than strong, not only because each stimulated fibre contracts less, but because a smaller number of fibres are excited (p. 155). The objects used for the study of muscular contraction contain many fibres, and it is not in Fig. 252. — Curve of a Single Muscular Contraction or Twitch taken on Smoked Glass with Spring Myograph and photographed. Vertical Jine A rr.arks the point at which the muscle was stimulated; time tracing shows t$-(7 of a second (reduced). general possible to distribute the stimulus equally to all. This is true for smooth muscle as well as for striped. Finally, increase of the load per unit of cross-section of the muscle diminishes above a certain limit the ' height of the lift.' Influences which affect the Time-Relations of the Muscular Contrac- tion.— Many circumstances affect the form of the muscle curve and its time-relations . (a) Influence of the Load — Isotonic and Isometric Contraction. — The first effect of contraction is to suddenly stretch the muscle, and the more the muscle is loaded the greater will this elongation be. So that at the beginning of the actual shortening part of the energy of contraction is already expended without visible effect, and has to be recovered from the elastic reaction during the ascent of the lever. The contraction of a muscle loaded by a weight which is not increased or diminished during the contraction is said to be isotonic, for here the tension of the muscle is the same throughout, and its length alters . When the muscle is attached very near the fulcrum of the lever, so that it acts upon a short arm, while the long arm carrying the writing-point is prevented from moving much by a spring, the muscle can only shorten itself very slightly; but the changes of tension in it will be related to those in the spring, and therefore to the curve traced by the writing-point. Such a curve is called isometric, since the length of the muscle remains almost unaltered. In the body muscles usually contract under conditions more nearly allied to those of the isometric than to those of the isotonic contraction. The work done by a muscle in raising a weight is equal to the product of the weight by the height to which it is raised. Beginning with no load at all, it is found that the weight can be increased up to a certain limit without diminishing the height of the contraction; perhaps the height may even increase. Up to this limit, then, the work evidently increases with the load. If the weight is made still greater, the con- Fig. 253. — Contrac- tions ol Smooth Mus- cle: Cat's Bladder (C. C. Stewart). Stimulated with pro- gressively stronger induction shocks. The lowest line is the time trace (lo-second intervals). Immedi- ately below the mus- cular contractions are marked the points at which the stimuli were thrown in. 748 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES traction becomes less and less, but up to another limit the increase oi weight more than compensates for the diminution of ' lift,' and the work still increases. Beyond this, further increase of weight can no Fig. 254. — Influence of Load on the Form of the Muscle Curve, i. curve taken with unloaded lever; 2, 3, 4, weight successively increased; 5. abscissa line: time trace, •jfaf second (reduced). longer make up for the lessening of the lift, and the work falls off till ultimately the muscle is unable to raise the weight at all. The ' absolute contractile force ' of an active muscle may be measured by determining the weight which, brought to bear upon the muscle at Fig. 255. — Influence of Temperature on the Striated Muscle Curve. 2, air tempera- ture; i. 25° — 30° C. ; 3, 7° — 10° C. ; 4, ice in contact with muscle. The fifth curve was taken at a little above air temperature. the instant of contraction, is just able to prevent shortening without stretching the muscle. It, of course, depends, among other things, on the cross-section of the muscle. During the contraction the absolute force diminishes continually, so that a smaller and smaller weight is MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 749 sufficient to stop any further contraction the more the muscle has already shortened before it is applied. At the maximum of the con- traction the absolute force is zero. Hence a muscle works under the most favourable conditions when the weight decreases as it is raised, and this is the case with many of the muscles of the body. During flexure of the forearm on the elbow, with the upper arm horizontal, a weight in the hand is felt less and less as it is raised, since its moment, which is proportional to its distance from a vertical line drawn through the lower end of the humerus, continually diminishes. (b) Influence of Temperature on the Muscular Contraction. — Increase of temperature of the muscle up to a certain limit diminishes the latent period and the length of the curve, and increases the height of the contraction, but beyond this limit the contractions are lessened in height (Fig. 255). Marked diminution of temperature causes, in general, an increase in the latent period and length, and a decrease in the height of the contraction. In the heart the effect of cold in strengthening the beat is often very marked. Temperature affects the contraction curve of smooth muscle much in the same way as that of striated muscle (Fig. 256). (c) Influence of Previous Stimulation — Fatigue. —If a muscle is stimulated by a series of equal shocks thrown in at regular intervals, and the contractions recorded, it is seen that at first each curve overtops its prede- cessor by a small amount.* This phenomenon, which is regularly _ observed in fresh skeletal muscle Fig. 256. — Influence of Temperature on the Smooth Muscle Curve: Cat's Bladder (C. C. Stewart). Contractions at different temperatures with the same strength of stimulus. The temperatures (Centigrade) are marked on the curves. (Fig. 260), although it was at one time supposed to be peculiarly a property of the muscle of the heart (Fig. 261), is called the staircase,' and seems to indicate that within limits the muscle is benefited by contraction and its excitability increased for a new stimulus. Soon, however, in an isolated preparation, the contractions begin to decline in height, till the muscle is at length utterly exhausted, and reacts no longer to even the strongest stimulation (Figs. 258, 259, 288). A conspicuous feature of the contraction-curves of fatigued muscle is the progressive lengthening, which is much more marked in the descending than in the ascending periods; in other words, * Guthrie has recently described an interesting series of phenomena illus- trating the influence of previous stimulation on the irritability of muscle. Under given conditions, for example, strong stimulation increases the response of skeletal muscles to subsequent single stimuli, it may be, by hundreds per cent. 75o THE PHYSIOLOGY OF THE CONTRACTILE TISSUES relaxation becomes more and more difficult and imperfect (Fig. 288, p. 814). In smooth muscle (cat's bladder or ring from frog's stomach) fatigue can be very easily demon- strated in the same way, and the curves present similar features, with the exception that, instead of be- coming longer in fatigue, the suc- cessive contractions become shorter. It is by no means so easy to fatigue a muscle still in connection with the circulation as an isolated muscle. But even the latter, if left to itself, will to some extent re- cover, and be again able to con- tract, although exhaustion is now more readily induced than at first. Fig. 258. — Fatigue Curve of Muscle-. Frog's Gastrocnemius. The arrange- ment with which the curve figured was obtained was a so-called auto- matic muscle interrupter (Fig. 257). A wire on the lever is made to close and open the primary circuit of an inductorium, the muscle or nerve being connected with the secondary. Every time the needle touches the mercury the muscle is stimulated automatically. A' Fig. 257. — Automatic Muscle Interrupter. K. battery; P, primary; S, secondary coil; A, axis of lever; N. needle; Hg, mercury cup. In man, muscular fatigue can be studied by means of an arrange- ment called an ergograph (Fig. 262). A record of successive con- t-'ig. 259. — Fatigue Curve taken on a Slowly-moving Drum (reduced to Half): Frog's Gastrccnemius. Excited through the sciatic nerve by maximal shocks once i« six second*. MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 751 •tractions, say, of one of the flexor muscles of a finger, in raising a weight (isotonic method) or in deforming a spring (isometric method) is taken on a drum. When the contractions are repeated every second, or every half-second, distinct evidence of fatigue is seen on the tracing after a longer or shorter period, according to the conditions.* What is the cause of muscular fatigue ? An exact answer is not possible in the present state of our knowledge, but we may fairly con- clude that in an isolated preparation it is twofold : (i) Waste products, among which some are so directly related to the onset of fatigue as to deserve the name of ' fatigue sub- stances,' are formed by the active muscle faster than they can be re- moved, oxidized or otherwise decom- posed. (2) The material necessary for contraction is used up more quickly than it can be reproduced or brought to the place where it is required. That the accumulation of fatigue products has something to do with the exhaustion is shown by the fact that the muscles of a frog, exhausted in spite of the continuance of the circulation, can be restored by bleeding the animal, or washing out the vessels with physiological salt solution, while injection of a watery extract of exhausted muscle into the bloodvessels of a Fig. 260. — ' Staircase ' in Skeletal Muscle : Frog. Stimulation by an automatic arrangement. Fig. 26?. — 'Staircase' in Cardiac Muscle. Contractions recorded on a much more 4iuilfely moving drum than in Fig. 260. The contractions were caused by stimu- 'lati^j a heart reduced to standstill by the first Stannius' ligature (p. 199). The ..contractions gradually increase in height. curarized muscle renders it less excitable (Ranke). This observer supposed that it was specially the removal of the acid products of contraction which restored the muscle. Such acid products as carbon dioxide and lactic acid, or the lactates which it may form with bases in the blood, lymph or tissues, when they act on muscle * Recent observations (by Ryan and Agnew) with improved methods have emphasized the necessity for care in the interpretation of ergographic records. This caution is timely, inasmuch as in modern war the question of the rela- tion of fatigue to industrial practice acquires first-class importance. 752 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES in more than a certain concentration, produce the same effects on its power of contraction as are produced by fatigue, and there is some reason to suppose that lactic acid is the most influential of the fatigue substances. In smaller concentration, on the contrary, they increase the excitability of the muscle, and, according to Lee, the phenomenon of the ' staircase ' is due to the augmenting action of these, and perhaps other fatigue substances, before they have accu- mulated sufficiently to cause fatigue. The lack of oxygen holds a conspicuous place among the con- ditions which permit an excessive accumulation of fatigue substances, and may contribute also to the failure of the processes normally going on in the muscle which replenish the store of materials needed for contraction. An isolated muscle is necessarily an asphyxiated muscle, and the favourable action of an atmosphere of oxygen on restoration of its contractile power after exhaustion Fig. 262. — Ergograph (Mosso's, modified by Lombard). (Fig. 123, p. 269) shows that asphyxia is itself an important*ilctor in the onset of fatigue. Injection of arterial blood, or even of an oxidizing agent like potassium permanganate, into the* vessels of an exhausted muscle causes restoration (Kroneckeif. The depletion of the available store of carbo-hydrate in the form of glycogen (and dextrose) seems to be another factor in fatigue, although not the chief direct cause of the phenomena associated with that condition. Seat of Exhaustion in Fatigue. — When a fatigued muscle responds no longer to indirect stimulation, it can still be directly excited. The seat of exhaustion must therefore be either the nerve-trunk or the nerve-endings. It is not the nerve-trunk which is first fatigued, for this still shows the negative variation (p. 824) on being excited. And if the two sciatic nerves of a frog or rabbit be stimu- MECHANICAL PHENOMENA OF MVSCULAk CONTRACTION 75$ lated continuously with interrupted currents of equal strength, while the excitation is prevented from reaching the muscles of one limb till those of the other cease to contract, it will be found that when the ' block ' is removed the corresponding muscles contract vigorously on stimulation of their nerve. The passage of a constant current through a portion of the nerve or the application of ether between the point of stimulation and the muscles may be used to prevent the excitation from passing down (p. 813). Or a dose of curara just sufficient to paralyze the motor inner vation may be given to a rabbit, and the animal kept alive by artificial respiration. The sciatic is now stimulated for many hours. As soon as the influence of the curara begins to wear off, the muscles of the leg contract. The possible seats of fatigue caused by voluntary muscular con- traction are (i) the muscle; (2) the nerve-endings (or the receptive substances in the muscles, p. 739) ; (3) the nerve-trunk ; and (4) the path of the voluntary motor impulses in the central nervous system, which includes the pyramidal cells in the motor region of the cerebral cortex (p. 875), the fibres of the pyramidal tract, and the motor cells in the anterior horn of the spinal cord. The two weak links in this chain appear to be the motor nerve- /• endings and the muscles. The nerve-fibres, whether peripheral or central, are certainly the strongest link. Ergographic experi- ments have hitherto yielded results too discordant to justify any very definite statement as to the point at which the chain snaps in complete fatigue, if, indeed, it al vays necessarily breaks at the same point. The muscles and motor en.iings appear to be always affected. The position of the nerve centres, including the synapses (p. 852), is in doubt. That the synapses easily lose their power of con- ducting nerve impulses under the influence of repeated excitations is indicated by the experiments of Sherrington on fatigue of reflex mechanisms in which two or more afferent paths can cause discharge along a common efferent path (p. 902). When excitation of one of tke afferent paths has ceased to be effective, the reflex contrac- tions can still be obtained on exciting the other. In this case the molvor aieuron from cell-body to nerve-ending and the muscle are eliminated as the seats of the fatigue block. Whether the tem- porary loss of conduction in this case is comparable to the fatigue of muscle, or is a perfectly different phenomenon (' pseudo-fatigue ' of Lee), scarcely bears on our present question. For if ' pseudo- fatigue ' of afferent synapses can cause a reflex to miss fire, this at least shows that the conductivity of the synapse is very easily affected by repeated excitation, just as it is known to be very easily affected by anaemia. The fact that a muscle, completely fatigued by direct electrical stimulation, can still be voluntarily contracted, has been supposed to indicate that the voluntary excitation is more effective t54 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES than any artificial stimulus. But the alternative explanation that the electrical stimuli cannot be applied to a muscle in situ, so as to cause uniform excitation, and therefore uniform fatigue, of all the fibres of the muscle, is more probable (Hough). It has been shown that the injection of the blood of an animal exhausted by running or other muscular effort into the circulation of a normal animal produces in the latter all the symptoms of fatigue. Here the fatigue-producing substances will have the opportunity of acting on both the central and the peripheral mechanisms. There are reasons for believing that the fatigue process is fundamentally the same in different tissues. The fatigue substances produced in Fig. 253. — Influence of Mental Fatigue on Muscular Contraction, i, series of con- tractions of flexors of middle finger before, and 2, series of contractions imme- diately after, a period of three and a half hours' hard mental work. In both cases the muscles were stimulated directly every two seconds by an electrical current, and caused to raise a certain weight till temporary exhaustion occurred. In the first series fifty-three contractions were found possible, in the second only twelve (Maggiora). muscle, and not immediately eliminated or transformed during active muscular exertion, may therefore very well be a factor in inducing fatigue of the central nervous mechanisms in addition to the formation of fatigue products, and the using up of necessary material in these mechanisms themselves. Conversely, active and long-continued mental exertion may occasion muscular fatigue (Fig. 263). The sensation of fatigue is alluded to in Chapter XVIII. (d) The Influence of Drugs on the Contraction of Muscle. — The total work which a muscle can perform, its excitability and the absolute force of the contraction, may all be altered either in the plus or the minus sense by drugs. But in connection with our present subject those drugs which conspicuously alter the form and time-relations of the muscle-curve have most interest. Of these veratrine is especially MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 755 important. When a small quantity of this substance is injected below the skin of a frog, spasms of the voluntary muscles, well marked in the limbs, come on in a few minutes. These are attended with great stiffness of movement, for while the animal can contract the extensor muscles of its legs so as to make a spring, they relax very slowly, and some time elapses before it can spring again. If it be killed before the reflexes are completely gone, the peculiar alterations in the form of the muscle-curve caused by veratrine will be most marked. The poisoned 1 wwwwwwww mw/mwm Fig. 264. — Veratrine Curve compared with Normal: Frog's Gastrocnemius. The tuning-fork marks hundredths of a second. Between i and 2 a portion of the tracing corresponding to one and a half seconds has been cut out, and between 2 and 3 a portion corresponding to one second. The veratrine curve does not show a peak. At 3 it has not yet fallen to the base-line. muscle, stimulated directly or through its nerve, contracts as rapidly as a normal muscle, while the height of the curve is about the same, but the relaxation is enormously prolonged (Fig. 264). This effect seems to be to a considerable degree dependent on temperature, and it may temporarily disappear when the muscle is made to contract several times without pause. Barium salts, and, in a less degree, those of strontium and calcium, have an action on muscle similar to that of veratrine. Sometimes the curve shows a peak (Fig. 265), due to a rapid descent of the lever for a certain distance. This is followed by a slow relaxation. The peak appears to be analo- gous to the initial con- traction when a strong voltaic current is passed through a muscle, and the rest of the curve to the tonic contraction. (e) The individuality of Fig 265.— Veratrine Curve: Frog's Gastrocnemius. the muscle itself has an The curve shows a peak, the lever falling a littk influence on the muscle- before the sustained contraction begins. curve. Not only do the muscles of different animals vary in the rapidity of contraction, but there are also differences between the skeletal muscles of the same animal. In the rabbit there are two kinds of striped muscle, the red and the pale (the semitendinosus is a red, and the adductor magnus a pale muscle), and the contraction of the former is markedly slower than that of the latter. In many fishes and birds, and in some insects, a similar difference of colour and structure is present. Even where there is no distinct histological difference, there may be great variations in the length of contraction. In the frog, for instance, the hyoglossus muscle contracts much more slowly than the gastroc- 756 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES nemius. The wave of contraction, which in frogs' striped muscle lasts only about 0-07 second at any point, may last a second in- the forceps muscle of the crayfish, though only half as long in the muscles of the tail. In the muscles of the tortoise the contraction is also very slow. The muscles of the arm of man contract more quickly than those of the leg. Summation of Stimuli and Superposition of Contractions. — Hitherto we have considered a single muscular contraction as arising from a single stimulus, and we have assumed that the muscle has completed its curve and come back to its original length before the next stimulus was thrown in. We have now to inquire what happens when a second stimulus acts upon the muscle during the contraction caused by a first stimulus, or during the latent period before the contraction has actually begun; and what happens when a whole series of rapidly-succeeding stimuli are thrown into the muscle. First let us take two stimuli separated by a smaller interval than the latent period (p. 745). If they are both maximal — i.e., if each by itself would produce the greatest amount of contraction of which the muscle is capable when excited by a single stimulus — the second has no effect whatever; the contraction is precisely the same as if it had never acted. But if they are less than maximal, the contraction, although it is a single contraction, is greater than would have been due to the first stimulus alone; in other words, the stimuli have been summed or added to each other during the latent period, so as to produce a single result. Next let us consider the case of two stimuli separated by a greater interval than the latent period, so Fig. 266 -Superposition of Contractions. that the second falls lnto the i is the curve when only one stimu us j during the contraction pro- is thrown in; 2, when a second stimulus ,, 9- ™ . acts at the time when curve i has nearly duced by the first The result here reached its maximum height. 1S verY different : traces of two con- tractions appear upon the muscle- curve, the second curve being that which the second stimulus would have caused alone, but rising from the point which the first had reached at the moment of the second shock (Fig. 266). Although the first curve is cut short in this manner, the total height of the contraction is greater than it would have been had only the first stimulus acted ; and this is true even when both stimuli are maximal. Under favourable circum- stances, when the second curve rises from the apex of the first, the total height may be twice as great as that of the contraction which one stimulus would have caused (p. 816). It is worthy of note that striated muscle has no power of summation of subminimal stimuli each of which is just too weak to cause contraction. No matter how rapidly they are thrown in, the muscle remains at rest. It is otherwise with smooth muscle. Stimuli which are singly ineffective cause contraction when repeated. Tetanus. — Not only may we have superposition or fusion of two contractions, but of an indefinite number; and a series of rapidly following stimuli causes complete tetanus of the muscle, which remains contracted during the stimulation, or till it is exhausted (Fig: 267). MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 757 The meaning of a complete tetanus is readily grasped if, beginning with a series of shocks of such rapidity that the muscle can just completely relax in the intervals between successive stimuli, we gradually increase the frequency (p. 816). As this is done, the ripples on the curve become smaller and smaller, and at last fade out altogether. The maximum height of the contraction is greater than that produced by the strongest single stimulus ; and even after complete fusion has been attained, a further increase of the fre- quency of stimulation may cause the curve still to rise. Fig. 267. — Analysis of Electrical Tetanus (reduced to f ). Four curves showing the effect of increasing frequency of stimulation of the frog's gastrocnemius through its nerve. In the lowest curve the frequency is such that the muscle relaxes almost completely between the successive contractions. In the uppermost curve, with a frequency more than three times greater, the contractions are almost completely fused. In all the curves the fusion becomes more nearly complete as stimulation goes on, owing to the slower relaxation of the fatigued muscle. It is evident from what has been said that the frequency of stimulation necessary for complete tetanus will depend upon the rapidity with which the muscle relaxes; and everything which diminishes this rapidity will lessen the necessary frequency of stimulation. A fatigued muscle may be tetanized by a smaller number of stimuli per second than a fresh muscle, and a cooled by a smaller number than a heated muscle. The striped muscles of insects, which can contract a million^ times in an hour, require 300 stimuli per second for complete tetanus, those of birds 100, of man 40, the torpid muscles of the tortoise only 3. The pale muscles of the rabbit need 20 to 40 excitations a second, the red 758 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES muscles only 10 to 20; the tail muscles of the crayfish 40, but the muscles of the claw only 6 in winter and 20 in summer. The gastrocnemius of the frog requires 30 stimuli a second, the hyo- glossus muscle only half that number (Richet). The frequency of stimulation necessary for complete tetanus of unstriped muscle is much less than for striped muscle. Smooth tetanus of a band of muscle from the frog's stomach was obtained with strong opening induction shocks at the rate of i in 5 seconds. There appears also to be an upper limit beyond which a series of stimuli becomes too rapid to produce complete tetanus, and at which an interrupted current acts like a constant current, causing a single twitch at its commencement or at its end, but no contraction during its pas- sage. This limit does not depend upon the frequency of stimulation alone; the intensity of the individual excitations, the temperature of the muscle, and probably other factors, affect it. For Bernstein found that with moderate strength of stimulus tetanus failed at about 250 per second, and was replaced by an initial contraction; with strong stimuli at more than 1,700 per second, tetanus could still be obtained. Kronecker and Stirling saw tetanus even with 4,000 shocks a second. Kries in a cooled muscle found tetanus replaced by the simple initial twitch at 100 stimuli per second, although in a muscle at 38° C. stimu- lation of ten times this frequency still caused tetanus. Einthoven, exciting the nerve of a frog's nerve-muscle preparation with extremely frequent oscillatory condenser discharges, observed tetanus up to even a million vibrations a second, if the current intensity was at the same time very greatly increased (to more than 16,000 times the intensity needed with a constant current). These results are not really so dis- cordant as they appear ; for it is known that with electrical stimulation the number of excitations is not necessarily the same as the nominal number of shocks. By applying a telephone to a muscle excited through its motor nerve, it has been shown that the pitch of the note produced by the tetanized muscle corresponds exactly to the rate of excitation up to a certain frequency. This frequency is about 200 per second for frog's and about 1,000 per second for mammalian muscle under the best conditions. If the rate of excitation is still further increased, there is no corresponding increase in the pitch. Therefore, some of the stimuli are now producing no effect — ' falling flat,' so to speak (Weden- sky). A physical reason for this is the overlapping of the make and break shocks (Erlanger and Garrey) ; and a physiological reason, the alterations of conductivity and excitability, which even very brief currents leave behind them (Sewall), and which we shall have to discuss in another chapter. It is only while the actual shortening is taking place that a tetanized muscle can do external work. But, although during the maintenance of the contraction no work is done, energy is nevertheless being ex- pended for the metabolism of a muscle during tetanus is greater than during rest, and, among other changes, lactic acid is produced. There are great differences in the ease with which different muscles can be exhausted by tetanus. For example, the muscles which close the forceps of the crayfish or lobster have, as everyone knows, the power of most obstinate contraction. Richet tetanized one for over seventy minutes, and another for an hour and a half, before exhaustion came on, while a tetanus of a single minute exhausted the muscles of tLe crayfish's tail. The gastrocnemius of a summer frog kept up for twelve minutes, and a tortoise muscle for forty minutes. Continuous stimulation is not always necessary for the production of continuous contraction; in some conditions a single stimulus is suffi- cient. A blow with a hard instrument may cause a dying or exhausted, and in thin persons even a fairly normal, muscle to pass into long- continued contraction. This so-called ' idio-muscular ' contraction seems to depend, in part at least, on the great intensity of the stimulus. It can sometimes be obtained in the frog's gastrocnemius, particularly in spring after the winter fast. It is not a tetanus and is not propa- gated along the muscular fibres, as an electrical tetanus is, but remains localized at the spot where it arises. Similar non-tetanic contractions have already been mentioned, such as the tonic contraction during the passage of a strong voltaic current and the sustained veratrine contrac- tion. Ammonia causes also a long but non-tetanic contraction, and this, too, does not spread when the substance has acted only on a portion of the muscle. The contraction force of all these tonic contractions, as measured by the resistance necessary to overcome or prevent them, is less than the contraction force in electrical tetanus (Schenck). The rate at which the wave of muscular contraction travels may be measured by stimulating the muscle at one end, and recording, by means of levers, the movements of two points of its surface as far apart from each other as possible. Time is marked on the tracing by means of a tuning-fork, and the distance between the points at which the two curves begin to rise from the base-line divided by the time gives ' the velocity of the wave. Another method is founded upon the measure- ment of the rate at which the negative variation (p, 824) passes over the muscle, this being the same as the velocity of the contraction-wave. In frog's muscle it is about three metres a second, or six miles an hour. Rise of temperature increases, fall of temperature lessens it. When a muscle is excited through its nerve, the contraction springs up first of all about the middle of each muscular fibre where the nerve-fibre enters it, and then sweeps out in both directions towards the ends. But so long is the wave, that all parts of the fibre are at the same time involved in some phase or other of the contraction. The wave of contraction in unstriped muscle lasts a relatively long time at any given point, and in tubes like the intestines and ureters, the walls of which are largely composed of smooth muscle arranged in rings, the wave shows itself as a gradually-advancing constriction travelling from end to end of the organ. There is no evidence that the contraction of smooth muscular fibres is discontinuous — that is, :omposed of summated contractions like a tetanus ; it appears to be a greatly-prolonged simple contraction. An artificial stimulus, mechani- cal or electrical, causes, after a long latent period, a very definitely- localized contraction in a rabbit's ureter, which slowly spreads in a peristaltic wave in one or both directions along the muscular tube. Here, as in the cardiac muscle, the excitation passes from fibre to fibre, while in striped skeletal muscle only the fibres excited directly or through their nerves contract. That the rhythmical contraction of the heart is not a tetanus has already been seen. It is a simple con- traction, intermediate in its duration and other characters between the twitch of voluntary muscle and the contraction of smooth muscle. The contraction both of unstriped and of cardiac muscle is lengthened and made stronger by distension of the viscera in whose walls they occur, just as a skeletal muscle contracts more powerfully against resistance. 760 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Voluntary Contraction. — There is evidence that the voluntary contraction is a tetanus. One of the strongest buttresses of the theory of natural tetanus has been the muscle-sound, a low rumbling note which can be heard by listening with a stethoscope over the contracting biceps, or, when all is still, by stopping the ears with the fingers and strongly contracting the masseter and the other muscles concerned in closing the jaws.* Discovered about ninety years ago, first by Wollaston and then by Erman, half a century passed away before it was investigated more fully by Helmholtz. The latter observer, confirming the results of his predecessors, put down the pitch of the sound at 36 to 40 vibrations per second. He found, however, that little vibrating reeds with a rate of oscillation of about 19-5 per second were more affected when attached to muscle thrown into voluntary contraction, than those that vibrated at a smaller or a greater rate. He therefore concluded that the fundamental tone of the muscle corresponded to this frequency, although, since such a low note is not easily appreciated, the sound actually heard was really its octave or first harmonic (p. 310). The objection has been brought forward that the resonance tone of the ear also corresponds to a vibration frequency of 36 to 40 a second. In other words, this is the natural rate of swing of the elastic struc- tures in the middle ear, the rate they will most easily fall into if set moving by an irregular mixture of faint, low-pitched tones and noises, and not compelled to vibrate at some other rate by a distinct sound of definite pitch. Now, this resonance tone might be elicited by a quivering muscle if, among many diverse rates of oscillation of different portions of its substance, the rate of 36 to 40 a second anywhere ap- peared, and the note corresponding to the real rate of vibration of the muscle as a whole might be overpowered. Or, even if there were no regular rate of vibration of the whole muscle, but, instead, a series of irregular tremors or pulls due to irregularities in the contraction, con- nected with a want of co-ordination of all the fibres (Haycraft), the ear might from time to time pick out of the turmoil of feeble aerial waves those corresponding to its resonance tone, just as a tuning-fork or a piano-string attuned to a particular note would catch it up amid a thousand other sounds and strengthen it. But while this renders it highly probable that the resonance of the ear contributes to the production of the muscle-sound, and shows that we cannot from the pitch of the muscle-sound alone deduce the rate at which the muscle-substance is vibrating, it does not invalidate Helmholtz's objective observations with the oscillating reeds. And several observers (Schafer, Horsley, v. Kries) have noticed periodic oscillations, at the rate of 10 or 12 per second, in the curves taken from muscles (Fig. 268), contracted voluntarily against a small resistance. When the resistance is greater, the rate may be as much as 18 or 20 a second, and in quick and rapidly * In order that a muscular sound may be produced there must be a certain abruptness in the contraction. Thus, the slowly-contracting smooth muscles do not produce a sound, nor the slowly-contracting heart-muscle of cold- blooded animals. MECHANICAL PHENOMENA OF MUSCULAR CONTRACTION 761 repeated movements of the fingers even 40 a second. In habitual movements, such as those employed by a man in his trade, the tremors are much less coarse than in unaccustomed movements. For this reason the tremors of the left hand are greater than those of the right in executing a movement usually made with the latter (Eshner). In disease these tremors are often increased — e.g., in the clonic convulsions of epilepsy — tout the frequency is the same. Fig. 268. — Vibrations of Contracted Arm Muscles (Griffiths). The arm was stretched out, holding a weight of about 6 kilos. Similar vibrations, and at about the same rate, are seen in curves traced by muscles excited through stimulation of the motor areas of the surface of the brain. Since this rate remains the same whether the motor cortex, the corona radiata, or the spinal cord is excited, and, unlike the rate of response to excitation of peripheral nerves, is independent of the frequency of stimulation (so long as the rate of stimulation is greater than 10 or 12 a second), it has been supposed to represent the / rhythm with which impulses are dis- charged from the motor cells of the cord (Fig. 269). It is prob- able that the cortical centres discharge at about the same rate, for not only is it im- possible to articulate more rapidly than eleven syllables per second, but it is impossible to reproduce the act of articulation in thought at a greater rate than this (Richet). But while this rate of 10 or 12 a second does seem to represent a fundamental rhythm of the central discharge, there are facts which indicate that upon this relatively slow rhythm a quicker rhythm is superposed. In other words, each of these discharges is itself discontinuous, and made up of a number of separate impulses. Fig. 269.- -Contractions caused by Stimulation of the Spinal Cord. 762 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Thus, according to Piper, the total number of simple discharges, each associated with an electrical change in the muscle, as recorded by the string galvanometer, is 47 to 50 a second. The rhythm of strych- nine tetanus in the frog is about 8 to 12 per second. By means of the capillary electrometer (p. 729) large electrical oscillations at this rate can be demonstrated, each of which represents a short tetanic spasm, as is shown by the fact that a number of smaller electrical oscillations are superposed upon the large ones (Sanderson). The electrical changes suggest that each discharge causes a simple contraction much more prolonged than the twitch of a directly stimulated muscle. This removes the difficulty of understanding how such a small number as 10 contractions per second could be smoothly fused, and indicates that even the shortest possible voluntary movement, which can be executed in i"& to -£$ of a second, is not caused by a single impulse, but is a tetanus. For these brief movements the frequency of oscillation, as shown by the action currents, is the same as for sustained contractions. The electrical changes in the voluntarily contracted muscle seem to differ in amplitude or abruptness from those produced in experimental tetanus. For secondary tetanus (p. 833) is not caused by muscle in voluntary contraction. But this is also the case with the other pro- longed contractions caused by continuous artificial stimulation — e.g., Ritter's tetanus (p. 741) and the contraction produced by sodium chloride or ammonia. We need not hesitate to conclude, then, that the voluntary contraction is discontinuous, in the sense that it is not a perfectly smooth and uniform tonic contraction, although we still lack a decisive proof that it is maintained by a strictly intermittent outflow of nervous energy, and not by a continuous outflow causing a sustained contraction, which, it may be, remits and is reinforced at intervals. The apparent discrepancies as to the rate of discharge in the results obtained by different observers, and by different methods, far from exciting distrust of them all, really lend support to the idea of a fundamental and fairly constant rhythm in the outflow as soon as it is recognized that the higher rates are approximately multiples of the lower. Thus, the number deduced by Helmholtz from the ex- periment of the springs is twice the lowest rate calculated from graphic records of the contraction. The rates corresponding to the muscle- sound and to the frequency of the electrical oscillations are about four times this number. Now, in a vibrating elastic body like a contracting muscle, a simple mathematical relation of this sort might be expected to appear when determinations of the rate of oscillation and of accom- panying periodic changes are made by methods varying in principle and in delicacy. For instance, an arrangement suited to record and to count coarse vibrations could not be expected to give the same result as an arrangement suited to record and count fine vibrations. But if both the coarse and the fine vibrations were related to a fundamental rhythm, a simple proportion might be expected to exist between the two sets of results. (3) Thermal Phenomena and Transformation of Energy in the Muscular Contraction. — When a muscle contracts, its temperature rises ; the production of heat ,jn it is increased. This is most dis- tinct when the muscle is tetanized, but has also been proved for single contractions. The change of temperature can be detected by a delicate mercury or air thermometer; and, indeed, a ther- mometer thrust among the thigh-muscles of a dog may rise as much THERMAL PHENOMENA OF MUSCULAR CONTRACTION 763 as i° to 2° C. when the muscles are thrown into tetanus. In the isolated muscles of cold-blooded animals the increase of tempera- ture is much less ; and thermo-electrical methods, which are the most delicate at present known, have generally been used for its detection and measurement. They depend upon the fundamental fact of thermo-electricity, that in a circuit composed of two metals a current is set up if the junctions of the metals are at different tempera- tures. Where no very fine differences of temperature are to be measured, a single thermo-j unction of German silver and iron, or copper and iron, is inserted into a muscle or between two muscles. But the electromotive force, and therefore the strength of the thermo-electric cur- rent, is proportional for any given pair of metals to the number of junctions, and for delicate measurements it may be necessary to use several connected together in series. A thermopile of antimony - bismuth junctions gives a stronger current for a given difference of temperature than the same number of German silver-iron couples, but from its brittle nature is otherwise less convenient . The direction of the current in the cir- cuit is such that it passes through the heated junction from bismuth to anti- mony and from copper or German silver to iron. Knowing this direction, we are aware of the changes of temperature which take place from the movements of the mirror of the galvanometer with which the pile is connected. In the thermopiles employed in the recent ex- tensive investigations of Hill the alloy constantan is coupled with iron, the electromotive force of this combination being exceptionally great. The muscle which is to be excited is brought into close contact with one junction or set of junctions, the other set being kept at constant temperature. The image will now come to rest on the scale; and excitation of the muscle will cause a movement indicating an increase of temperature in it, the amount of which can be calculated from the deflection. In one form (Fig. 270) the thermopile con- stitutes a hollow cone, in which a muscle can be arranged so as to eliminate largely the errors due to differences of temperature of the muscle, or to the " slip " of the contracting muscle over the junctions. In this way Helmholtz observed a rise of temperature of 0-14° to o -18° C. in excised frogs' muscles when tetanized for a couple of minutes . S Fig. 270. — Conical Thermopile containing Gastrocnemius Mus- cle Reversed. C, copper leads to galvanometer; S, stimulating wire. The straight lines indicate iron, the crossed lines constan- tan, the external junctions em- bedded in the ebonite frame being at a, the internal junc- tions, b, in contact with the muscle. 764 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Heidenhain, with a very delicate pile, found a rise of 0-001° to 0-005° C. for a single contraction of a frog's muscle. On the assump- tion that the pile had time to take on the temperature of the muscle before there was any appreciable loss of heat, this would be equal to the production by every gramme of muscle of a thou- sandth to five-thousandths of a gramme-calorie (p. 676) of heat. From pick's observations we may take about three-thousandths of a gramme-calorie as the maximum production of a gramme of frog's muscle in a single contraction. Hill has shown that in the case of the single contraction or twitch the evolution of heat may be so rapid as to be practically instan- taneous, indicating that it depends upon some sudden ' explosive ' chemical reaction; or, on the other hand, under certain conditions it may last as long as two seconds — that is, from four to ten times as long as the contraction itself. In the absence of oxygen, when the muscle is left, for in- stance, in an atmosphere of hydrogen, the heat produc- tion becomes markedly pro- longed. When abundant oxygen is supplied, the dura- tion of the discharge of heat is decreased. In a tetanus the evolution of heat lags behind the excitation, and the discharge associated with each stimulus is not complete till 0-5 to 2-5 seconds after the stimulus. In a prolonged complete tetanus the heat production corresponding to the first tenth of a second of excitation is far greater than that corresponding to the second tenth of a second, and so on, until eventually a uniform dis- charge of heat, at a rate much smaller than the initial rate, is reached. When frogs' muscles are rapidly stimulated indirectly (through the nerves) till fatigue has occurred, the maximum value of the heat evolved approximates to 0-9 gramme - calorie per gramme of muscle, about 70 or 80 per cent, being liberated in the first two minutes (Peters). A fact of great significance in regard to the relation of the reaction upon which the heat production depends and the mechanical conditions in the acti^ muscle is that the production of heat is determined by the length of muscular fibres existing at the time when the heat is being evolved. From this it has been assumed that the production of heat in active muscle is a surface effect, and Fig. 271. — A, a single copper-iron thermo- electric couple ; B,two pairs, one inserted into the tissue b, the other dipping into water in a beaker a. The temperature of the water may be adjusted so that the galvanometer shows no deflection. The temperature of the tissue is then the same as that of the water. THERMAL PHENOMENA OF MUSCULAR CONTRACTION 765 not an effect taking place uniformly throughout the muscle substance and related accordingly to the volume or mass of the muscular substance (Blix). Much evidence has been accumulated in favour of this hypothesis. For example, a muscle contracting isometrically (p. 747) produces more heat the greater is the initial tension (the more it is stretched at the begining of the excitation) — 'that is, the greater its length during contraction (Heidenhain). When a muscle is allowed to shorten in a tetanus, the heat production as compared with that of an isometric contraction of the same dura- tion, and evoked by the same strength of stimulus, is diminished by as much as 40 per cent. Relation between the Development of Mechanical Energy and Heat Production in Active Muscle. — There is no simple relation between the external work done in a muscle twitch and the heat set free. The efficiency of the muscular machine, as estimated by the proportion of the work done to the total energy degraded, varies with a number of factors — e.g., the load, the number of fibres excited, and the intensity of the excitation of each fibre, the two lattjer factors depending upon the strength of the stimulus. The greater the resistance, so long as the muscle can overcome it so as to do its utmost amount of external work,* the larger is the proportion of energy which appears as work, the smaller the proportion which appears as heat. For every muscle, under given conditions, there is a certain load which can be raised more advan- tageously than any other; but even in the most favourable case, an excised frog's muscle never does work equal to more than £ of the heat given off. Generally the ratio is much less, and may sink as low as -£g. In the intact mammalian body the muscles work somewhat more economically than the excised frog's muscles at their best; for both experiment and calculation show (p. 687) that in a normal man under the most favourable conditions as much as 1 of the energy is converted into work. According to Zuntz and Katzenstein, 35 per cent, of the total energy appeared as muscular work in climbing a mountain, and in bicycling only 25 per cent. Movements which have been much practised are more economically performed than unaccustomed ones, and this explains the superior efficiency of the muscles concerned in climbing, for no movements can possibly be more familiar than those con- cerned in locomotion. So far as this indication goes, it would seem that in the treatment of obesity unfamiliar, and therefore physio- logically expensive, forms of exercise should be recommended, in so far, of course, as they do not injuriously react upon the general condition, especially upon the circulation.' * This statement, based on experiments with excised frog's muscles, is not, of course, inconsistent with the fact mentioned on p. 687, that in the intact body the fraction of the energy transformed into heat is greater in hard than in moderate work. 766 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES When a muscle, excited by maximal stimuli, is made to lift con- tinuously increasing weights, both the work done and the heat given out increase up to a certain limit. The muscle, as it were, burns the candle at both ends. The heat-production reaches its maximum some- what sooner than the work. It is certain that when work is done by a muscle an equivalent amount is subtracted from its sum-total of energy, and under proper conditions this can be actually demonstrated by the deficiency in the heat-production. This is done by means of a contrivance called a work-adder. It consists of a wheel, the rotation of which raises a weight attached to a cord wound round its axle. The muscle acts on the periphery of the wheel, and by rotating it raises the weight a little at each contraction. At the end of the contraction the wheel is pre- vented from moving back by a catch. The work done in a series of contractions is calculated from the total height to which the weight has been raised. Suppose a frog's gastrocnemius is made to contract a certain number of times while attached to the work-adder, and that simultaneously the heat- production is measured by means of a thermo- pile. Let H represent the heat actually produced, and h the heat equivalent of the work done. Now let the muscle be disconnected from the adder and made to raise the same weight, directly attached to it, by a series of contractions elicited in precisely the same way as the previous ones, except that the weight is allowed to fall with the muscle when it relaxes after each contraction. Here heat corresponding to the external work disappears from the muscle during the contraction just as in the first experiment, but this heat is returned to the muscle during the relaxation, since on the whole no external work is done. The heat produced in the second experiment is found, as a matter of fact, allowing for unavoidable errors, to be equal to H'+h. According to Hill, the true ' efficiency ' of the muscle is not the ratio W/H, where W is the external work and H the total heat liberated, but T/H where T is the maximum increase of tension set up during the twitch when the muscle is contracting isometrically. This fraction T/H is constant whatever be the initial tension, the number of fibres excited, or the strength of excitation of each fibre. For the theory of the muscular contraction the tension auring an isometric muscle twitch, which represents the potential energy suddenly developed in conse- quence of the excitation, is accordingly much more important than the height of the contraction, which is related to the work actually done. The essential thing in muscular contraction may be the abrupt develop- ment of this tension through a chemical reaction which liberates certain substances at some membrane or surface in the muscle. The potential energy once in being may or may not be transformed into work, and if so transformed the change may be accomplished economically or waste- fully, according to the conditions of the contraction. The ratio T/H decreases in fatigue, and with the time during which the muscle has been deprived £fi its blood-supply. Hill has calculated the absolute value of the hfcat-production in tetanus of a sartorius or semimem- branosus muscle of the frog. This quantity, reckoned per centimetre of length of the muscle, per gramme weight of the tension developed and per second of maintenance of the tension, is relatively constant at about 0-000015 'gramme-calorie. Including the recovery processes of oxidation following the contraction the total heat-production would amount to about 0-000025 calorie. The potential energy possessed by a muscle of a length of a centimetre when maintaining a tension of a gramme is abowt 0-000004 calorie. So that to maintain this state of potential energy six or seven times as much energy must be liberated CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 767 per second. The maintenance of prolonged tension is, therefore, from the point of view of the mechanical result, an exceedingly wasteful process, with a very low efficiency in comparison with the high efficiency in a rapid twitch. This enables us to see how important a part in heat- production, and therefore in temperature regulation, the tonusof muscle and the prolonged contractions of shivering may possess (p. 695). We are as yet in the dark as to the precise relation of the energy which appears as heat and of that which is converted into work. The ultimate source of both is, of course, the oxidation (and cleavage) of the food substances. It was at one time a favourite theory that in a muscle, as in a heat-engine, the chemical energy is first converted into heat, and part of the heat then transformed into work. There is no evidence that this is the case. It is, indeed, impossible that such differences of temperature can exist as would be compatible with the known efficiency of the muscular machine. Hypotheses based on the assumption that the chemical energy is immediately changed into work, perhaps through the pro- duction of surface effects, have met with increasing favour, but data are as yet too few for the formulation of any really satisfactory theory. The close relation between the heat-production and the formation of lactic acid in contraction which have been shown to exist, is a suggestive fact whose full significance will only be revealed by further investigation. The restitution processes by which the original state of the muscle is restored after contraction are, of course, intimately related to those concerned in the actual shorten- ing ; but unless we know how, and in consequence of what chemical or physical changes, the equilibrium of the resting muscle has been disturbed, we cannot know how, or in consequence of what chemical or physical changes, it is restored. SECTION IV. — CHEMICAL PHENOMENA OF THE MUSCULAR CONTRACTION. The composition of dead mammalian muscle of the striped variety may be stated, in round numbers, as follows, but there are considerable variations, even within the same species: Water - 75 per cent. Proteins - - 20 ,, Fats, lecithin, and cholesterin - - 2 „ Nitrogenous extractives, creatin (about 0-4 peri cent.), carnosin, phospho-carnic acid, inosinic acid, purin bodies, such as uric^cid, hypoxan- thin, xanthin, etc. - Carbo-hydrates (glycogen, dextrose, maltose) Non-nitrogenous organic substances (lactic acid, inosit) Pigment (myohaematin or myochrome, a haemoglobin not precisely identical with that of blood). Inorganic substances less than i per cent, (chlorides, carbonates, phosphates, and sulphates of potassium, sodium, iron, calcium, magnesium) . Potassium is absent from the nuclei (Frontispiece). 758 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Of the nitrogenous extractives, creatin (p. 597) and carnosin are present in greatest quantity, muscle containing 0-2 to 0-4 per cent, of kreatin. Carnosin (CgHj^^C^) is a substance with basic properties, and can be split up into histidin and /3-amino-propionic acid, an amino-acid not identical with alanin (or a-amlno-propionic acid), but having the NH2 coupled to the tf instead of the a carbon atom (p. 566^. There is more water in the muscles of young than of old animals (v. Bibra), and more in tetanized than in rested muscle (Ranke). The fats are variable in amount, and belong to a small extent to the actual muscle-fibres. For even when the visible fat is separated with the utmost care, nearly i per cent, of fat still remains (Steil). The glycogen content varies extremely in different muscles and in the same muscle under different nutritive and functional conditions. Thus, in one and the same dog the biceps brachii contained 0-17 and the quadriceps femoris 0-53 per cent. In dogs on a diet rich in carbo- hydrate and protein the percentage in the whole skeletal musculature ranged from 0-7 to 3-7, and in the heart from o-i to 1-2. The average for human muscles has been given as 0-4 per cent. In lean horse-flesh Pfliiger found 0-35 per cent, of glycogen, but no sugar. The total nitrogen was 3-21 per cent, of the moist tissue. The lactic acid of muscle and other tissues is the d- lactic acid , which rotates the plane of polarization to the right. By the action of certain bacteria on cane- sugar /-lactic acid is obtained, which is left rotatory. The optically inactive fermentation lactic acid is obtained by the fermentation of lactose. Smooth muscle is somewhat richer in water than the striated variety from the same species, because skeletal muscle is richer in fat. Glycogen is either absent or present only in traces in the smooth muscle (of the stomach and bladder). Lactic acid, creatin, and creatinin are also found in much smaller amount than in striped muscle (Mendel and Saiki) . As in striated muscle, hypoxanthin is the conspicuous purin base occurring in the free form — i.e., obtainable in muscle extracts. The most remarkable difference in the quantitative relations of the inorganic constituents is that in striated muscle potassium preponderates over sodium and magnesium over calcium, whereas in the smooth variety this relation is reversed. It would be natural to expect that the proteins, which bulk so largely among the solids of the dead muscle, and which are so obvi- ously important in the living muscle, should be affected by contrac- tion. But up to the present time no quantitative difference in the proteins of resting and exhausted muscle has ever been made out. The quantity of creatin (and creatinin) is said by some authorities to be increased. The following chemical changes have been defi- nitely established. In an active muscle — (a) More carbon dioxide is produced, (b) More oxygen is consumed. (c) Lactic acid is formed, (d) Qfc^>gen is used up. (e) The substances soluble in water diminish in ai^^Ht; those soluble in alcohol increase. Production of Carbon Dioxide and Consumption of Oxygen during Contraction. — This subject has already been dealt with in part in connection with tissue respiration (p. 266). The fact that muscular exercise increases the carbon dioxide output and the oxygen absorp- tion at the pulmonary surface, shows that oxidation processes involving ultimately the combustion of carbon-containing substances CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 7&> are associated with the activity of the muscular tissue, but does not of itself prove that the final steps of the oxidation occur in the muscles themselves. This has been demonstrated, however, by observations on isolated muscles. When well supplied with oxygen, these, in addition to the stock of carbon dioxide in solution, in the form of carbonates and in other combinations, which they possess at the moment of isolation, continue to produce carbon dioxide, and this production is markedly increased by stimulation. The best evidence is to the effect that only preformed carbon dioxide is given off by isolated muscles in the absence of oxygen. They can go on contracting indeed, as previously stated, in an atmosphere of hydrogen or nitrogen, and may seem to be producing carbon dioxide, but the increased output appears to be due simply to an accelerated decomposition of already existing carbonates, or perhaps other combinations in which carbonic acid is loosely held, brought about by lactic acid, which in the absence of oxygen, is not trans- formed as it is under normal conditions, and accumulates in the muscular substance, uniting with bases, and thus displacing carbonic acid. Formation of Lactic Acid — Reaction of Muscle. — To litmus-paper fresh muscle is amphicroic — that is, it turns red litmus blue and blue litmus red. This is due, partly at least, to the phosphates. Mono- phosphate (tribasic phosphoric acid, H3PO4, in which one hydrogen atom is replaced, say, by sodium or potassium) reddens blue litmus, while diphosphate (where two hydrogen atoms are replaced) turns red litmus blue. Litmoid (lacmoid) differs from litmus in not being affected by monophosphates. Diphosphates turn red litmoid blue, and so does fresh muscle, which has no effect on blue litmoid. A cross-section of fresh muscle is about neutral (sometimes faintly acid) to turmeric paper, which is turned yellow by monophosphates. A muscle which has entered into rigor or has been fat \gued by prolonged stimulation is distinctly acid to blue litmus and to brown turmeric, reddening the former and turning the latter yellow, but does not affect blue litmoid. Perfectly fresh resting muscle excised with avoidance of all un- necessary manipulation contains very little lactic acid (as little as 0-02 per cent, expressed as zinc lactate). Mechanical injury, heating, and chemical irritation cause a marked increase in the amount. Under anaerobic conditions — in an atmosphere of hydrogen, for instance — lactic acid is spontaneously developed in the resting muscle so long as irritability persists, but not longer. In air, which for even^mall excised muscles corresponds to a partial asphyxia, there is a small increase in the lactic acid, but its pro- duction is very slow in comparison with that in the hydrogen atmosphere. In pure oxygen not only is there no accumulation of lactic acid for a long time after excision, but a portion of the amount originally present in the resting excised muscle disappears. The same is true of the lactic acid formed in a muscle fatigued by stimulation when it is afterwards placed in an atmosphere of pure 49 770 THE PHYSIOLOGY OP THE CONTRACTILE TISSUES oxygen. There is no doubt that the production of lactic acid in functional activity and its transformation into other substances are processes that go on also in the muscles of the intact body. The formation of the acid in the excised muscle, far from being a sign of death, is an index of the ' survival ' of a process by which it is normally formed, as the accumulation of it is an index of the crippling, in the absence of oxygen, of a mechanism by which it is normally transformed. The lactic acid which accumulates in the excised muscle in rigor and activity does not remain free, since blue litmoid paper is not reddened as it would be by free lactic acid. It causes a repartition of the bases at the expense of the sodium carbonate and disodium phosphate, the latter being changed into monophosphate, which, in part at least, accounts for the acid reaction to turmeric (Roh- mann). It is of great interest that this oxi dative transformation of lactic acid only occurs in muscle whose structure is so far pre- served that its irritability is not lost. In minced or triturated muscle it does not take place. The relations between the heat production, the formation of carbon dioxide, and the production of lactic acid indicate that liberation of lactic acid from some precursor is an essential stage in the sudden, ' explosive ' reaction or series of reactions which precedes and induces the mechanical response to stimulation. This stage takes place whether oxygen be present or absent, and it seems to be accompanied by a considerable liberation of energy, at the expense of which alone the anaerobically contracting muscle works. It is most probable that the liberation of lactic acid follows the same course in the muscle abundantly supplied with oxygen, although it has not been shown that oxidative processes, resulting in the forma- tion of carbon dioxide, do not contribute also at this stage to the energy which is transformed into the mechanical effect. But while in the absence of oxygen the reaction stops at the formation of lactic acid, when oxygen is available the cycle is completed by restitution processes which lead to the disappearance of the lactic acid either by restoration to its original position in the precursor from which it is derived, or perhaps in the case of a portion of the tactic acid to its combustion to carbon dioxide and water. For these restitution processes oxygen is essential, and it is to be supposed that the energy required for the rebuilding of the lactic acid precursor, or, to speak more generally, for the restoration of the muscle to its original state in readiness for a fresh contraction, is derived largely from oxidations in which carbon dioxide makes its appearance. The Precursor of Lactic Acid. — What material is the lactic acid formed from ? There are reasons for thinking that lactic acid is an intermediate substance which in metabolism serves as a link between CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 771 the products of protein decomposition and carbo-hydrates, and between carbo-hydrates and fat. From what we know of the production of lactic acid both outside the body and in the intestine from carbo-hydrates, it might seem a most plausible suggestion that in the active muscle it comes from glycogen. Glycogen is the one solid constituent of muscle which has been definitely proved to diminish during activity. It accumulates in a resting muscle, especially in a muscle whose motor nerve has been cut; but rapidly disappears from the muscles of an animal made to do work while food is withheld; or from the muscles of an animal poisoned by strychnine.which causes violent muscular contractions. But the best evidence points the other way — e.g., in rigor mortis lactic acid is produced just as in muscular contraction. Nay, more, the amount of lactic acid (as much as 0-5 per cent, expressed as zinc lactate) produced in full heat rigor (at 40° to 45° C.) is con- stant for similar excised muscles. This ' acid-maximum ' is the same when fresh muscle is at once put into rigor; or when fatigue is first induced, with formation of lactic acid, before rigor; or, finally, when the lactic acid of the fatigued muscle is caused to disappear under the influence of oxygen, and heat rigor is then brought about in the muscle (Fletcher and Hopkins). Yet in rigor mortis the quantity of glycogen is unaltered (Boehm). Further, under certain conditions an excised muscle is capable of producing a quantity of lactic acid much greater than could be derived from the glycogen contained in it. An indirect argument against the view that the lactic acid pre- cursor is glycogen has been based by Hill on the results of his studies on the heat production of surviving muscle. From the amount of heat evolved, he calculates that the precursor of lactic acid must have a heat value 10 per cent, greater than that of lactic acid. Now, the heat of combustion of dextrose is only about 3 per cent, more than that of lactic acid. He concludes that the precursor which yields lactic acid is a body of greater energy than dextrose. This, of course, does not preclude the possibility that the complex, whatever it is, from which lactic acid is liberated, contains a carbo- hydrate group. But it would not be profitable to pursue these speculations at present. The facts just mentioned suggest that it is the same precursor which yields the lactic acid developed with the onset of rigor. Further evidence of the close relations between the chemical changes occurring in contraction and those occurring in rigor will be developed in considering the latter phenomenon. The Substances metabolized in Muscular Contraction. — If the liberation of lactic acid were assumed to be the immediate cause of the mechanical changes in muscular contraction, if the nature of the body which yields lactic acid were known, and if it were proved, which is far from being the case, that the whole of the energy con- 77* THE PHYSIOLOGY OF THE CONTRACTILE TISSUES cerned in initiating and carrying out the mechanical effect is derived from the decomposition of this precursor, the question would still confront us, What are the materials at the expense of the energy of which the muscle is restored to its original condition ready for another contraction ? If the lactic acid is used over and over again, it is indeed the metabolism of these substances which will be chiefly represented in the waste products given off by the muscle ; the lactic, acid complex will merely represent a chemical machine through which the energy of these other substances is transformed into mechanical energy, and they will constitute the ultimate source of energy of the muscular contraction. In this sense the muscular glycogen, whether it yields lactic acid or not, is almost certainly one source of energy for the active muscle, being converted into dextrose, of course, before utilization. Dextrins and maltose, the intermediate products of this decomposition, have been detected in muscle, more maltose, indeed, than dextrose being present (Osborne), since the dextrose is rapidly oxidized. Glycogen cannot be the only source of muscular energy, for its amount is too small. For example, the heart of an average man, which weighs 280 grammes, contains about 60 grammes of solids, and among these certainly not more than i gramme of glycogen. In twenty-four hours it produces 120 calories of heat (pp. 138, 688), equivalent to the complete com- bustion of a little less than 30 grammes of glycogen. To supply this amount, the whole store of glycogen in the heart would have to be used and replaced every fifty minutes. But the accumulation of glycogen is immensely slower in the muscles of a rabbit made glycogen-free by strychnine, and therefore we have to look around for some other source of energy to supplement the glycogen. We have already brought forward evidence (p. 610) that, under ordinary circumstances, not a great deal, at any rate, of the energy of muscular contraction comes from the proteins. Of carbo-hydrates, the only one except the glycogen of the heart muscle which is at all adequate to the task of supplying so much energy is the dextrose of the blood. The quantity of blood passing through the coronary circulation has been estimated at 30 c.c. per 100 grammes of cardiac muscle per minute (Bohr and Henriques), which would be equivalent for an average man to about 120 litres in twenty-four hours. This quantity of blood will contain at least 1 20 grammes of dextrose, and about 32 grammes will suffice to supply all the heat produced by the heart. There is no reason to suppose that this dextrose must first be changed into muscular glycogen, which only represents a certain amount of reserve carbo-hydrate. Of proteins a little less than 30 grammes would be needed, of fat a little more than 12 grammes. We see, therefore, how intense must "be the metabolism that goes on in an actively contracting muscle. On any probable assumption as to the source of muscular energy, a quantity of material equal to half of its solids must be used up by the heart in twenty-four hours. Or, to put it in another way, the heart requires not less than two-fifths of its weight of ordinary solid food in a day. The body as a whole requires £$ to ^ of its weight. The general conclusions to which physiologists have been led as to the relative importance of the different food substances for CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 773 muscular work have been previously given (p. 612), and need not be repeated here. It may be added that the various food substances yield muscular energy in isodynamic relation. In other words, a given amount of muscular work requires the expenditure of approxi- mately the same quantity of chemical energy, whether it comes almost entirely from protein, or chiefly from carbo-hydrates, or chiefly from fat. Some observers have stated that the taking of even a comparatively small quantity of sugar vastly increases the capacity for muscular work as measured by the ergograph (p. 750). But although it is not to be doubted that sugar is under normal circumstances one of the most important substances used up in muscular contraction, the claim that sugar is, par excellence, the food for muscular exertion has not yet been made out. Physico-Chemical Conditions of Muscular Contraction. — For excised fresh muscle A (p. 427) has been estimated at 0-68° C. But this is probably higher than in the living body, for after excision waste products, with their relatively small and numerous molecules, are still for a time produced, and are no longer removed by the blood. In salt solutions isotonic with the muscle substance — e.g., for the frog's gastrocnemius at room-temperature a 0-75 per cent, solution of sodium chloride — the resting muscle neither gains nor loses water for some hours. The active muscle behaves quite differently. When a muscle immersed in isotonic salt solution is tetanized, water enters it, leading to an increase in weight and a diminution in specific gravity (Ranke, Loeb, Barlow). The same occurs even when blood is circulated through active muscles, the blood becoming poorer in water (Ranke). This may be explained by the increase of osmotic pressure in the muscle substance which must accompany the decomposition of large molecules into small. As fatigue progresses, a movement of water in the reverse direction occurs, and the muscle rapidly loses water. Exposure of the fatigued muscle for a sufficient time to an atmosphere of oxygen restores the osmotic proper- ties of the resting muscle. Striking differences have also been demon- strated in the behaviour of resting and fatigued muscle to hypotonic solutions or water. Hales observed long ago that, on injecting large quantities of water into the bloodvessels of a dog, so as to replace the blood, marked swelling of the muscles occurred. This physiological fact is well known to the pork-butchers in China, who have given it a practical, if not a very praiseworthy, application in sophisticating their product by increasing its weight (MacGowan). So long as the muscular fibres are uninjured they are permeable or impermeable for exactly the same compounds as other animal and vegetable cells. All substances easily soluble in media like ether or olive oil readily penetrate them (Overton). To most salts they are relatively impermeable, as is shown by the fibres retaining their original volume in isotonic solutions of them. In particular, they cannot easily take up or retain the salts of the blood-plasma, otherwise the observed qualitative differences — e.g., the preponderance of potassium in the muscle and sodium in the plasma — could not be maintained. There are facts which indicate that temporary changes in the permeability to ions, not only of muscular fibres, but also of nerve fibres and other excitable structures, are concerned in their stimulation. Potassium salts after a time seem to produce an effect upon frog's muscle, which alters its permeability so that it takes up water from hypertonic solutions. 774 Calcium salts have the opposite effect (Loeb). Sodium (and in a minor degree lithium) salts have a peculiar relation to the contraction of skeletal muscle, for which they appear to be indispensable. Yet sodium chloride produces a paralyzing action on the frog's motor nerve-endings, so that after perfusion with a solution of that salt stimulation of the motor nerve causes no contraction, or with a slighter degree of paralysis contraction only after a long interval. The effect can be counteracted by solutions containing calcium salts (Locke, Gushing). Rigor Mortis. — When a muscle is dying, its excitability, after perhaps a temporary rise at the beginning, diminishes more and more until it ultimately responds to no stimulus, however strong. The loss of excitability is not in itself a sure mark of death, for, as we have seen, an inexcitable muscle may be partially or com- pletely restored; but it is followed, or, where the death of the muscle takes place very rapidly, perhaps accompanied, by a more decisive event, the appearance of rigor. The muscle, which was before soft and at the same time elastic to the touch, becomes firm; but its elasticity is gone. The fibres are no longer translucent, but opaque and turbid. If shortening of the muscle has not been opposed, it may be somewhat contracted, although the absolute force of this contraction is small compared with that of a living muscle, and a slight resistance is enough to prevent it. The reaction is now distinctly acid to litmus. This is rigor mortis, the death-stiffening of muscle. An insight into the real meaning of this singular and sometimes sudden change was first given by the experiments of Kiihne. He took living frog's muscle, freed from blood, froze it, and minced it in the frozen state. The pieces were then rubbed up in a mortar with snow containing I per cent, of common salt, and a thick neutral or alkaline liquid, the ' muscle-plasma,' was obtained by filtration. This clotted into a jelly when the temperature was allowed to rise, but at o° C. remained fluid. The clotting was accompanied by a change of reaction, the liquid becoming acid. An equally good, or better, method is to use pressure for the extraction of the plasma from the frozen fragments of muscle. A low temperature is essential, otherwise the plasma will coagulate rapidly within the injured muscle. A similar plasma can be expressed from the skeletal muscles of warm-blooded animals (Halliburton), and with greater difficulty from the heart. When the muscle, after exhaustion with water, is covered with a solution of a neutral salt, a 5 per cent, solution of magnesium sulphate or 10 per cent, solution of ammonium chloride being the best, certain proteins are extracted which clot or are precipitated much in the same way as the muscle-plasma obtained by cold and pressure; and the process is hastened by keeping them at a temperature of 40° C. In the extracts of mammalian muscle three chief proteins are present : paramyosinogen (v. Furth's myosin), coagulating by heat at 47° to 50° C. ; myosinogen (v. Fiirth's myogen), coagulating at 55° to 60° C., usually about 56°); and serum-albumin, coagulating about 73°. The CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 775 serum-albumin belongs to the blood and lymph, and is not a constituent of the muscle-fibre. The most recent work on the subject is that of Botazzi, who obtained muscle juice without the addition of water or salt solutions, by rubbing muscles up with sand, and then subjecting the triturated material to a pressure of many atmospheres. He finds that, leaving out of account the serum-albumin, muscle juice contains only one protein in solution, and this corresponds upon the whole in its properties to myosinogen. A second protein, and only these two have been proved to exist in muscle juice, is not in solution, but in the form of very fine granules revealed by the ultramicroscope. This corresponds in a general way to paramyosinogen. Botazzi supposes that it repre- sents the substance of the muscular fibrils. The granules show a ten- dency even at the ordinary temperature to agglutinate and to be pre- cipitated. The process is hastened by dilution with water, removal of the salts by dialysis, addition of acids, and the agglutination and pre- cipitation are accomplished very rapidly at 45° to 55° C., giving rise to ' heat coagulation.' The protein in solution (myosinogen) is in- soluble in distilled water when thoroughly freed from salts, and is pre- cipitated by dialysis, but not so easily as paramyosinogen. The total proteins in the juice obtained by pressure varied from 5-3 to 9-5 per cent., a great deal of the muscle protein being, of course, left in the residue. The granules (paramyosinogen) constituted from a third to two-thirds of the protein in different experiments, and the true pro- portion must have been considerably higher, since on account of their small size the loss in separating them by filtration was great. The ' myosin ' precipitate, which rapidly forms in muscle-plasma at body temperature, is sometimes called the muscle-clot, and the liquid which is left the muscle-serum, but it would probably be better to avoid these terms, as they suggest an analogy with the coagulation of blood-plasma, which is apt to be misleading. A similar precipitate or clot seems to be formed in the interior of the muscular fibres in natural rigor and in the rapid rigor produced by heating a muscle to a little above the body- temperature. But in natural rigor the whole of the paramyosinogen and myosinogen do not undergo the change, since a certain amount of these substances can as a rule be extracted from dead muscle by saline solutions. Thus, in rabbit's muscles, before the onset of rigor mortis, 87-3 per cent, of the total protein was found to be soluble in 10 per cent, ammonium choride solution, and only 12-7 per cent, coagulated; while after rigor had occurred, 71-5 per cent, was coagulated, and only 28-5 per cent, remained soluble (Saxl). It is not known whether in the living muscle paramyosinogen and myosinogen exist as such. It has, indeed, been stated that, if a tracing is taken from a muscle which is gradually heated, it first shortens at the temperature of coagulation of para- myosinogen, and then again at that of myosinogen, and that in frog's muscle there is an additional shortening at 40°, the temperature at which in extracts an additional heat precipitate occurs. The conclusion has been drawn that these substances must be present as such in the living fibres, and that the successive shortenings are mechanical phe- nomena due to their heat coagulation. Similar shortenings have been described in nerve and liver tissue at about the temperatures at which the proteins in extracts of these tissues are coagulated by heat. But Meigs has shown that the supposed correspondence is far from being exact, and that muscles whose proteins have been already coagulated in a mixture of alcohol and salt solution still show the typical shortening on being heated. The heat shorter ing is, therefore, dependent on some other process than aggregation of the particles of coagulable protein. 776 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES Certain analogies between rigor and muscular contraction were early pointed out. In both there is (i) shortening; (2) heat-pro- duction; (3) formation of lactic acid; (4) discharge of carbon dioxide; (5) electrical changes. As regards the production of lactic acid, there is reason to believe that the process is fundamentally the same as in contraction, and the study of rigor, especially of certain of the artificially induced forms — e.g., heat rigor — in relation to the liberation of lactic acid, carbon dioxide, and heat, has thrown light upon the changes normally occurring in muscle. Another analogy might be forced into the list by anyone who was deter- mined to see only rigor in contraction: the rigor passes off as the contraction passes off, although the ' resolution ' of a rigid muscle takes days, the relaxation of an active muscle a fraction of a second. The disappearance of rigor is not dependent on putrefaction; it takes place when growth of bacteria is prevented (Hermann). Possibly it is connected with autolytic processes due to intracellular ferments (p. 599). Why does coagulation of myosin occur at the death of the muscle ? To this question no clear answer can be given. Some have looked on the process as analogous to the clotting of blood when it is shed, and it has even been suggested that just as a fibrin ferment is developed when the leucocytes and blood-plates begin to die, a myosin ferment, which aids coagulation, is developed in dead or dying muscle. But no proof has been given of the existence of such a ferment. And it is easy to make too much of the apparent analogy between the clotting of muscle and the clotting of blood, for there are differences as well as resemblances. For instance, the addition of potassium oxalate does not prevent coagulation of muscle extracts, as it does of blood and blood-plasma. If the development of lactic acid in the muscle is not the primary cause of the coagulation which constitutes the essential feature of rigor mortis, it seems to be closely related to it. For when excised muscles are abundantly supplied with oxygen, no lactic acid accumulates in them, and the final loss of excitability of the muscle is not followed by rigor. In any case, direct precipitation of hitherto unclotted muscle proteins may be induced by the acid, or the acid salts formed in its presence. Deficiency of oxygen is associated with the occurrence of rigor mortis, as it is with the accumulation of lactic acid, and a developing rigor can be abolished by oxygen, and its onset long or indefinitely delayed. When strict aseptic technique is observed an excised sartorius muscle of the frog may remain irritable in sterile Ringer's solution, even without oxygenation, for as long as three weeks (Mines). Various influences affect the onset of rigor. Fatigue hastens it; heat has a similar effect; the contact of caffeine, chloroform, and other drugs causes most pronounced and immediate rigor, CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 777 Blood applied to the cross-section of a muscle first stimulates the fibres with which it is in contact, and then renders them rigid. But it is to be remembered that normally the blood does not come into direct contact even with the sarcolemma, much less with its contents. The effect of heat is of special interest. A skeletal muscle of a frog, like the gastrocnemius, if dipped into physiological saline solution at 40° or 41° C. goes into rigor at once; the frog's heart requires a temperature 3° or 4° higher; the distended bulbus aortae can withstand even a temperature of 48° for a short time. An excised mammalian muscle passes into immediate rigor at 45° to 50°. In heat rigor the reaction of the muscle becomes strongly acid owing to the formation of lactic acid, and the evolution of carbon dioxide is also increased. The total discharge of carbon dioxide in heat rigor induced at 40° amounts to 35 to 40 c.c. per 100 grammes of muscle. An additional 15 to 20 per cent, is obtained on heating to 75° C. to completely coagu- late the proteins, and a further 15 to 20 per cent, on heating to about 100° C. When a muscle is scalded by being suddenly immersed in boiling salt solution, lactic acid is not formed, but carbon dioxide to the amount of 60 to 70 per cent, is discharged. An excised muscle kept in oxygen for many hours, during which it has discharged several times as much carbon dioxide as is ever liberated by heating, still yields the normal discharge on heating whether to 40° C. or to 100° C. On the other hand, previous survival in an anaerobic atmosphere (of nitrogen) reduces greatly or abolishes the yield of carbon dioxide at 40° C., although not that at 100° C., the sum of the carbon dioxide given off to the atmosphere of nitrogen and that given off on heating to 100° C. being about equal to the total amount which would have been dis- charged by a freshly-excised muscle on heating first to 40° C. and then to 100° C. If acid is added to a fresh muscle at about o° C. even more carbon dioxide is liberated than in heat rigor, while the yield of lactic acid is, even after many hours, very little increased above the normal amount for fresh resting muscle. When the acidified muscle after the discharge of the carbon dioxide is now heated to 40° C., the yield of lactic acid is increased, but only traces of carbon dioxide are given off. From these and similar observations, Fletcher concludes that the carbon dioxide discharged during heat rigor at 40° C. is pre-existent carbon dioxide set free from carbonates or other compounds by the lactic acid known to be produced in heat rigor. The carbon dioxide discharged at 75° and 100° C. he regards as held by muscle colloids or in combination with amino-acid groups. These results tend to discredit the ' inogen ' theory (p. 270), with its assumption that ' intramolecular oxygen ' is stored away in the muscle, which was largely based upon erroneous observations on the discharge of carbon dioxide from heated muscles. According to this theory, carbon dioxide and lactic acid were supposed to arise from a common precursor into which oxygen had been previously introduced. 778 THE PHYSIOLOGY OF THE CONTRACTILE TISSUES The production of heat in heat rigor is also of great interest. Hill has shown that it amounts to from 0-6 to i-o gramme calorie per gramme of muscle. Of this no more than 0-05 calorie can be due to the heat of neutralization of lactic acid by the sodium bicarbonate in the muscle, with which it reacts as soon as it is liberated. The rest of the heat is associated with the chemical reaction by which lactic acid is formed from its precursor, a reaction which, there is every reason to believe, is the same as that which occurs in muscular contraction. The heat production can only be due in very slight degree to the physical alteration (clotting or precipitation) of the muscle proteins. The so-called rigor caused by water, which is not a true rigor, causes no increase in the carbon dioxide given off. Chloroform, on the other hand, produces a marked increase in the carbon dioxide production, and this is evidently related to its action in hastening the onset of rigor. Rigor mortis is to some extent in- fluenced by the nervous system, for section of its nerves retards the onset of rigor in the muscles of a limb. Ante-mortem stimula- tion of the peripheral ends of the vagi, even with currents too weak to cause a perceptible effect upon the heart-beats, prolongs the period of spontaneous contraction and the irritability of the ven- tricles after death, and retards the onset of rigor (Joseph and Meltzer). Cold rigor is obtained when frog's muscles are cooled to —15° C. The muscles remain perfectly translucent. They do not recover their irritability on thawing, but if cooled only to — 7° C. they recover (Folin). In a human body rigor generally appears not earlier than an hour, and not later than four or five hours, after death. In ex- ceptional cases, however, it may come on at once, and the annals of war and crime contain instances where a man has been found after death still holding with a firm grip the weapon with which he had fought, or which had been thrust into his hand by his murderer (so-called cataleptic rigor). It is related that after one of the battles of the American Revolutionary War some of the dead were found with one eye open and the other closed as in the act of taking aim. A high temperature favours a rapid onset; a body wrapped up in bed will, other things being equal, become rigid sooner than a body lying stripped in a field. Muscular exhaustion, as we have said, is another favouring condition: hunted animals and the victims of wasting diseases go quickly into rigor. It is a rule, but not an invariable one, that rigor, when it comes on quickly, is short, and lasts longer when it comes on late. All the muscles of the body do not stiffen at the same time; the order is usually from above downwards, beginning at the jaws and neck, then reaching the arms, and finally the legs. After two or three days the rigor disappears in the same order. The position of the CHEMICAL PHENOMENA OF MUSCULAR CONTRACTION 779 limbs in rigor is the same as at death; the muscles stiffen without any marked contraction. This can be strikingly shown on a newly- killed animal by cutting the tendons of the extensors of one foot and the flexors of the other ; when natural rigor comes on, the feet remain just as they were. If heat rigor, however, is caused, the one foot becomes rigid in flexion and the other in extension; and the contraction-force is considerable, although not so great as that of an electrical tetanus in a living muscle. The Possibility of Recovery of Muscles after Rigor. — When the circu- lation in the hind legs of rabbits is interrupted by compression or ligation of the abdominal aorta (Stenson's experiment), the muscles lose their excitability, but speedily recover, if they have not been deprived of arterial blood for too long a time, when the blood is again allowed to reach them. A longer interruption of the circulation leads not only to total inability to respond to stimulation, but also to rigor, and most observers are agreed that, as regards the skeletal muscles at least, this is the irrevocable end of excitability. Brown-Sequard, indeed, stated that after the full development of rigor in the rabbit's muscles (Stenson's experiment), and also in the hand of an executed criminal through which an artificial circulation was established, recovery ensued. But probably the rigor was incomplete or did not involve all the fibres. In heart muscle the conditions appear to be somewhat different, and Heubel has alleged that rhythmical contractions of the frog's heart can be restored by filling its cavity with blood, after rigor has been caused by heat and in other ways, and we have already seen that the same is true of the mammalian heart alter the onset of rigor. Excised frog's muscles which have undergone rigor mortis become less stiff when exposed to an atmosphere of oxygen. CHAPTER XIV NERVE THE voluntary movements are originated by efferent or outgoing impulses from the brain, which reach the muscles along their motor nerves. The involuntary movements and the secretions are in many cases able to go on in the absence of central connec- tions, but are normally under central control. Afferent impulses are continually ascending to the cord and brain from the skin, joints, bones, muscles, and organs of special sense like the eye and the ear. Everywhere the connection between the nervous centres and the peripheral organs, and between different parts of the central nervous system, is made by nerve- fibres. Those which run outside the brain and cord are called peripheral nerve-fibres to distinguish them from the intracentral fibres of the central nervous system itself. In this chapter we propose to consider certain of the general properties of nerve-fibres. Most of our knowledge of these proper- ties has been derived from experiments on the peripheral, and particularly the peripheral motor nerves ; but there is every reason to believe that the main results are true of all nerve-fibres, afferent and efferent, peripheral and central. What we call nerve-fibres were known and named, and many im- portant facts in their physiology discovered, long before their true morphological significance was recognized. The researches of recent years have shown that every nerve-fibre is, as regards its essential con- stituent the axis-cylinder, a process of a nerve-cell. The nerve-cells, each of which, including all its processes, may be conveniently termed a neuron, are the essential elements of the nervous system. The cell- bodies of most of the neurons are situated in, or in close relation to, the spinal cord and the brain, and therefore the detailed description of them will be reserved till we come to treat of the central nervous system (see p. 850 and Figs. 328 to 340). It is enough to say here that in general a nerve-cell gives off two kinds of piocesses: (i) one or more dendrites or protoplasmic processes, which repeatedly bifurcate like the branches of a tree into thinner and thinner twigs, and extend only for a relatively short distance from the cell-body; (2) an axis-cylinder process or axon, which as a rule runs for a considerable distance without altering its calibre, and either gives off no branches (as in the peripheral nerves) or only a comparatively small number of lateral twigs (col- 780 THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 781 laterals). Ultimately the axis-cylinder process and its collaterals, if it has any, end by breaking up into a brush, a plexus or a feltwork or basketwork of fibrils. The axons of different nerve-cells vary greatly in length. Some terminate within the grey matter of the brain or spinal cord not far from their origin; others run in the white tracts of the central nervous system or in the peripheral nerves for half the height of a man. All except the shortest axis-cylinder processes become clothed at a little distance from the cell-body with a protective covering, which continues to invest them (and their collaterals) throughout the rest of their course, disappearing only when they begin to break up at their terminations. An axis-cylinder process (spoken of simply as the axis-cylinder, when considered apart from the nerve-cell) constitutes, with its covering, a nerve-fibre. The axis-cylinder is the essential conducting part of the fibre, for it is present in every nerve-fibre, running from end to end of it without break, and towards the periphery it is alone present. It is made up of fine longitudinal fibrils embedded in interstitial substance (Fig. 329, p. 851). Such a fibrillar structure is best shown after treatment of the nerve- fibres with certain reagents, although it is certain that it exists pre- formed in the living fibres. SECTION I. — THE NERVE- IMPULSE OR PROPAGATED DISTURBANCE: ITS INITIATION AND CONDUCTION. So far as we know, the only function of nerve-fibres is to conduct impulses from nerve-centres to peripheral organs, or from peripheral organs to nerve-centres, or from one nerve-centre to another. In the normal body these impulses never, or only very rarely, originate in the course of the nerve- fibres; they are set up either at their peripheral or at their central endings. By artificial stimu- lation, however, a nerve-impulse may be started at any part of a fibre, just as a telegram may be dispatched by tapping any part of a telegraph wire, although it is usually sent from one fixed station to another. Nature of the Nerve- Impulse. — What the nerve-impulse actually consists in we do not know. All we know is that a change or dis turbance of some kind, of which the most evident token is an electrical change, passes over the nerve with a measurable velocity, and gives tidings of itself, if it is travelling along efferent fibres — that is, out from the central nervous system — by the contraction or inhibition of muscle or by secretion; if it is travelling along afferent fibres — that is, up to the central nervous system — by sensa- tion, or by reflex muscular or glandular effects. Whether the wave which passes along the nerve is a wave of chemical change (such, to take a very crude example, as runs along a train of gunpowder when it is fired at one end), or a wave of mechanical change, a peculiar and most delicate molecular shiver, if we may so phrase it, or a shear in a definite direction along the colloidal substance of the axis-cylinder (Sutherland), there is no absolutely definite experimental evidence to decide. An 7&z NERVE electrical change accompanies the nerve-impulse travelling at the same rate, and although this is to be distinguished from the impulse itself, there is little doubt that the latter is essentially connected with a disturbance of the electrical equilibrium of the nerve- substance. An attempt has been made to settle the question by determining the temperature coefficient of the velocity of conduction of the impulse — i.e., the quantity which measures the change of velocity for a given change of temperature. For most physical processes the quotient velocity at T»+ 10° . — TT T-~ . where Tn is any given temperature, is not greater than 1-2, while for frog's sciatic nerve the temperature coefficient for the most part lies between 2 and 3 (Snyder). The mean value of a large number of observations is 1-79, with T«= 8° to 9° C. (Lucas). For the pedal nerve of the giant slug the mean value of the temperature coefficient is 1-78 (Maxwell). In other words, while for the majority of physical processes an increase of 10° C. increases the velocity of the process by at most one-fifth, the same increase of temperature increases the velocity of conduction of the nerve-impulse by four- fifths, or even more. While it is true that it may not be entirely safe to apply such a criterion to a biological process which need not be either entirely chemical or entirely physical, and very likely is a complex one, the suggestion, so far as it goes, is undoubtedly in favour of the chemical hypothesis. That chemical changes go on in living nerve we need not hesitate to assume; and, indeed, if the circulation through a limb of a warm- blooded animal be stopped for a short time, the nerves lose their excitability. Even the nerves of cold-blooded animals gradually become inexcitable and incapable of conduction when placed in an oxygen-free medium, as the oxygen already contained in the tissue is exhausted. The excitability and conductivity of the nerve are restored by oxygen. It is clear, then, that even a resting nerve requires oxygen, and it can be shown that the loss of function is acceleiated by stimulation in the absence of oxygen. But the metabolism is very slight compared with that in muscle or gland. Until recently even in active nerve no measurable production of carbon dioxide had ever been observed, nor, in fact, had any chemical difference between the excited and the resting state ever been unequivocally made out. However, it has been announced that by the aid of an extremely delicate method of estimating small quantities of carbon dioxide, a measurable production of carbon dioxide can be detected even in resting frogs' nerves, and that this pro- duction is increased two to three times on stimulation (Tashiro). This result is somewhat puzzling in view of the fact that neither in cold- blooded nor in mammalian nerves is there any sensible rise of temper- ture during stimulation. With the apparatus shown in Fig. 272 (an electrical resistance thermometer or bolometer whose use depends upon the fact that the electrical resistance of a metallic conductor varies with its temperature) an increase even of 0-0003° C. in the temperature of the sciatic nerves of dogs could not be detected during tetanization. Rolleston failed to find evidence of a rise of even 0-0002° C. in frog's nerves during stimulation. And according to the latest investigation with a more suitable and much more sensitive thermo-electric arrange- ment, the passage of a single nerve impulse along a frog's nerve cannot be associated with an increase of temperature in the nerve of even the hundredth million of a degree (A. V. Hill). The difficulty of inducing fatigue in nerves under ordinary conditions has been considered a strong B THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 783 support of the physical nature of the conduction process. Neverthe- less, it is possible to show by special methods that nerve can be tempor- arily fatigued, although it recovers very rapidly. When a medullated nerve is stimulated, a brief period ensues during which it refuses to respond to a second stimulus. This refractory period is normally very short — not more than 0-002 second for the frog's sciatic. But it can be greatly prolonged by cold, asphyxia, or anaesthesia, especially by the alkaloid yohimbine (Tait and Gunn), and when the refractory period is thus prolonged, fatigue phenomena are readily induced by stimulation. And while the nerves of warm-blooded animals at body temperature and those of cold-blooded animals at about 32° C. can hardly be shown to undergo fatigue when tetanized in atmospheric air, fatigue phe- nomena are easily elicited when the temperature is lowered even although air is supplied (Thorner). Stimulation of Nerve. — With some differences, the same stimuli are effective for nerve as for muscle (p. 737) ; but chemical stimula- tion is not in general so easily obtained. The so-called thermal Fig. 372. — Electrical - Resistance o p Thermometer (Natural Size) as > ¥ ¥ 3 used for investigating heat-pro- duction in mammalian nerves in situ. A, a piece of hard rubber in the hook-shaped part of which the fine platinum wire P is fixed, and covered with insulating varnish; c, c, thick copper wires connected • with P, fastened in grooves, and » covered with paraffin. Above they end in contact with the small binding posts, B1P B2. B is a hard rubber sliding piece, with a slot s. When B is in position the screw, a, projects through the slot. By a nut on this screw B is fixed on A when the nerve has been arranged in the ^' groove. P stimulation is not a real stimulation due to the sudden change of temperature. The irregular contractions of the muscle caused by the local application of heat to the nerve are dependent on desicca- tion of the nerve. Chemical Stimulation. — When hyper- or hypo tonic solutions are em- ployed, the withdrawal or entrance of water may be an important factor. For salts which penetrate the fibres with equal difficulty this factor can be eliminated by applying them as isotonic solutions. There is evidence that chemical stimulation proper, as distinguished from the stimulation produced by changes in the water content of the fibres by osmosis, is connected with the electrical charges on the dissociated ions of the salts (p. 428). Electrical stimulation, indeed, may only be a variety of chemical stimulation (Loeb, Mathews, etc.). Mechanical Stimulation may be applied to a nerve by allowing a small weight to fall on it from a definite height or by permitting mercury to drop upon it from a vessel with a fine outflow tube. A regular tetanus may thus be obtained. Tigerstedt found that the smallest amount of ,vork spent on a frog's nerve which would suffice to excite it was a little less 784 NERVE than a gramme-millimetre — that is, the work done by a gramme falling through a distance of a millimetre, or (taking an erg as equivalent to i^rro gramme-centimetre) about 100 ergs. No doubt a great part of this is wasted, as a much smaller quantity of work done by a beam of light on the retina or by an electrical current on an isolated nerve, both of which may be supposed to act more directly on the excitable con- stituents, suffices to cause stimulation. Thus, the work done by the minimal, natural or specific, stimulus for the retina in the form of green light may be as little as — g erg (S. P. Langley), or only one-ten-thousand - millioneth part of the minimum work necessary for mechanical stimula- tion. Again, with electrical stimulation (closure of a voltaic cuirent, or condenser discharges) it has been shown that an amount of work equal to — 4 erg may be enough to cause excitation of a frog's nerve. This is ten thousand times as great as the minimal luminous stimulus, but a million times less than the minimal mechanical stimulus. The laws of electrical stimulation for nerve are essentially the same as those we have already discussed for muscle (p. 741). The voltaic current stimulates a nerve, as it does a muscle, at closure and opening. During the flow of the current, so long as its intensity remains constant, there is as a rule no excitation, or at least none which is propagated along the nerve, so that the muscles supplied by it remain uncontracted. But under certain conditions — for example, when the nerve is more excitable than usual (as is the case with nerves taken from frogs which have been long kept in the cold) — a closing tetanus may be seen while the current continues to pass through the nerve, and an opening tetanus after it has ceased to flow, just as when the current is led directly through the muscle. Sensory nerve-fibres, too, are stimulated by a voltaic current during the whole time of flow. Induction shocks are relatively more powerful stimuli for nerve than the make or break of a voltaic current. The opposite, as we have seen, is true of muscle; and, upon the whole, we may say that muscle is more sluggish in its response to stimuli, and is excited less easily by very brief currents, than nerve is. An apparent illustration of this difference is the fact that the nervous excitation has no measurable latent period, while muscular excitation has. But it is quite possible that, if the conditions of experi- ment were as favourable in nerve as in muscle, a sensible latent period might be found here too. In nerve as in muscle, strength of stimulus and intensity of response correspond within a fairly wide range, when we take the height of the muscular contraction or the amount of the negative variation (p. 824) as the measure of the nervous excitation. Summation of stimuli, super- position of contractions, and complete tetanus, are caused by stimulating the muscle through its nerve, just as by stimulating the muscle itself (p. 756). Excitability of Nerve. — It has usually been stated that the ex- citability of frog's nerve, as measured by the muscular response to stimulation, is increased by rise of temperature, and diminished by fall of temperature. It has, however, been shown that this increase of excitability is only apparent, and due to the strengthening of the current by diminution of the resistance, since the resistance of all animal tissues, like that of electrolytic conductors in general, diminishes as the temperature rises (Gotch). When precautions THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 785 are taken to keep the current intensity the same at the various temperatures compared, it is found that cooling of a (frog's) nerve, even to 5° C., increases the excitability for currents of long duration (several hundredths of a second). It has, indeed, been shown both for muscle and for nerve that the cooler tissue requires a smaller current strength for its excitation when the current is of long dura- tion. With brief currents this effect is masked, either partially or completely, by the greater increase of current strength needed in the case of the cooler tissue to compensate fora given decrease in duration (p. 789) (Lucas and Mines). This is the reason that for induction shocks or voltaic currents of short duration, the excitability of the nerve seems to be increased by a rise of temperature (up to about 30° C. in the case of frog's nerve), and diminished by cooling. Drying of a nerve at first increases its excitability; and the same is true of separation of a nerve from its centre. In the latter case the increase of irritability begins at the proximal end of the nerve, and travels towards the periphery. As time goes on, the excita- bility diminishes, and ultimately disappears in the same order (Ritter-Valli Law). At a certain stage it may be found that a given stimulus causes a smaller and smaller contraction the farther down the nerve — that is, the nearer to the muscle — it is applied. On this was based the now abandoned ' avalanche theory/ according to which the impulse continually unlocked new energy as it passed along the nerve, and so gathered strength in its course like an avalanche. It is now known that no material change takes place in the intensity of the excitation while it is being propagated along a normal uninjured nerve. For instance, experiments on the phrenic nerve, in its natural position, and with all its connections intact, have shown that with a given strength of stimulus the amount of contraction of the diaphragm is the same whether the nerve be excited in the upper, middle, or lower portion of its course. In the above experiment on the isolated, and therefore injured, nerve, the contraction varies in height with the distance of the point of stimulation from the muscle, not because the excitation grows as it travels, but because it is already greater at the moment when it sets out from a point near the central end of the nerve than at the moment when it sets out from a point near the muscle. Electrotonus. — Although the constant current does not, unless it is very strong or the nerve very irritable, cause stimulation during its passage, it modifies profoundly the excitability and conductivity of the nerve. In the neighbourhood of the kathode the excitability is increased (condition of katelectrotonus), while around the anode it is diminished (anelectrotonus). Immediately after the opening of the current these relations are for a brief time reversed, the excitability of the post-kathodic area (area which was at the kathode during the flow) being diminished, and that of the post-anodic 50 786 NERVE increased. In the intrapolar area there is one point the excita- bility of which is not altered. This indifferent point, as it is called., shifts its position when the intensity of the current is varied, moving towards the kathode when the current is increased, towards the anode when it is diminished. It is only under certain definite conditions that these phenomena, first described by Pfluger, appear in their purity and uncomplicated by other changes. The nerve should be quite fresh, the current a weak or at most a moderately strong one, and the stretch of nerve employed should be as far as possible from the cross section, and from the cross sections of branches. The middle region of the frog's sciatic nerve is the best. When all these conditions are fulfilled, the whole stretch of nerve in katelectrotonus — i.e., the part on both sides of the kathode and at the kathode itself shows an increased stimulation effect, the more pronounced the nearer to the kathode the point of stimulation. This condition, however, only lasts an instant. Then the excitability begins to sink sharply first at the kathode, then on both sides of it, till it ultimately becomes decidedly less than the initial excitability. This secondary depression of excitability, always most marked at the very kathode, is just as con- Fig. 273 . — Katelectrotonus. Weak tetanus of muscle (the right-hand elevation), greatly intensified in katelectrotonus of the motor nerve (th« left-hand elevation). Fig. 274. — Anelectrotonus. Strong tetanus of muscle (left-hand ele- vation), lessened in strength by anelectrotonic condition of the mo- tor nerve (right-hand elevation). stant a phenomenon as the preliminary increase. The stronger the cur- rent the more profound is the depression, the more quickly it is de- veloped, and the greater is the distance to which it spreads along the nerve. With a certain strength of current the depression appears so rapidly that the preliminary increase of excitability may be completely missed. When the current is opened the excitability quickly increases again, but with strong currents it may remain depressed for a while. At the anode changes in the reverse direction may be observed, although they are less pronounced than at the kathode. Thus at the anode during the passage of the current the initial depression of the excitability tends to give place to an increase (Werigo). These statements have been made on the strength of experiments in which the height of the muscular contraction was taken as the index of the excitability of the nerve at any given point. It is difficult, however, to disentangle the effects of alterations in the excitability from the effects of alterations of conductivity — i.e., of the power of a portion of the nerve to conduct an impulse set up elsewhere. Whether these two properties are distinct or not is a question which will be considered a little later on. But it is perfectly clear that in deducing conclusions THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE as to the effect produced on a nerve by excitation at a given point from the resultant effect on the muscle to which the nerve is attached (or on a galvanometer or electrometer if we are following the effect by means of the electrical changes), we must know whether the change set up in the nerve at the point of excitation can pass freely along the nerve to the muscle or to the point at which it is led off to the galvanometer. Now, changes of conductivity are certainly produced in a nerve by the constant current, which even outlast its flow. For all currents above a certain strength the conductivity at the kathode and in its neigh- bourhood is eventually dim- inished, and with currents still only moderately strong the block deepens into im- passability. The conduc- tivity at the anode is, during this stage, higher than at the kathode, so that at the time of full kathodic block the nerve - impulse still passes through the region around the positive pole. With still stronger currents the con- 275._Diagram Of Changes of Excitability ductivity here, too, dimm- and Conductivity produced in a Nerve by a Voltaic Current. E, changes of excitability during the flow of the current, according to Pfliiger. These are seen most typically with the weaker currents. In particular the in- creased excitability at and around the kathode when the current is strong very quickly gives place to depression. The ordinates drawn from the abscissa axis to cut the curve repre- sent the amount of the change. C(i), changes of conductivity found shortly after the closure and during the flow of a moderately strong current. Conductivity greatly reduced around kathode; little affected at anode. C(2), changes of conductivity during flow of a very strong current. Conductivity reduced both in anodic and kathodic regions, but less in the former. C1, changes of conductivity just after opening a moderately strong current. Con- ductivity greatly reduced in region which was formerly anodic; little affected in region for- merly kathodic. ishes, until the anode as well as the whole intrapolar re- gion is blocked. After the opening of the current, the relation between kathodic and anodic conductivity is reversed, for now the post- kathodic region conducts the nerve-impulse relatively bet- ter than the post -anodic. It will be seen that these changes of conductivity up- on the whole run parallel to the (secondary) changes of excitability, depression of excitability correspond ing to depression of conductivity, and vice versa. With the relatively strong currents re- quired to produce decided effects on the conductivity, any preliminary change in the same sense as the (so-called primary) effects on the excitability (increase at kathode, decrease at anode) might be expected to be fleeting, and therefore less easy to detect. The above facts serve to explain the manner in which the effects of stimulation of a nerve with the constant current vary with the strength 788 NERVE and direction of the stream. These effects, so far as the contraction of the muscles supplied by the nerve is concerned, have been formulated in what has been somewhat loosely termed the law of contraction. In this formula the direction of the current in the nerve is commonly dis- tinguished by a thoroughly bad but now ingrained phraseology, as ascending when the anode is next the muscle, and descending when the kathode is next the muscle. Current. Ascending. Descending. M. B. M. B. Weak - C C Medium - C C C C Strong - C C — Here M means ' make,' B, ' break,' of the current; C means ' con- traction follows.' The explanation generally given is as follows: Wherever there is an increase of excitability sufficiently rapid and sufficiently large, stimula- tion is supposed to take place; where there is a fall of excitability, stimulation does not occur. Accordingly, at closure the kathode stimu- lates— the anode does not; while at opening, the anode, at which the depressed excitability jumps up to normal or more, is the stimulating pole; the kathode, at which it declines to normal or under it, is inactive. With a weak current, (i) contraction only occurs at make, and (2) the direction of the current is indifferent. The explanation of the first fact is that the make is a stronger stimulus than the break, and when £he current is weak enough the break is less than a minimal stimulus. No sensible change of conductivity is caused by weak currents, which suffices to explain (2). With a ' medium ' current, contraction occurs at make and break with both directions. Heie the break excitation is effective as well as the make. With anode next the muscle (ascending current), there is, of course, nothing to prevent the opening excitation, which starts at the anode, from passing down the nerve and causing contraction ; and since there is no block around the anode or in the intrapolar region with ' medium ' currents, there is nothing to keep the closing (kathodic) excitation from reaching the muscle too. With the kathode next the muscle (descending current), the closing excitation, which starts from the kathode, has no reigon of diminished conductivity to pass through, nor has the opening (anodic) excitation, for the kathodic block, caused by moderately strong currents, is removed as soon as the current is broken. With ' strong ' currents there are only two cases of contraction out of the four, just as with ' weak,' but for very different reasons. There is a break-contraction with ascending, and a make-contraction with descending current. With ascending current the anode is next the muscle, and the break-excitation starting there has nothing to hinder its course. The make-excitation, although as strong or stronger, has to pass through the whole intrapolar region and over the anode, and here the conductivity is depressed and the nerve-impulse blocked. With descending current the kathode is next the muscle, and there is no hindrance to the passage of the make-excitation. The break-excitation, however, has to traverse the whole intrapolar region, and this does not THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 789 at once, after a strong current, become passable. The break-excitation, accordingly, cannot get through to the muscle. A formula similar to the law of contraction has been shown to hold for the inhibitory fibres of the vagus (Bonders) , ' inhibition ' being substituted for ' contraction.' There is also some evidence that a similar law obtains for sensory nerves. It is not difficult to see that with currents of brief duration the break follows so quickly on the make that interference of their opposed effects may occur. This is the reason — or, at least, one reason — why, ; bove a certain frequency, a muscle or nerve ceases to respond to all of a series of rapidly recurring electrical stimuli (p. 758). It is also the reason why, with single very brief stimuli, a greater current intensity must be employed in order to cause excitation than when the duration of the stimulating current is greater (Wood worth, Lucas). The Law of Contraction for Nerves ' in Situ.' — When a nerve is stimu- lated without previous isolation — in the human body, for instance, through electrodes laid on the skin — the current will not enter and leave it through definite small portions of its sheath, nor will it be possible to make the lines of flow nearly parallel to each other and to the long axis of the nerve, as is the case in a slender strip of tissue when there is a considerable dis- tance between the electrodes. On the contrary, when, as is usual in electro-therapeutical treatment, a single electrode — say, the positive — is placed over the position of the nerve, and the other at a distance on some convenient part of the body, the current will enter the nerve by a broad fan of stream-lines cutting it more or less obliquely, and pass out again into the surrounding tissues; so that both an anode (surface of en- trance) and a kathode (still larger surface of exit) will correspond to the single positive pole. Similarly, the single nega- tive electrode will correspond to an anodic surface where the now narrowing sheaf of lines of flow enters the nerve, and a smaller kathodic surface, where they emerge. Even if the two elec- trodes were on the course of the nerve, the stream-lines would still cut it in such a way that each electrode would correspond both to anode and kathode (Fig. 276). It is impossible under these circumstances to take account of the direction of a current in a nerve, or to connect direction with any specific effect. When we place one of the electrodes over the nerve and the other at a distance, the law of contraction only appears in a disguised form; for since a kathode and an anode exist at each pole, there is, with a current of sufficient strength (' strong current '), excitation at each, both at make and break. The negative make contraction is, however, stronger than the positive, for the excitation corresponding to the latter arises at the secondary kathodic surface, where the sheaf of current-lines spreading from the positive electrode passes out of the nerve. Now, this is much larger than the primary kathodic surface, through which the narrow wedge of stream-lines passes to reach the negative electrode, and the current density at the latter is accordingly much greater. The Fig. 276. — Diagram of Lines of Flow of a Current passing through a Nerve. A, an isolated nerve ; B, a nerve in situ. Secon- dary anodes ( + ) are formed where the current re-enters the nerve below the negative elec- trode after passing through the tissues in which it is embedded and secondary kathodes ( - ) where the current passes out of the nerve into the surrounding tissues below the positive elec- trode. 790 NERVE positive break-contraction is, for a similar reason, stronger than the negative. With a ' weak ' current, the only contraction is a closing one at the kathode; with a ' medium ' current there are both opening and closing contractions at the positive pole, and a closing but no opening con- traction at the negative (Practical Exercises, p. 846). Conductivity of Nerve. — The disturbance which is called the nerve-impulse, once set up, is propagated along the fibres. Are the changes in the nervous substance involved in the initiation of the disturbance at a given point identical with those involved in its transmission from one point to the next, or are they different ? This is a question which has been much discussed, and many attempts have been made to prove that the two processes can be dissociated by acting on nerves with substances like carbon dioxide, ether, and alcohol, which gradually suspend their functions, by cutting nerves off from the circulation and allowing them to die gradually, by depriving them of oxygen and in other ways. Many of the results obtained from such experiments seem at first sight to be favourable to the view that the local change is different from the propagated disturbance. Nevertheless, careful examination of the results on which such statements are based indicates that none of them supplies a crucial test of the question at issue. For example, when a stretch of frog's sciatic nerve is treated with ether or another of the narcotics which act on nerve, and the strength of stimulus determined which is necessary to elicit a contraction when applied to an untreated portion more remote from the muscle than the narcotized area, this strength is found, for some time after the application of the narcotic, to be just the same as it was previous to the application. ' The conductivity ' of the narcotized stretch appears to be unaltered. On the othxar hand, the stimulus, when applied within the narcotized region, must be strengthened, and the narcotic appears to have diminished the ' excitability ' of the nerve. When the narcotic has acted for a longer time, the reverse effect appears. No stimulus, however strong, applied to the central non-narcotized stretch will cause a contraction, the ' conductivity ' having been apparently totally abolished by the narcotic, whereas a strong stimulus applied in the narcotized region will still cause a contraction, showing that ' excitability ' still remains. As to the facts there is general agreement; it is their interpretation which is in doubt. Now, it has been shown that in passing along a narcotized nerve the propagated disturbance diminishes in pro- portion to the length of nerve which it has to traverse. Accordingly in the second stage of narcosis the failure of the stimulus applied to the upper part of the nerve to elicit a contraction is explained most naturally as due to the extinction of the disturbance, which must pass through the whole narcotized region, whereas the dis- turbance set up by stimulation in that region succeeds in reaching THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 79i the muscle, since it has a shorter stretch of narcotized nerve to traverse. This experiment, then, in reality affords no proof that excitability and conductivity can vary independently. Facts are also known, to which allusion need not be made here, but which greatly modify the ordinary interpretation of the experimental results obtained in the first stage of narcosis, and upon the whole it may be said that these direct methods of determining the question have failed to yield a satisfactory answer. Indirect evidence exists, however, that the local process initiated by stimulation is not quite the same as the process involved in the propagation of the disturbance (Lucas); Thus, a brief current too weak to set up a propagated disturbance, nevertheless causes so me change at the pomt of stimulation, since a second current, also too weak to be effective by itself, will, when thrown in a short time after the first, cause a disturbance which is propagated along the nerve. There is good reason to believe that the change produced by the first current is not the same in kind as that produced by the second, only weaker, but that it is inherently different in quality. Above all, it is a local change incapable of being itself propagated, but constituting the necessary prelude to the starting of the propagated disturbance. Lucas has called this preliminary local effect the ' local excitatory process.' There are many facts which indicate that the capacity of different functional or anatomical groups of nerve-fibres for responding to stimu- lation and for conducting the nerve-impulse can be differently affected by one and the same influence. For example, pressure abolishes the conductivity of sensory fibres sooner than that of motor fibres. Cocaine locally applied to a nerve diminishes or abolishes its con- ductivity, according to the dose. It exercises a selective action as regards nerve-fibres of different kinds, picking out and paralyzing sensory fibres before motor; vagus fibres conducting upwards before those conducting downwards, vaso-constrictors before vaso -dilators, and broncho-constrictors before broncho-dilators (Dixon). The conduction or propagation of a definite disturbance or impulse is a phenomenon not confined to nervous tissue. It is also characteristic- ally seen in muscle, although there the mechanical effect which con- stitutes the normal response to the arrival of the propagated disturbance obtrudes itself and tends to divert attention from the latter. It is unlikely that the conduction process in muscle should be essentially different from that in nerve, and in muscle, as in nerve, there is evidence that it is associated with only a small, perhaps not even a detectable, liberation of heat. The main heat-production in muscle is essentially a feature not of conduction, but of contraction. Con- duction in muscle can be completely dissociated from the contraction process in various ways. For example, if a portion of a muscle is immersed for a time in distilled water, so-called water rigor ensues, and the altered muscle has lost the power of contraction. It will never- theless conduct the impulse which on reaching the unaltered part of the muscle causes it to contract normally. Double Conduction. — When a nerve (or muscle) is stimulated artificially, the excitation runs along it in both directions from the 792 NERVE point of stimulation; so that nerve-fibres which in the intact body are afferent can conduct- impulses towards the periphery and efferent fibres can conduct impulses away from the periphery. In the normal state, however, double conduction must seldom occur, for efferent fibres are connected centrally, and afferent fibres peripherally, with the structures in which their natural stimuli arise. In general, too, an impulse, if it did pass centrifugally along an afferent fibre, would not give any token of its existence, for the peripheral organ would not be able to respond to it; and there is no ground for assuming that the central mechanisms connected with afferent fibres are better fitted to answer such foreign and un- accustomed calls as impulses reaching them along normally efferent nerves. There is good evidence that muscular excitation is not carried over to the motor nerve- fibres; in other words, the wave of action flows from the nerve to the muscle, but cannot be got to flow backwards. Excitation of the central end of an efferent (anterior) spinal root is not transferred to the corresponding afferent (posterior) root, the connection between the efferent and afferent neurons presenting the character of a physiological ' valve,' which permits impulses to pass only in one direction. We have seen that vaso-dilator impulses possibly pass out to the limbs over fibres which, morphologically speaking, are afferent fibres (p. 181). And we shall see that a nutritive influence is exerted over the afferent fibres of the spinal nerves by the ganglion cells of the posterior root ganglia (p. 796), an influence which must spread along these fibres in the opposite direction to that of the normal excitation. The best proofs of double conduction in nerves, with artificial stimu- lation, are: (i) The propagation of the negative variation or action current in both directions. This holds for sensory as well as for motor fibres, as du Bois-Reymond showed on the posterior roots of the spinal nerves of the frog and the optic nerves of fishes. (2) Stimulation of the posterior free end of the electrical nerve of Malapterurus (p. 841) causes discharge of the electric organ, although the nerve-impulse travels nor- mally in the opposite direction. (3) If the lower end of the frog's sartorius is split into two, gentle stimulation of one of the tongues causes contraction of individual fibres in the other. This is supposed to be due to conduction of the nerve-impulse up a twig of a nerve-fibre distributed to the one tongue, and down another twig of the same fibre going to the other tongue. A similar experiment can be done on the gracilis of the frog. This muscle is divided by a tendinous inscription into two parts, each supplied by a branch of a nerve which divides after entering the muscle. Stimulation of either twig is followed by contraction of both parts of the muscle (Kuhne). Bert's much-quoted experiment on the rat is valueless as a proof of double conduction. He caused union of the point of the tail with the tissues of the back, then divided the tail at the root, and found that stimulation of what was now the distal end caused pain. From this he concluded that the sensory fibres of the ' transposed ' tail conducted in the direction from root to tip. But the conclusion is not warranted, for sensation disappeared in the tail after the section, and did not THE NERVE-IMPULSE OR PROPAGATED DISTURBANCE 793 return till some months later, when the nerve-fibres, after degenerating, would have been replaced by new sensory fibres growing down from the dorsal nerves (Ranvier). For a similar reason the so-called union of the peripheral end of the cut hypoglossal nerve (motor) with the central end of the cut lingual (sensory) proves nothing as to double conduction, nor as to the possibility of motor nerves taking on a sensory function. For while sensation is after a time restored in the affected portion of the tongue, this is due to the growth of sensory fibres from the central stump of the lingual down through the degenerated hypoglossal, and not to the conduction upwards of sensory impulses by the motor fibres of the latter. Every fibre of a nerve is physiologically isolated from the rest, so that an impulse set up in a fibre runs its course within it, and does not pass laterally into others (law of isolated conduction). In connection with this physiological fact there is the anatomical fact that nerve- fibres do not normally branch in the trunk of a peripheral nerve. (But see p. 802.) It has, however, been shown that bifurca- tion of nerve-fibres may occur in the spinal cord (Sherrington). The axis-cylinder of a peripheral nerve-fibre only begins to branch where complete isolation of function is no longer required, as within a muscle. The experiment of Kuhne on double conduction, men- tioned above, shows that an excitation set up in one twig or one fibril of an axis-cylinder which has branched can spread to the rest. Velocity of the Nerve- Impulse. — We have said that the nerve- impulse travels with a measurable velocity. It is now time to describe how this has been ascertained (p, 818). For motor fibres the simplest method is to stimulate a nerve successively at two points, one near its muscle, the other as far away from it as possible, and to record the contractions on a rapidly-moving surface (pendu- lum or spring myograph) (p. 746). The apparent latent period of the curve corresponding to the nearer point will be less than that of the curve corresponding to the point which is more remote, by the time which the impulse takes to pass between the two points. The distance between these points being measured, the velocity is known. Helmholtz found the velocity for frog's nerves at the ordinary temperature of the air to be a little under, and for human nerves, cooled so as to approximate to the ordinary temperature, a little over 30 metres per second. For observations on man the contraction curves of the flexors of one of the fingers or of the thumb may be recorded, first with stimulation of the brachial plexus at the axilla, and then with stimulation of the median or ulnar nerve at the elbow. Probably at the same temperature there is little difference in the rate of transmission in the nerves of warm- blooded and cold-blooded animals, but temperature has a con- siderable influence (p. 782). By cooling a frog's nerve Helmholtz reduced the rate to ^ of its value at the ordinary temperature. In the human arm he found a variation from 30 to 90 metres per second, according to the temperature, 50 metres 794 NERVE being about the normal rate. This is greater than the speed of the fastest train in the world. According to Piper's recent measurements the velocity in human medullated nerve is even greater than Helmholtz concluded, about 120 metres a second under ordinary conditions. The rate is independent of the intensity of the excitation. The velocity with which the negative variation is propagated (p. 824) is the same as that of the nerve-impulse. In sensory nerves there is no reason to believe that the velocity of the nerve-impulse differs from that in motor nerves, but experiments on man really free from objection are as yet wanting. The usual method is to stimulate the skin first at a point distant from the brain, and then at a much nearer point. The person experimented on, as soon as he feels the stimulation, makes a signal, say, by closing or opening with the hand a current connected with an electric time- marker, writing on a moving surface. There is, of course, a measurable interval between the excitation and the signal, and this being in general longer the more remote the point of stimulation is from the brain, it is. assumed that the excess represents the time taken by the nerve-impulse to pass over a length of sensory nerve equal to the difference in the length of the path. But there is this difficulty, that the propagation of the impulse from the point of stimulation to the brain is only one link in the chain of events of which the signal marks the end. The impulse has first to be transformed into a sensation, and then the will has to be called into action, and an impulse sent down the motor nerves to the hand. And while the time taken by the excitation in travelling • up and down the peripheral nerve-fibres is probably fairly constant, the time spent in the intermediate psychical processes is very variable. SECTION II. — CHEMISTRY, DEGENERATION, AND REGENERATION OF NERVE. Chemistry of Nerve. — Our knowledge of this subject is still scanty; and most of what we do know has been obtained from analyses, not of the peripheral nerves, but of the white matter of the central nervous system Proteins are present, especially in the axis-cylinder. The proteins of nervous tissue include two globulins, one coagulated by heat at 47° C., the other at 70° to 75° C., and a nucleo -protein coagulating at 56° to 60° C. Very important constituents are certain substances soluble in organic solvents, like benzol and ether, and comprising cholesterin, certain phosphatides (kephalin and lecithin], and certain cerebrins or cerebrosides. The cerebrins are glucosides containing nitrogen, but no phosphorus, and they yield a reducing sugar (galactose) on hydrolysis. In the nervous tissue there is also present, according to some authorities, a compound called protagon. Others consider it a mere mixture of phos- phatides and cerebrosides. The lipoids of nerve-fibres belong largely to the medullary sheath, but they are not confined to it, since non-medul- lated nerves also yield a considerable quantity of lipoids (11*5 per cent, of the solids as against 46-6 per cent, for medullated nerves). Non- medullated nerves (splenic nerves of the ox) are distinguished from medullated nerves (human sciatic) by the high proportion of their total lipoids constituted by the phosphatides (kephalin and lecithin) and cholesterin. Thus, in non-medullated fibres 47 per cent, of the lipoid CHEMISTRY OF NERVE 795 extract consisted of cholesterin, and 23-7 per cent, of kephalin; while in the medullated fibres cholesterin made up only 25 per cent, of the extract, and kephalin 12-4 per cent. On the other hand, the cere- brosides are present, both relatively and absolutely, in much larger quantity in medullated than in non-medullated nerves. In both varieties of fibres kephalin, and not lecithin, is the chief phosphorus- containing body (Falk). The medullary sheath further contains a kind of network of a peculiar resistant substance, neurokeratin. The neurilemma consists of substances insoluble in dilute sodium hydroxide. Gelatin is obtained from the connective tissue which binds the nerve- fibres together. There may also be ordinary fat in the meshes of the epineurium connecting the bundles. Small quantities of xanthin, hypoxanthin, and other extractives, can also be obtained from nerve. According to Halliburton's analyses, the water in sciatic nerves amounts to 65*1 per cent., and the solids to 34-9 per cent. The proteins make up 29 per cent, of the solids. For an analysis of the white matter of the brain, see Chapter XVI. Nerve-cells contain no potassium, according to Macallum; and this is true both of the dendrites and the axons. In medullated nerves, how- ever, potassium compounds are present external to the axons, chiefly at the nodes of Ranvier (Frontispiece) and in the neurokeratin framework of the sheath. The only chemical difference between living and dead nervous tissue which has been made out with any degree of certainty is that the former is neutral or faintly alkaline, and the latter acid, in reaction to such indicators as litmus. This is especially true of the grey matter of the central nervous system, although the white matter also is often found acid. The change of reaction is due to the accumulation of lactic acid. Such a change has not hitherto been clearly demonstrated in peripheral nerves, either after death or after prolonged stimulation. The (non- medullated) splenic nerves of the dog, even after stimulation for six hours, never became acid (Halliburton and Brodie). Degeneration of Nerve. — Nerve-fibres are ' bound in the bundle of life ' with the nerve-cells from which their axis-cylinders arise; the connection between cell and axon once severed, the nerve- fibre dies inevitably. This is an illustration of a general law that no portion of a cell can live once it is separated from the nucleus. We shall see later on that changes also occur in the nerve-cell whose axon has been divided from it, although they are of a different nature (rather a slow atrophy than an acute degeneration), and do not necessarily lead to the destruction of the cell. We must regard the neuron not only as a morphological unit, a single cell from nucleus to remotest end-brush, but also as a functional and nutritive unit, the fortune of any portion of which is not in- different to the rest. Thus, when a man's arm is amputated the arm fares worse than the man, for the arm dies. But the man is not unaffected. He lives, but he suffers much temporary disturb- ance and some permanent loss. What is left of him is not quite the same as it was. The acute changes that occur in severed nerve- fibres are most conveniently studied in the peripheral nerves, although essentially similar phenomena take place also in the fibres of the central nervous system. 796 NERVE A spinal nerve is composed of efferent fibres whose cells of origin are in the grey matter of the anterior horn, and afferent fibres whose cells of origin are in the posterior root ganglion. When such a nerve is cut below the junction of its roots, muscular paralysis and impairment of sensation at once follow in the region supplied by the nerve; but for a time the nerve remains excitable to direct stimulation. The excitability gradually diminishes, and in a few days is completely gone. If portions of the nerve distal to the lesion are examined at different periods after section, a remarkable process of degenera- tion (commonly spoken of as Wallerian degeneration) is seen to be going on. In the medul- lated fibres this begins on the second or third day with a swelling of the axis-cylinder, which breaks up into detached pieces (fragmentation), and as- sumes a granular appearance. The medullary sheath also under- goes fragmentation at the lines of Lantermann, and a little later separates into clumps and drop- lets of myelin. The nuclei under the neurilemma increase in size, proliferate by mitosis, and in- sinuate themselves between the fragments of the medullary sheath and axis-cylinder, which ultimately disappear, leaving the nerve- fibre represented only by a kind of mummy of connective tissue, in which the neurilemma with its abnormally numerous Fig. 277. — Degeneration of Nerve-Fibres after Section (Barker, after Thoma). I, normal fibre; II, degenerating fibre; III, further stage of degeneration; S, neurilemma ; m, medullary sheath; A, axis-cylinder; L, Lantermann's line or cleft; R, node; mt, drops of myelin; a, remains of axis-cylinder ; w, prolifera- ting cells of neurilemma. nuclei can still be recognized. The fragmentation of the myelin sheath is not dependent upon the proliferation of the nuclei, since it occurs also in nerves removed from the body and kept under con- ditions in which the nuclei do not proliferate (Feiss and Cramer). The protoplasm around the nuclei of the neurilemma also increases in amount, and undergoes other changes, which will be more particularly referred to in describing the regeneration of nerve. The degenerative process begins near the cut end, and extends gradually to the peri- DEGENERATION OF NERVE 797 phery, and more rapidly in warm- than in cold-blooded animals. At any rate, that is the interpretation generally given to the fact that at a given period after section the changes — especially the breaking- up of the myelin — are more pronounced near the proximal end of the peripheral stump. In a mammal degeneration is far advanced in a fortnight, although the last remnants of the myelin may not be absorbed for months. In the degenerated nerve (cat's sciatic) the percentage of phosphorus undergoes a diminution from about the third day. About the eighth day the loss of phosphorus — i.e., of the phosphatides (lecithin, kephalin) — is markedly accelerated, coinciding with the appearance of a strong Marchi* staining re- action. By the twenty-ninth day the degenerated nerve is prac- tically devoid of phosphorus. A progressive increase in the water and a diminution in the total solids also culminate about the same time (Mott and Halli- burton). In the portion of the nerve-fibre still connected with the nerve-cell the degeneration only extends as far back as the next node of Ranvier, and seems to be due to the direct effect of the injury. In non-medullated fibres, such as the fibres arising from the cells of the superior cervical ganglion (Tuckett), the degeneration is confined to the axis-cylinders. It begins in about twenty-four hours after section, and the loss of excitability and conductivity is complete by the fortieth hour. It follows from what has been said as to the position of the cells of origin of the root fibres of the spinal nerves that section of the anterior root causes degeneration on the peripheral, but not on the central side of the lesion, f Only the anterior root fibres in the mixed nerve degenerate. Section of the posterior root above the ganglion causes degeneration of the central stump, but not of the portion still connected with the ganglion, nor of the posterior root fibres below the ganglion or in the mixed nerve. Section of the posterior root below the ganglion causes degeneration of the fibres of the root below the section and in the mixed nerve, but not above it. * The chief constituents of Marchi's solution are potassium bichromate and osmic acid. It stains medullated nerve-fibres black in the earlier stages of degeneration. | A few fibres in the peripheral stump of the anterior root do not degenerate, and a few fibres in the central stump do. These are the 'recurrent fibres,' whose course is described on p. 892. Fig. 278. — Degeneration of Spinal Nerves and their Roots after Section. The shading shows the degenerated portions. 798 NERVE Regeneration of Nerve. — Degeneration of nerve is followed, if its divided ends are not kept artificially apart, by a process of re- generation, already distinct under favourable conditions in from three to four weeks after the section, and indeed in some cases commencing as early as the second week. This consists in the outgrowth of new axis-cylinders, in the form of fine fibres, from the ends of the divided axis-cylinders of the central stump of the nerve. These push their way into and along the degenerated fibres, ultimately acquire a medullary sheath, and develop into complete nerve-fibres, restoring first sensation, and later 'on volun- tary motion, to the paralyzed part. The process needs several months for its completion, even in warm-blooded animals. It takes place under the influence of the nucleated portion of the neuron (the cell-body), and is never completed if the peripheral and central portions of the nerve are permanently separated by a substance through which the new axis-cylinders cannot grow or by a gap too wide for them to bridge over. When the 'cut ends of the nerve are carefully sutured together, the conditions for com- plete and speedy regeneration are rendered more favourable — a fact which finds its application in the surgical treatment of injured nerves. The cycle of chemical changes described in the degenerating nerve is retraced in the reverse order. In the cat's sciatic the first sign of the return of the phosphorus was seen with the beginning of the normal myelin reaction about the sixtieth day after section. At the one-hundredth day the phosphorus content was almost as great as that of the normal nerve (a little under i per cent, of the solids for the regenerated, as compared with a little over i per cent, for the normal nerve). It is not as yet well understood how the regenerating fibres are directed in their growth, so that they join their centres to the appro- priate end-organs without mistake. That they have a high capacity for finding their way is indicated by the results of cross-suturing such nerves as the median and ulnar — i.e., of uniting the central end of the one with the peripheral end of the other. Howell and Huber found that after this operation in the dog, both co-ordinated volun- tary motion and sensation returned in large measure in the parts supplied by the nerves. Here the motor fibres of the median nerve must, of course, have made connection with muscles previously supplied by the ulnar, being guided to them along the nerve-sheaths of the latter. Doubtless the old nerve-sheaths serve to some extent as mechanical guides by offering to the new axons a path of least resistance. And when a nerve-trunk containing motor and sensory fibres is simply crushed so as to destroy all physiological continuity, but is not cut, no distortion of the motor and sensory ' patterns ' of the nerve — in other words, no ' straying ' of the fibres from their old paths — can be detected on regeneration. When the REGENERATION OF NERVE 799 nerve is cut and then sutured, a certain amount of distortion of the pattern is inevitable. The mechanical apposition of central and peripheral stumps is, of course, much more nearly perfect in the crushed nerve than in the cut nerve, however exact the suturing may be (Osborne and Kilvington). Yet, even after crushing or liga- tion of nerves, or after section and suturing, the regenerating fibres do not pass straight through the scar tissue from the central to the peripheral stump, but cross and mingle, ap- parently in the most inextricable confusion (Feiss) (Fig. 279). This is due to the prolifera- tion of cells in the scar which run in all direc- tions, and show no signs of following the Fig. 279. — Regenerating Fibres crossing in the Scar after Ligation of a Dog's Sciatic Nerve 165 Days previously. Weigert-Pal stain. Drawn under oil- immersion (Feiss). parallel arrangement of the nerve-sheaths in the central or the distal segment. These being formed before the regenerating nerve- fibres, the latter must necessarily grow also in all direc- \Sk±.fyMt*\ tions in the scar. J^A This, however, is a ,NeUro«vo.i ~*-,'''^.-%r*' '• local phenomenon. Beyond the scar the , 4~^;-->--,r_ «:_ .-'/• arrangement of the IPJ \ ,.y regenerating axis- ',/ cylinders recovers ^ its regularity, and H^^ the amount of dis- tortion of the nerve pattern, as indicated by histological ex- amination and functional tests, is by no means so great as the com- plete effacement of the pattern in the scar might appear to promise. That the degenerated peripheral stump directs the growth of the axons from the central stump in some other than a merely mechanical way is evident from the experiments of Langley on Ext. Fig. 280. — Semidiagrammatic Representation of Longi- tudinal Section through Neuroma or Scar produced by ligating the Sciatic Nerve with Catgut, and crushing it with a hasmostat just above its Division into the Ex- ternal and Internal Popliteal Nerves. Weigert-Pal preparation (Feiss). 8oo NERVE regeneration of the cervical sympathetic in the cat after section below the superior cervical ganglion. The nerve contains fibres of various functions which reach it from the upper thoracic nerves. The anterior roots of the first and third thoracic nerves supply the cervical sympathetic mainly with fibres which end in the ganglion around cells that give off dilator fibres for the pupil. The fibres connected with the cells in the ganglion which send vaso-motor fibres to the vessels of the ear are for the most part contained in the anterior roots of the second and fifth thoracic nerves; and the fibres connected with the cells that give origin to the pilo-motor fibres for the hairs of the face and neck in the anterior roots of the fourth to the seventh. Stimulation of any one of the upper thoracic roots accordingly causes a specific effect, which, according to Langley, is in general the same after regeneration as before section of the cervical sympathetic. We must assume, therefore, that each regenerating fibre seeks out either the ganglion cell with which it was originally connected, or one belonging to the same class. No mere mechanical guidance of the growing axons by the old neuri- lemmas will suffice to explain this selective growth. It is necessary to postulate, in addition, an attraction of a chemical or physico- chemical nature (chemiotaxis), dependent upon a specific relation between the new axons and the scaffolding of the peripheral stump or the ganglion cells. But it is not possible at present to form any very precise conception of the properties on which the chemiotactic phenomena depend. And the specificity is not an absolute one. Under certain conditions these pre-ganglionic nerve-fibres (that is to say, nerve-fibres running from the spinal cord to end around the sympathetic ganglion cells) can form connections with nerve-cells of a different class — e.g., pupillo-dilators with cells whose axons end in the erector muscles of the hairs. Further, after section of the sympathetic above the superior cervical ganglion, the post- ganglionic nerve-fibres (i.e., the fibres coming off from the cells of the ganglion) may also, if the opportunity be favourable during regeneration, exchange their old end-organs for new ones ; pilo- motor fibres, for instance, finding their way into the iris and becoming pupillo-dilators. After excision of the superior cervical ganglion, the cervical sympathetic does not recover its function. Accordingly the pre-ganglionic fibres cannot form direct functional connection with the post-ganglionic fibres, but can become connected with them only indirectly through the ganglion cells: Nor can efferent post-ganglionic fibres achieve regenerative union with a cerebro- spinal (somatic) motor nerve, although they can themselves re- generate, as has been shown, e.g., in the case of the vaso-constrictors of the limbs. On the other hand, union easily takes place between pre-ganglionic fibres and efferent somatic fibres, and vice versa. For example, the cervical sympathetic can unite with the phrenic 8oi nerve, and cause contraction of the diaphragm, or with the recurrent laryngeal nerve, and cause movement of the vocal cords, or with the spinal accessory, and cause contraction of the sterno-mastoid muscle. Conversely, the phrenic nerve, when united with the cer- vical sympathetic, can, when stimulated, produce the usual effects observed on exciting the latter nerve (Langley and Anderson). Central and Autogenetic Theories of Regeneration. — Although the establishment of connection with the central end of the cut nerve is necessary for complete regeneration, it must not be supposed that no share whatever is taken in the process by the peripheral stump. Even while it remains completely isolated from the central nervous system, changes occur which are often described as the third or final stage of degeneration, but which are more correctly interpreted as forming a stage in the regenerative .cycle. Spindle-shaped cells or fibres with elongated nuclei make their appearance, produced by the proliferation of the nuclei of the primitive sheath already described, and the increase of the proto- plasm in which these nuclei are embedded. These so-called axial strand fibres or this fibrillated protoplasm may appear long before the remains of the degenerated axis-cylinder and myelin sheath have been completely removed. It is generally acknowledged that in the adult they do not develop beyond this, so long as the peri- pheral portion of the nerve remains completely isolated, but neither do they disappear even after a very long interval. When strict precautions against union with other nerve-trunks were taken, the radial nerve of an adult cat was found in this resting-stage nearly a year and a half after division, and the same was true after two years and a half in a nerve divided in a human being. The fibres are incapable of being excited or of conducting nerve impulses. The precise relation between these axial strand fibres of the peri- pheral stump and the myelinated fibres found there after regenera- tion has been much debated. All are agreed that nerve-fibrils sprout from the central stump, and the weight of evidence is in favour of the long-accepted view that it is by the growth of these fibrils along the peripheral stump that the new axons are formed, and that all the changes in the distal portion of the nerve, however important for directing and perhaps sustaining the growth of the central fibrils, are subsidiary to this. But some maintain that the outgrowing central fibrils meet and unite with corresponding fibrils sprouting from the peripheral stump, and that the new axis- cylinders arise from the fibrils of the axial strand. It is said that very shortly after being brought into connection with the central portion of the same or of another nerve by careful suturing the spindle cells begin to lengthen, and form non-medullated fibres, like those of the sympathetic. Four weeks after union the afferent fibres, although still non-medullated, are capable of being stimulated 51 802 NERVE mechanically and electrically, and of conducting impulses towards the centre. In about eight weeks they become medullated, but at first are of small calibre (Head and Ham) . Bethe, the most strenuous defender of the inherent regenerative power of the isolated peri- pheral stump (autogenetic theory), has even stated that complete regeneration occurs in young animals in nerves entirely separated from their centres. There is no doubt that this result is due to some error of technique or of interpretation. The controversy turns largely upon the precautions judged necessary to prevent the ingrowth of central fibres. And while it is comparatively easy to make sure, by removing a large part of it, that the central end of the nerve under observation shall remain completely unconnected with the peripheral end, it is often a matter of the greatest difficulty to prevent the union of the distal stump with central fibres from other sources — e.g., from the nerves cut in the wound. Many of the results which seemed to favour the autogenetic theory were cer- tainly due to this cause. The most conclusive evidence in favour of central and against autogenetic regeneration, because the most direct and uncompli- cated, has been afforded by the demonstration that the development of axis-cylinders occurs in vitro in a suitable plasmatic medium, in the absence of any other elements than the nerve-cells from which they arise (Harrison). This observer, working with the medullary plates of tadpoles, in which the nerve-cells originate in the embryo, showed further that peripheral nerves do not develop when the nerve-centres are removed, and that the sheath-cells of Schwann are not essential to the growth of axis-cylinders, since in their absence the latter continue to grow and reach their normal length. It has also been proved that nerve-fibres grow out from pieces of the cere- bellum and spinal ganglia (Fig. 281) of young mammals when cul- tivated on clotted plasma outside of the body (Fig. 337, p. 857). Many fibres sprouting out from the spinal ganglia attain a length of more than half a millimetre in forty-eight hours, and their growth need not be accompanied by either neuroglia or connective tissue. When portions of normal peripheral nerve are incubated in plasma no growth of axis cylinders -is ever observed, nor does the Wallerian degeneration occurring in peripheral nerves incubated in Ringer's solution take place (F;g. 282). On the other hand, if nerves in which Wallerian degeneration has been produced by section are placed in plasma on the fifth day after section or later, growth of the elements of the sheath of Schwann may be seen, again unaccompanied by any growth of axis cylinders (Ingebrigtsen). A fact of great physiological interest, and also of practical impor- tance, in connection with the anastomosis of nerves for the relief of certain forms of paralysis, is the bifurcation of axons in regeneration, when the conditions are such that the axons of the central stump are offered more than one path along which to regenerate. If, for instance, a limb nerve-trunk containing motor fibres is cut, and its central end REGENERATION OF NERVE 803 sutured both to its own distal end and to the distal end of an adjacent nerve-trunk, the sum of the nerve-fibres in the two distal trunks after regeneration has occurred is greater than the number of fibres in the central stump (Kilvington). That this is due to splitting of axons is Fig. 281. — Two-day-old Culture of Spinal Ganglia of a Guinea Pig Six Days Old (Ingebrigtsen). Fig. 282- — Microphotograph of Nerve Fibres from the Sciatic Nerve of a Rabbit, incubated in Plasma for Twelve Days. Hardened in formalin and stained with hasmatoxylin. No trace of Wallerian degeneration (Ingebrigtsen). shown by the fact that an axon reflex (p. 913) can be elicited on dividing one of the distal trunks and stimulating its central end after complete separation of the proximal or parent stem from the central nervous system. Even when the second path offered to the regenerating motor axon is a sensory path, bifurcation of the axon occurs, one branch 8o4 NERVE passing down along the previous motor path to its proper muscular termination, and the other passing down the sensory path. Although there is no evidence that efferent fibres can unite with afferent fibres, a degenerated afferent path can therefore serve as a chemiotactic scaffold- ing or guide for the growth of regenerating motor axons, though not such an efficient one as a degenerated motor path. Sensory fibres, however, cannot regenerate along motor paths or make functional union with the receptive substance of skeletal muscle. Regeneration of the fibres of the central nervous system either does not in general occur, or is exceedingly difficult to realize. This lends support to the doctrine of the importance of the neurilemma in regeneration, since its elements are scantily developed in the fibres of the brain and cord (p. 860) . Regeneration of the fibres which proceed from the cells of the spinal ganglia along the posterior roots into the cord may take place after the roots have been cut, so that the normal reflexes through the res- piratory, cardiac, and vaso-motor centres may be once more obtained. Degeneration of Muscle. — Experimental section or, in man, traumatic division or compression of a nerve leads not only to its degeneration, but ultimately, if regeneration of the nerve does not take place, to degeneration of the muscles supplied by it as well. The muscle- fibres dwindle to a quarter of their normal diameter; the stripes disappear; the longitudinal fibrillation fades out; and at length only hyaline moulds of the fibres are left, filled, and separated by fatty granules and globules and surrounded by engorged capil- laries. Amidst the general decay, the muscular fibres of the terminal ' spindles ' with which the afferent nerves of muscles arc connected alone remain unchanged (Sherrington). Certain dis- eases of the cord which interfere with the cells of the anterior horn cause degeneration of motor nerves, and ultimately of muscles. The motor nerve-endings degenerate sooner than the sensory. Both may, under suitable conditions, regenerate (Huber). Reaction of Degeneration. — Muscles whose motor nerves have been separated from their trophic centres show, when a certain stage in degeneration has been reached, a peculiar behaviour to electrical stimulation, called the ' reaction of degeneration.' To the constant current the muscles are more excitable, and the contraction slower and more prolonged than normal. When a current, either constant or induced, is passed through a normal muscle, the muscular fibres may be stimulated either directly, or indirectly through the intramuscular nerves. Under ordinary conditions the nerves respond more readily than the muscular fibres, especially to momentary stimuli like induction shocks, and therefore the so-called direct stimulation of uncurarized muscle is as a rule an indirect stimulation. When the muscle is curarized and the nerves thus eliminated, the excitability to induced currents is found to be diminished. The same is the case in a muscle which exhibits the reaction of degeneration after section of its motor nerve, only the loss of excitability to induced currents is greater, and may even be complete. The common statement that the closing anodic contraction is stronger than the closing kathodic — the opposite of the ordinary law — is subject to so many exceptions that it has no diagnostic value. The nerves are inexcitable either to constant or induced currents. The reaction of degeneration is only obtained from paralyzed muscles when the paralyzing lesion is situated in the cells of the anterior TROPHIC NERVES 805 horn irom which the motor nerves take origin, or below that level. Accordingly, it is sometimes of use in localizing the position of a lesion. For instance, a group of muscles might be paralyzed by a lesion in the grey matter of the brain or in the nerve-fibres connecting this with the grey matter of the anterior horn of the cord, or in the grey matter of the anterior horn itself, or in the peripheral nerve-fibres leading from this to the muscles. In the first two cases the reaction of degeneration would be absent, although the muscles, if the lesion was of long standing, would be atrophied to some extent; in the last two there would be acute atrophy of the muscles, and the reaction of degeneration would be obtained. Trophic Nerves. — There is no question that nerves exert a very important influence upon the nutrition of the parts supplied by them, in influencing the specific function of those parts. So that in this sense all nerves are trophic nerves. The fact that the proper nutrition of nerve-fibres and striated muscular fibres is dependent on their connection with nerve-cells has been by some writers generalized into the doctrine that all tissues are provided with ' trophic ' nerves, which, apart from any influence of functional activity, regulate the nutrition of the organs they supply. But the evidence for this view, when weighed in the balance, is found wanting; and it may be said that up to the present no unequivocal proof, experimental or clinical, has ever been given of the existence of specific trophic fibres, anatomically distinct from other efferent or afferent nerves. It is true that in various diseases and injuries of the nervous system nutritive changes in the skin, and sometimes in the bones and joints, are apt to appear. But it is very difficult in such cases to disentangle the effects produced by accidental injuries acting on structures whose normal sensibility is lost or lessened, or whose circulation is deranged, from true trophic changes. The most that can be said is that there is some evidence that the power of the skin to resist injury, and the capacity of recovering from it, are diminished by interference with its nerve-supply, so that a large sore may result from a trifling lesion, and healing may be slow and difficult. Experimentally it has been found that division of the trigeminus nerve within the skull is sometimes followed by cloudiness of the cornea, going on to ulceration, and ulti- mately inflammation and destruction of the eyeball. Ulcers also form on the lips and on the mucous membrane of the mouth and gums ; and the nasal mucous membrane on the side corresponding to the divided nerve becomes inflamed. But in this case the sensibility of the eye is lost, and reflex closure of the eyelids ceases to prevent the entrance of foreign bodies. The animal is no longer aware of the contact of particles of dust or bits of straw or accumulated secretion with the conjunctiva, and makes no effort to remove them. The lips, being also without sensation, are hurt by the teeth, particularly as the muscles of mastica- tion on the side of the divided nerve are paralyzed, and decomposed food, collecting in the mouth, and inhaled dust in the nose, will tend still further to irritate the mucous membranes. There is thus no more need to assume the loss of unknown trophic influences in order to explain the occurrence of the ulcerative changes than there is to explain the production of ordinary bed-sores, bunions or corns on parts peculiarly liable to pressure. And, as a matter of fact, if the eye be 8o6 NERVE artificially protected, after section of the trigeminal nerve, the ophthalmia either does not occur or is much delayed. In man, too, a case has been recorded in which both the fifth and the third nerves were paralyzed. The eye was still shielded by the contraction of the orbicularis oculi supplied by che seventh nerve, as well as by the drooping of the upper eyelid that accompanies paralysis of the third. It remained perfectly sound for many months, till at length the tumour at the base of the brain which had affected the other nerves involved the seventh too. The eye was now no longer com- pletely closed; inflammation came on, and vision was soon permanently lost (Shaw). In another case a patient lived for seven years with complete paralysis of the fifth nerve, yet the eye remained free from disease and sight was unimpaired (Cowers). The so-called ' trophic ' effects following division of both vagi we have already discussed (p. 286) so far as they are concerned with the respiratory system. The degenerative changes sometimes seen in the heart are perhaps due to its being overworked in the absence of nervous restraint on its functional activity. The nutritive alterations in muscles and salivary glands after section of motor and secretory nerves seem to depend in part on functional and vaso-motor changes. In the paralyzed muscles nutrition is not only interfered with in consequence of their inactivity, as would be the case even if the paralysis were due to a lesion above the level of the anterior cornual cells, but the already poorly nourished fibres are continually pressed upon by the capillaries, which are dilated owing to the division of the vaso-motor nerves. The degeneration may also be in part ascribed to the loss of a reflex tonic influence exerted on the muscles by the spinal cord, through the ordinary motor nerves (p. 917). When all allowance has been made for these factors, the rapid and characteristic degeneration of the striated muscles, after their connection with the central nervous system is severed, is still inexplicable, except on the assumption that their nutrition is specially related to the integrity of their efferent nerves. In other words, it is necessary to suppose, not, indeed, that distinct trophic nerves exist for the muscles, but that an influence or impulses, which can be termed trophic or nutritive, do normally pass out to them from the spinal cord along their motor nerves. Section of the cervical sympathetic in young rabbits and dogs increases the growth of the ear and of the hair on the same side. But it is impossible to separate these consequences from the vaso-motor paral- ysis; and the same is true of the hypertrophy following section of the vaso-motor nerves of the cock's comb and of the nerves of the bones. After section of the superior laryngeal the vocal cord on the side of the section is at once rendered motionless, and remains so, but the muscles, notwithstanding their inaction, do not degenerate. And Mott and Sherrington have found that, although section of the posterior roots in monkeys is followed after a time (three weeks to three months) by ulceration over certain portions of the foot, no corresponding lesions occur in the hand. They believe, therefore, that the lesions are not due to the withdrawal of a reflex trophic tone, but are accidental injuries in positions specially exposed to mechanical or microbic insults. One of the best examples of interference with the proper nutrition of a part produced by a lesion in the nerves supplying it is an eruption (herpes zoster), limited to the skin supplied by the nerve-fibres coming from one or more spinal ganglia, and depending on an (infectious) inflammatory change in the ganglia. It has been suggested that the vesicles are formed either because the passage of afferent impulses normally concerned in the nutrition of the skin is interfered with or CLASSIFICATION OF NERVES 807 because the skin is bombarded by antidromic (p. 181) impulses dis- charged from the inflamed ganglia. But an alternative hypothesis is that a toxine spreads out along the nerves from the ganglia, just as in traumatic tetanus the toxine is known to pass in the opposite direc- tion along the nerves from the seat of injury to the central nervous system. Classification of Nerves. — Omitting the group of ' trophic ' nerves, and the even more problematical' thermogenic ' fibres (which some have supposed to preside over the production of heat, and therefore to assist in the regulation of the temperature of the body, but of whose existence as distinct and specific nerve-fibres with no other function there is not the slightest proof), peripheral nerves may be classified as follows: Centripetal or afferent fibres. ' Smellv Taste. Hearing. Sight. Touch (light touch). Pressure (perhaps in- cluding the nerves oi muscular sense). Warmth— Cold. Pain. Calibre of small arteries (pressor, depressor). Action of heart. Respiratory movements. Visceral movements. Glandular secretion. Ordinary skeletal muscles. Skeletal muscles Visceral , , Vascular / Vaso -constrictor vdocuidr .. i f^ * . . I Cardio-augmentor. Erector muscles of hairs (pilo-motor fibres). ( Visceral muscles 2. Inhibitory nerves for \ ( Vaso-dilator. I Vascular ,, \ Cardio-inhibi- 3. Secretory nerves [ tory. * It is not known whether the afferent portion of a reflex arc is always com- posed of fibres included in the first two categories, although undoubtedly in some cases it is. Centrifugal or efferent fibres. I. Nerves of special sensation 2. Nerves of general sensation 3.* Possibly nerves other than those included under i and 2, concerned in reflex changes in i. Motor nerves for PRACTICAL EXERCISES ON CHAPTERS XIII. AND XIV. i. Difference of Make and Break Shocks from an Induction Machine. — Connect a Daniell or other cell B (p. 724) with the two upper binding- screws of the primary coil P, and interpose a spring key K in the circuit. Connect a pair of electrodes with the binding-screws of the secondary coil (Fig. 283). 8o8 MUSCLE AND NERVE Electrodes can be very simply made by pushing copper wires through two glass tubes, filling the ends of the tubes with sealing-wax, and binding them together with waxed thread. The projecting points may be filed, and the nerve laid directly on them, or they may be tipped with small pieces of platinum wire soldered on. (a) Push the secondary away from the primary, until no shock can be felt on the tongue when the current from the battery is made or broken with the key. Then bring the secondary gradually up towards the primary, testing at every new position whether the shock is per- ceptible. It will be felt first at break. If the secondary is pushed still further up, a shock will be felt both at make and at break. From this we learn that for sensory nerves the break shock is stronger than the make. The same can easily be demonstrated for motor nerves and for muscle. (6) Smoke a drum and arrange a myograph, as shown in Fig. 287. But omit the brass piece F, and do not connect the primary through the drum, as there shown, but connect it as in Fig. 283. Pith a frog (brain and cord), and make a muscle-nerve preparation. To make a Muscle-Nerve Preparation. — Hold the frog by the hind legs back upwards; the front part of the body will hang down, making an angle with the posterior portion. With strong scissors divide the back- bone anterior to this angle, and cut away all the front portion of the body, which will fall down of its own weight. Make a circular incision at the level of the tendo Achillis, and another at the lower end of the femur, through the skin. The sciatic nerve must now be dissected out, as follows : Remove the skin from the thigh, and, holding the leg in the Fig. 283. — Arrangement of Coil for Single Shocks. left hand, slit up the fascia which connects the external and internal groups of muscles on the back of the thigh. Complete the separa- tion with the two thumbs. Cut through the iliac bone, taking care that the blade of the scissors is well pressed against the bone, otherwise there is danger of severing the sciatic plexus. Now divide in the middle line the part of the spinal column which remains above the urostyle. A piece of bone is thus obtained by means of which the nerve can be manipulated without injury. Seize this piece of bone with the forceps, and carefully free the sciatic plexus and nerve from their attachments right down to the gastrocnemius muscle, taking care not to drag upon the nerve. The muscles of the thigh will contract, as the branches going to them are cut. This is an instance of mechanical stimulation. Now pass a thread under the tendo Achillis, tie it, and divide the tendon below it. Strip up the tube of skin that covers the gastrocnemius, as if the finger of a glove were being taken off. Tear through the loose connective tissue between the muscle and the bones of the leg, and divide the latter with scissors just below the knee. Cut across the thigh at its middle. Fix the preparation on the cork plate of the myograph by a pin passed PRACTICAL EXERCISES 809 through the cartilaginous lower end of the femur, and attach the thread to the upright arm of the lever by one of the holes in it. Hang not far from the axis by means of a hook a small leaden weight (5 to 10 grammes) on the arm of the lever which carries the writing-point, and move the myograph plate or the muscle-nerve preparation until this arm is just horizontal. Fasten the electrodes from the secondary coil on the cork plate with an indiarubber band ; lay the nerve on them ; and cover both muscle and nerve with an arch of blotting-paper moistened with physiological salt solution, taking care that the blotting- paper does not touch the thread. Or put the preparation in a moist chamber* (Fig. 322, p. 843). Muscle troughs of various kinds may also Fig. 284. — Lucas's Muscle Trough. A, trough made of hard rubbsr; B, a hard rubber boss with a hole drilled in it to receive the pin which fastens the gastrocnemius preparation; H, H, electrodes cased in hard rubber except at the ends, which in the trough carry platinum wires; C, a brass plate mounted on one side of the trough, carrying a lever with a vertical arm F ending in a hook, which is attached by a loop of thread to the tendon of the preparation; G, the writing arm of the lever; K, M, holes in G for loading the muscle. C can be slid horizontally by means of the slots in it, and clamped by the screw E. I, tube for running off the solution. be used, which permit immersion of a muscle (or nerve) in Ringer's solution. A convenient form is shown in Fig, 284, but a trough suffi- cient for the purposes of the student can be easily improvised in any laboratory. Adjust the writing-point to the drum. Begin with such a distance between the coils that a break contraction is just obtained on opening the key in the primary circuit, but no make contraction. The lever will trace a vertical line on the stationary P- 748. shows a series of curves obtained in this way. 7. Influence of Load on the Muscle- Curve. — Arrange everything as in 6. Take a tracing first with the lever alone, then with a weight oi 10 grammes, then with 50, 100, 200, and 500 grammes (Fig. 254, p. 748). 8. Influence of Fatigue on the Muscle-Curve. — Arrange as in 7, but leave on the same weight (say 10 grammes) all the time. Place the nerve on the electrodes. Leave the short-circuiting key open. The nerve will be stimulated at each revolution of the drum, and the writing- point will trace a series of curves, which become lower, and especially longer, as the preparation is fatigued. Two or four curves can be taken at the same time, if both ends of one or of two brass slips be arranged so as to make contact with the projecting wire at an interval of a semicircumference or quadrant of the drum (Fig. 287). (For specimen curve, see Fig. 288, p. 814.) 9. Seat of Exhaustion in Fatigue of the Muscle-Nerve Preparation for Indirect Stimulation. — When the nerve of a muscle-nerve preparation has been stimulated until contraction no longer occurs, the muscle can, under ordinary conditions, be made to contract by direct stimulation. The seat of exhaustion is, therefore, not the general contractile sub- stance of the muscular fibres themselves. To determine whether it is the nerve-fibres or some structure or substance intermediate between them and the ordinary contractile substance of the muscle, perform the following experiments : (a) Pith a frog; make two muscle-nerve preparations; arrange them both on a myograph plate, which has two levers connected with it. Attach each of the muscles to a lever in the usual way, and lay both nerves side by side on the same pair of electrodes. Cover with moist blotting-paper. The electrodes are connected with the secondary of an induction machine arranged for tetanus. With a camel's hair brush 3i4 MUSCLE AND NERVE moisten one of the nerves between the electrodes and the muscle with a mixture ol equal parts of ether and alcohol, diluted with twice its volume of water, to abolish the conductivity. Or put the mixture in a small bottle, in which dips a piece of filter-paper. The projecting end of the filter-paper is pointed, and the nerve is laid on the point. As soon as it is possible to stimulate the nerves without obtaining contraction in this muscle, proceed to tetanize both nerves till the contracting muscle is exhausted. If the other muscle begins to twitch during the stimu- lation, more of the ether mixture must be painted on the nerve. As soon as the stimulation ceases to cause contraction in the non-etherized preparation, wash off the mixture from the other nerve with physio- logical salt solution, and soon contraction may be seen to take place in Fig. 288. — Fatigue Curve of Skeletal Muscle: Gastrocnemius of Frog. Indirect stimulation; taken with arrangement shown in Fig. 287 (p. 812). Time-tracing, J-JT of a second. the muscle of this preparation. This shows that the nerve-trunk is still excitable. Now, both nerves have been equally stimulated, and there- fore the exhaustion in the non-etherized preparation was not due to fatigue of the nerve-fibres, but of something between them and the contractile substance of the muscle. 10. Influence of Veratrine on Muscular Contraction. — Arrange a drum as in Fig. 287. Pith a frog (brain only), expose the sciatic nerve in one thigh, and isolate it for \ inch from the surrounding tissues. Pass under it a strong thread, and ligature everything except the nerve. Now inject into the dorsal or ventral lymph -sac a few drops of o-i per cent, solution of sulphate of veratrine. In a few minutes make two muscle-nerve preparations from the posterior limbs. First put the preparation from the unligatured limb on the myograph plate. Lay PRACTICAL EXERCISES 815 the nerve on electrodes connected through a short-circuiting key with the secondary of an induction machine arranged as in Fig. 287. Put the writing-point en the drum and set it off (fast speed). Open the short-circuiting key till the nerve has been once stimulated, then close it again. The curve obtained differs from a normal curve, in that the period of descent (relaxation) is exceedingly prolonged. Now connect the preparation from the ligatured limb with the lever, and take a tracing of a single contraction. Put on a time-tracing with the electrical tuning-fork (see Figs. 264, 265, p. ,755). ii. Measurement of the Latent Period of Muscular Contraction. — (i) For this the drum must travel at a faster speed than usual. It is most convenient to use a drum rotated very rapidly by a cord attached to a falling weight or by the recoil of a stretched rubber band or spring. The arrangement for automatic stimulation described in Experiment 6 (p. 813) may be employed. Or an electro-magnetic signal may be connected in the primary circuit of the induction coil so that when the primary is closed or opened the writing-point of the signal moves. Arrange the writing-point of the signal on the drum in the same vertical line as the writing-point of the muscle lever, and in the same line place the writing-point of a vibrating electric tuning-fork. The coil is adjusted for single opening shocks as in Experiment 5 (p. 811). Pith a frog, and make a muscle-nerve preparation. Arrange it on the myograph plate. The muscle, or the nerve very near the muscle, is to be excited by a single opening shock while the drum is moving. When the curve has been traced, the latent period is got by drawing a vertical line through the point at which the curve just begins to rise from the abscissa line, and another through the signal mark. The number of vibrations of the tuning-fork included between these two verticals gives the latent period. Or (2) use the spring myograph (Fig. 251, p. 746), raising it oa blocks of wood. Smoke the glass plate over a paraffin flame, or cover it with paper, and smoke the paper. Connect the knock-over key of the myo- graph with the primary circuit of an induction coil. Arrange a muscle- nerve preparation on the myograph plate. Place electrodes below the nerve as near the muscle as possible, and connect by a short-circuiting key with the secondary. Bring the writing-point in contact with the smoked surface of the spring myograph, so as to get the proper pressure. See that the writing-point of the tuning-fork is in the right position for tracing time. Then push up the plate so as to compress the spring, till the rod connected with the frame which carries the plate is held by the catch. With the short-circuiting key closed, press the release and allow an abscissa line to be traced. Again shove back the frame till it is caught. Push home the rod by means of which the prongs of the tuning-fork are separated, and rotate it through 90°. Close the knock-over key, open the short-circuiting key, shoot the plate again, and a muscle-curve and time-tracing will be recorded. Again close the short-circuiting key, withdraw the writing-point of the tuning-fork, push back the plate, close the trigger key, then open the short-circuiting key, and, holding the travelling frame with the hand, allow it just to open the knock- over and stimulate the nerve. The writing-point now records a vertical line (or, rather, an arc of a circle), which marks on the tracing the moment of stimulation. The latent period is obtained by drawing a parallel line (or arc) through the point of the muscle-curve where it just begins to diverge from the abscissa line. The value of the portion of the time-tracing between these two lines can be readily determined, and is the latent period. 8i6 MUSCLE AND NERVE 12. Summation of Stimuli. — Arrange two knock-over keys on the spring myograph at such a distance from each other that the plate travels frcm one to the other in a time less than the latent period. Connect each key with the primary circuit of a separate induction coil having a couple of Daniells in it. Join two of the binding-screws of the secondaries together; connect the other two through a short- circuiting key with electrodes, on which the nerve of a muscle-nerve preparation is arranged. Push up the secondaries till the break shocks obtained on opening the two knock-over keys are maximal. Then shoot the plate as described in n, first with one trigger key closed, and then with both. The curves obtained should be of the same height in the two cases, as a second maximal stimulus falling within the latent period is ignored by the nerve or muscle. Repeat the experiment with submaximal stimuli — i.e., with such a distance of the coils that opening of either trigger key does not cause as strong a contraction as is caused when the coils are closer. The curve will now be higher when the two shocks are thrown in successively than when the nerve is only once stimulated. This shows that (submaximal) stimuli can be summed in the nerve. The same could be demonstrated for muscle (p. 756). 13. Superposition of Contractions. — Smoke a drum arranged for auto- matic stimulation as in Fig. 287. Adjust the brass points with a distance of, say, i centimetre between them, so that a second stimulus may be thrown into the nerve at an interval greater than the latent period of muscle. Put two Daniells in the primary circuit. Lay the nerve of a muscle-nerve preparation on electrodes connected through a short-circuiting key with the secondary. Allow the drum to revolve (fast speed) ; open the short-circuiting key till both brass points have passed the projecting wire, then close it. Now bend back the second brass point, and take a tracing in which the first curve is allowed to complete itself. This will not rise as high as the second curve obtained when the two stimuli were thrown in. Repeat the experiment with varying intervals between the brass points — that is, between the two successive stimuli. Put on a time-tracing with the electrical tuning- fork. (For specimen curve, see Fig. 266, p. 756.) 14. Composition of Tetanus. — (a) Adjust a muscle-nerve preparation on a myograph plate, the nerve being laid on electrodes connected through a short-circuiting key with the secondary of an induction machine, the primary circuit of which contains a Daniell cell and is arranged for an interrupted current (Fig. 93, p. 200). The lever should be shorter than that used for the previous experiments, or the thread should be tied in a hole farther from the axis of rotation, so as to give less magnification of the contraction. Set the Neef's hammer going, let the drum revolve (slow speed), and open the key in the secondary. The writing-point at once rises, and traces a horizontal or perhaps slightly-ascending line. Close the short-circuiting key, and the lever sinks down again to the abscissa line. If it does not quite return, it should be loaded with a small weight. This is an example of complete tetanus. (b) Connect the spring shown in Fig. 289 with one of the upper terminals of the primary coil, and the mercury cup with the other. Fasten the end of the spring in one of the notches in the upright piece of wood by means of a wedge, so that its whole length can be made to vibrate. Let the drum off, set the spring vibrating by depressing it with the finger, then open the key in the secondary. The muscle is thrown into incomplete tetanus, and the writing-point traces a wavy curve at a higher level than the abscissa line. Close the short-circuiting key, and the iever falls to the horizontal. Repeat the experiment with PRACTICAL EXERCISES 817 the spring fastened, so that only f , £, £, J of its length is free to vibraie. The rate of interruption of the primary circuit increases in proportion to the shortening of the spring, and the tetanus becomes more and more complete, till ultimately the writing -point marks an unbroken straight line. Put on a time- tracing by means of an electro - magnetic marker connected with a metronome beating seconds or half-seconds (Fig. 88, p. 195). (For speci- men curves, see Fig. 267, p. 757-) 15. Contraction of Smooth Muscles — (i) Spontaneous Rhythmical Contrac- tions. — Immerse in oxygenated Ringer's solution a ring of Fig. 289. — Arrangement for Tetanus. A, upright with notches, in which the spring S is fastened (shown in section); C, horizontal board to which A is attached, and in a groove in which the mercury-cup E slides. The primary coil P is connected with E, and through a simple key, K, with the battery B, the other pole of which is connected with the end of the spring. The wires from the secondary coil, P', go to a short- circuiting key, K', from which the wires F go off to the electrodes. oesophagus obtained immediately after death from a cat, or, still better, from a chicken. Or a segment of rabbit's intestine may be employed as described on p. 456. Use the arrangement described on p. 456. In the case of the cat's oeso- phagus the ring should be taken from the lower half of the oesophagus, since the upper portion contains purely striated muscle. Obtain tracings of the rhythmical con- tractions on a slowly- moving drum (Fig. 290). (2) Fix one end of a piece of cat's oeso- phagus, 2 to 5 centi- metres long, to a muscle- clamp in a moist chamber, and the other end to a lever writing on a drum. Connect thin copper wires from the secondary coil of an inductorium with the Fig. 290. — Rhythmical Contractions of (Esophagus of two ends of the piece Chicken (Botazzl). of oesophagus. Take tracings to show (a) the curve of a single contraction caused by a single make or break shock, with estimation of the latent period, as in Experi- ment ii, p. 815; (b) summation, as in Experiment 12, p. 816; (c) genesis 52 8iS MUSCLE AND NERVE of tetanus, as in Experiment 14, p. 816; (d) the relations between strength of stimulus and amount of contraction. For this last experi- ment the drum should be stationary while the contraction is being recorded, and should be allowed to move a little between successive contractions. Begin with the secondary at such a distance from the primary that a contraction is just caused by a break shock. Then gradually increase the strength of the stimulus (always using the break) till maximum contraction is obtained. The gradual increase in the response is very clearly seen with the cesophageal preparation (Waller). For further experiments on the contraction of smooth muscle, see pp. 66 and 457. 1 6. Velocity of the Nerve-Impulse. — Use the spring myograph (Fig. 251, p. 746) or a very rapidly rotating drum. Make a muscle- nerve preparation from a large frog (preferably a bull-frog), so that the sciatic nerve may be as long as possible. Connect the knock-over key with the primary circuit of an induction machine, which should contain Fig. 291. — Arrangement for Measuring the Velocity of the Nerve-Impulse. A, travel ling plate of spring myograph; M, muscle lying on a myograph plate; N, nerve lying on two pairs of electrodes, E and E'; C, Pohl's commutator without cross- wires; K, knock-over key of spring myograph (only the binding-screws shown); K', simple key in primary circuit; B, battery; P, primary coil; S, secondary coil. a single Daniell cell. Arrange two pairs of fine electrodes under the nerve on the myograph plate, one near the muscle, the other at the central end. Connect the electrodes with a Pohl's commutator (with- out cross-wires), the side-cups of which are joined to the terminals of the secondary coil, as shown in Fig. 291. By tilting the bridge of the commutator the nerve may be stimulated at either point. Great care must be taken to keep the nerve in a moist atmosphere by means of wet blotting-paper or a moist chamber; but at the same time it must not lie in a pool of salt solution, as twigs of the stimulating current would in this case spread down the nerve; and we could never be sure that the apparent was always the real point of stimulation. The writing- points of the lever and tuning-fork having been adjusted to the smoked plate, as in n (p. 815), the bridge of the Pohl's commutator is arranged for stimulation of the distal point of the nerve, the plate is shot with the short-circuiting key in the secondary closed, and an abscissa line and time-curve traced. Then the writing-point of the fork is removed and the plate again shot with the key in the secondary open, and a PRACTICAL EXERCISES 819 muscle-curve is obtained. The commutator is now arranged for stimu- lation of the central end of the nerve, and another muscle-curve taken. Vertical lines are drawn through the points where the two curves just begin to separate out from the abscissa line. The interval between these lines corresponds to the time taken by the nerve-impulse to travel along the nerve from the central to the distal pair of electrodes. Its value in time is given by the tracing of the tuning-fork. The length of the nerve between the two pairs of electrodes is now carefully measured with a scale divided in millimetres, and the velocity calculated (p. 793). 17. Chemistry of Muscle. — Mince up some muscle from the hind-legs of a dog or rabbit (used in some of the other experiments), of which the bloodvessels have been washed out by injecting 0*9 per cent, salt solution through a cannula tied into the abdominal aorta until the washings are no longer tinged with blood. To some of the minced muscle add twenty times its bulk of distilled water, to another portion ten times its bulk of a 5 per cent, solution of magnesium sulphate. Let stand, with frequent stirring, for twenty-four hours. Then strain through several folds of linen, press out the residue, and filter through paper, (i) With the filtrate of the watery extract make the following observations : (a) Reaction. — To litmus-paper acid. (b) Determine the temperatures at which coagulation of the various proteins in the extract takes place, according to the method described on p. 9.* Put some of the water}' extract in the test-tube, and heat the bath, stirring the water in the beakers occasionally with a feather. Note at what temperature a coagulum first forms. It will be about 47° C. Filter this off, and again heat; another coagulum will form at 56° to 58°. Filter, and heat the filtrate; a third slight coagulum may be formed at 60° to 65° C., but this represents merely a residue of the myosinogen which was left in solution at the previous heating. A fourth precipitate (of serum-albumin) will come down at 70° to 73°. Saturate some of the watery extract with magnesium sulphate ; a large precipitate will be formed, showing the presence of a considerable amount of globulin. Filter off the precipitate and heat the filtrate; coagulation will again occur at very much the same temperatures as before, although the total amount of precipitate will be less. Note in particular that there is still some precipitate at 47° to 50°. Paramyo- sinogen possesses some of the characters of both globulins and albumins, for it is partially but not entirely precipitated by saturation with magnesium sulphate, and is not precipitated by sodium chloride. (2) (a) Test the reaction of the magnesium sulphate extract. It will usually be faintly acid to litmus. (b) Heat some of it. Precipitates will be obtained at the same tem- peratures as in (i) (b), but those at 47° to 50° and 56° to 58° will be more abundant. Of the two, that at 47° to 50° will usually be the larger when time is given for it to come down and the heating is gradual. (c) Dilute some of the magnesium sulphate extract with three times, another portion with four times, and another with five times, its volume of water in a test-tube, and put in a bath at 40° C. Coagulation or * It should be remembered that the temperature of heat-coagulation of any substance is by no means an absolute constant. It depends on the reaction, the proportion and kind of neutral salts present, perhaps on the strength of the protein solution and the manner of heating. A solution oi egg-albumin, e.g., can be coagulated at a temperature much below 70° when it is heated for a week. Small differences in the temperature of heat-coagula- tion, unless supported by well-marked chemical reactions, are not enough to characterize protein substances as chemical individuals. 820 MUSCLE AND NERVE precipitation will occur in one or all of these test-tubes. To another test-tube of the extract diluted in the proportion which has given the best ' muscle-clot ' add a few drops of a dilute solution of potassium oxalate, and place in a bath at 40°. Coagulation occurs as before. Filter off the clot from all the test-tubes. The filtrate is the ' muscle- serum,' and yields a precipitate of serum-albumin at 70° to 73° C. (3) Myosinogen, like other globulins, is insoluble in distilled water, but soluble in weak saline solutions. Saturation with neutral salts like sodium chloride and magnesium sulphate precipitates myosinogen, but not albumin, from its solutions; saturation with ammonium sulphate precipitates both. Verify the following reactions of myosinogen, using the original magnesium sulphate extract of the muscle : (a) Dropped into water, it is precipitated in flakes, which can be redissolved by a weak solution of a neutral salt (say 5 per cent, mag- nesium sulphate). (b) When a solution of myosinogen is dialyzed, it is after a time pre- cipitated on the inside of the dialyzer as the salts pass out. (c) If a piece of rock-salt is suspended in a solution, the myosin gradually gathers upon it, diffusion of the salt out through the precipi- tated myosin always keeping a saturated layer around it. (d) Saturate a solution containing myosinogen with crystals of magnesium sulphate, stirring or shaking at frequent intervals. The myosinogen is precipitated. (e) Without adding any salt, simply shake a myosinogen solution vigorously; a certain amount of the myosinogen will be precipitated and the solution will become turbid. This reaction can also be ob- tained with solutions of other proteins, such as albumins (Ramsden). Extracts qualitatively similar to those obtained from the muscles of a freshly-killed animal can be got from muscles that have entered into rigor, but the quantity of the various proteins going into solution is less. 18. Reaction of Muscle in Rest, Activity, and Rigor Mortis. — (a) Take a frog's muscle, cut it across, and press a piece of red litmus-paper on the cut end; it is turned blue. Yellow turmeric paper is not affected. (b) Immerse another muscle in physiological salt solution (0-75 per cent, for frog's tissues) at 40° to 42° C. It becomes rigid. The reaction becomes acid to litmus-paper, and also turns brown turmeric paper yellow. (c) Plunge another muscle into boiling physiological salt solution. It becomes harder than in (b). (d) Stimulate another muscle with an interrupted current from an induction machine (Fig. 93, p. 200), till it no longer contracts. The reaction is now acid to litmus-paper. Brown turmeric paper may also be turned yellow. (e) To demonstrate the formation of lactic acid in muscle in heat rigor or fatigue, perform the following experiment: Pith a frog, and afterwards leave it for half an hour at rest, so that the lactic acid pro- duced in the movements connected with the pithing operation may disappear from the muscles. See that the circulation in the hind-limbs is not interfered with by pressure or flexion. Then remove both hind- limbs. Carefully, but rapidly, remove the muscles of one from the bones with as little manipulation as possible. Immediately place them in a small mortar cooled in ice, and containing some sand and 20 or 30 c.c. of ice-cold 95 per cent, alcohol, and quickly grind them up. Produce heat rigor (p. 777) of the muscles of the other hind-limb, or fatigus them with induction-shocks, and then grind them up under alcohol in the same way. P'ilter the alcoholic extracts, and then PRACTICAL EXERCISES 821 evaporate them to dryness on the water-bath. Rub up the residues with a few c.c. of hot water. Add to each aqueous extract a small quantity (say a decigramme) of finely powdered charcoal. Then heat each extract to boiling in a test-tube, and filter. Evaporate the filtrates to dryness, and apply Hopkins 's Reaction for Lactic Acid. — The reagents required are (i) a very dilute alcoholic solution of thiophene (C4H4S) (10 to 20 drops in 100 c.c.); (2) a saturated solution of copper sulphate; and (3) ordinary strong sulphuric acid. Have ready a glass beaker containing water briskly boiling. Place about 5 c.c. of strong sulphuric acid in a test-tube, with i drop of the copper sulphate solution.* Add to the mixture a few drops of the solution to be tested, and shake well.f (In the case of the muscle extracts the dry residues aro dissolved in the 5 c.c. of strong sulphuric acid, the acid transferred to test-tubes, and the test proceeded with by the addition of the copper sulphate solution, etc.) Now place the test-tube in the boiling water for one to two minutes. Then cool it well under the cold-water tap, and add 2 or 3 drops of the thiophene solution from a pipette. Replace the tube in the boiling water, and immediately observe the colour. If lactic acid is present, the liquid rapidly takes on a bright cherry-red colour, which is only permanent if the test-tube be cooled immediately after its appearance. The tube should always be cooled as described, before addition of the thiophene, as the gradual appearance of the colour on re-warming makes the test more delicate. (The extract of the resting limb generally gives a negative, that of the other a strongly positive, reaction.) * The copper sulphate is added to hasten the oxidation that follows. f For practice use a i per cent, alcoholic solution of lactic acid. The test cannot be applied directly to material which chars with the strong sulphuric acid used . In this case preliminary extraction of the lactic acid is necessary. Alcohol should be used as the solvent, or if ether is employed it must first be well washed to remove aldehyde-yielding products, since the colour-change is due to an aldehyde reaction with thiophene. CHAPTER XV ELECTRO-PHYSIOLOGY A LITTLE more than a hundred years ago the foundation both of electro- physiology and of the vast science of voltaic electricity was laid by a chance observation of a professor of anatomy in an Italian garden. It is indeed true that long before this electrical fishes were not only popularly known, but the shock of the torpedo had been to a certain extent scientifically studied. But it was with the discovery of Galvani of Bologna that the epoch of fruitful work in electro-physiology began. Engaged in experiments on the effect of static and atmospheric elec- tricity in stimulating animal tissues, he happened one day to notice that some frogs' legs, suspended by copper hooks on an iron railing, twitched whenever the wind brought them into contact with one of the bars (p. 842). He concluded that electrical charges were developed in the animal tissues themselves, and discharged when the circuit was completed. Volta, professor of physics at Pa via, fixing his attention on the fact that in Galvani's experiment the metallic part of the circuit was composed of two metals, maintained that the contact of these was the real origin of the current, and that the tissues served merely as moist conductors to complete the circuit ; and after a controversy lasting for more than a decade, he finally clinched his argument by constructing the voltaic pile, a series of copper and zinc discs, every two pairs of which were separated by a disc of wet cloth, or paper moistened with salt solution. The pile yielded a continuous current of electricity. ' So,' said Volta, ' it is clear that the tissue in Galvani's experiment only acts the part of the cloth.' Galvani, however, had shown in the meantime that contraction without metals could be obtained by dropping the nerve of a preparation on to the muscle (p. 842) ; and it soon began to be recog- nized that both Galvani and Volta were in part right, that the tissues produce electricity, and that the contact of different metals does so too. Although it is curious to note how completely the growth of that science of which Volta's discovery was the germ has overshadowed the parent tree planted by the hand of Galvani, yet animal electricity has been deeply studied by a large number of observers, and many interesting and important facts have been brought to light. Since it is in muscle and nerve that the phenomena of electro- physiology are seen in their simplest expression, and have been chiefly studied, we shall develop the fundamental laws with reference to muscle and nerve alone, and afterwards apply them to other excitable tissues. i. All points of an uninjured resting muscle or nerve are approxi- mately at the same potential (or iso- electric). In other words, if any 822 DEMARCATION CURRENT 823 two points are connected with a galvanometer by means of un- polarizable electrodes, little or no current is indicated. (Although it is scarcely possible to isolate a muscle without its showing some current, the more carefully the isolation is performed, the feebler is the current; and between two points of the inactive, uninjured ventricle of the frog's heart no electrical difference has been found. Frogs' nerves kept ten to twenty hours after excision in physiological salt solution to which a little calcium salt and frog's blood have been added, are absolutely iso-electric.) 2. Any uninjured point of a resting muscle or nerve is at a different potential from any injured point. The difference of potential is such that a current will pass through the galvanometer from uninjured to injured point and through the tissue from injured to uninjured point (current of rest, or demarcation current, or injury response) (Fig. 292). 3. Any unexcited point of a muscle or nerve is at a different potential from any excited point, and any less excited point is at a different Fig. 292. — A, uninjured, B, injured, portion of nerve; G, galvanometer. The large arrows show direction of demarcation current or current of rest, the small arrows direction of negative variation or action current. Fig. 293. — Diagram of Currents of Rest in a Regular Muscle, or Muscle Cylinder. E, equator. The dotted lines join points at the same po- tential, between which there is no current. potential from any more excited point. The difference of potential is such that a current will pass through the galvanometer to the excited from the unexcited or less excited point (action current, or negative variation, or excitatory electrical response). It has been customary in physiological writings to speak of the electrical change in injured or active tissue as a negative one, because when the tissue is led off to a galvanometer the current passes from the galvanometer to the injured or excited portion of the tissue. It may be called with greater precision ' galvanometrically negative.' It is in this sense that we shall employ the term. The best object for experiments on the demarcation current is a straight-fibred muscle like the frog's sartorius. If this muscle be taken, and the ends cut off perpendicularly to the surface, a muscle-prism or muscle-cylinder is obtained (Fig. 293). The strongest current is got when one electrode is placed on the middle of either cross-section, and the other on the ' equator ' — that is, on a line passing round the longi- tudinal surface midway between the ends. The direction of this current is from the cross-section towards the equator in the muscle. If the electrodes are placed on symmetrical points on each side of the equator, there is no current. 824 ELECTRO-PHYSIOLOGY Current of Action, or Negative Variation. — When a muscle or nerve is excited, an electrical change sweeps over it in the form of a wave. Suppose two points, A and B (Fig. 294), on the longi- tudinal surface of a muscle to be connected with a capillary electro- meter (p. 729), the movements of the mercury being photographed on a travelling surface — for example, a pendulum carrying a sensitive plate. Let the muscle be excited at the end, so that the wave of excitation will be propagated in the direction of the arrow. The wave will reach A first, and while it has not yet reached B, A will Fig. 294. — Diagram to illustrate Propagation of the Electrical Change along an Active Muscle or Nerve. Suppose AB to be a horizontal bar representing the muscle or nerve. Let C be a curved piece of wood representing the curve of the electrical change at any point. Let W, W' be two glass cylinders connected by a flexible tube, the whole being filled with water. Suppose the rims of the cylinders originally to touch AB at the points A and B, and let them be movable only in the vertical direction. The level of the water being the same in both, there is no tendency for it to flow from one to the other. This represents the resting state of the tissue when A and B are symmetrical points. Now let C be moved along the bar at a uniform rate. The cylinder W, being free to move down, but not horizontally, will be displaced by C, and, if it is kept always in contact with its curved margin, will, after describing the curve of the electrical variation, come again to rest in its old position at A. B will do the same when C reaches it. But since C reaches A before B, the level of the water in B will at first be higher than that in A, and water will flow from B to A as the current flows through the galvanometer. This will correspond to the time during which the point of the tissue represented by A would be galvanometrically negative to a point represented by B. Later on, when C has reached the position shown by the dotted lines, the level of the water in A will be higher than that in B, and a flow will take place in the opposite direction to the first flow. This corresponds to a second phase of the electrical variation. become negative to B. If there is a resting difference of potential between A and B, this will be altered, the new and transitory differ- ence adding itself algebraically to the old. When the wave reaches B, it may already have passed over A altogether, and B now be- coming negative to A, there will be a movement of the meniscus of the electrometer in the opposite direction. This is called the NEGATIVE VARIATION 825 diphasic current of action. If the wave has not passed over A before it reaches B, as would in general be the case in an actual experiment, there will be first a period during which A is relatively negative to B (first phase) ; this will end as soon as B has become iso-electric with A, and will be succeeded by a period during which B is rela- tively negative to A (second phase). Since the wave takes time to reach its maximum, it is evident that a well-marked first phase will be favoured when the interval between its arrival at A and at B is long, for in this case A will have a chance of becoming strongly negative while B is still normal. Similarly, if A has again become normal, or nearly normal, before the maximum negative change has passed over B, a strong second phase will be favoured. The heart- muscle, accordingly, where the wave of contraction, and its accom- panying electrical change, move with comparative slowness, te better suited for showing a well-marked diphasic variation than skeletal muscle, and still better suited than nerve. In the gastroc- T*^* • • • r» Fig. 295. — Photographic Electrometer Curves from Sartorius Muscle (Sanderson). The darkly-shaded curve represents the diphasic variation of the uninjured muscle; the lightly-shaded curve the monop basic variation of the muscle after injury of one end. The toothed curve at the top is the time-tracing registered by photographing the prong of a tuning-fork vibrating five hundred times a second. nemius muscle of the frog, when excited through its nerve, the elec- trical response begins about T> **' R Fig. 314. — Electro-Cardiogram from Man (String Galvanometer) (Lewis). From a case of paroxysmal tachycardia. The heart-rate was 200 a minute. The upper notched line is the time-trace in one-fifth seconds. electro-cardiograms, (Fig. 309) is reproduced for their historical interest. In their main features it is obvious that they agree with the records obtained by the string galvanometer. The electro-cardiograms are distinctly affected by exercise and by the position of the body, and very markedly in disease.. The galvanometer may be connected with the two ELECTRO-CARDIOGRAM 837 hands, or, better, with the right hand and the left foot. The two feet are the most unfavourable combination. The reason is obvious from the direction of the long axis of the heart, which determines the direction of the lines of flow of currents due to differences of potential between base and apex (Fig. 310). Relation of the Waves of the Electro-Cardiogram to the Mechanical Events in the Cardiac Cycle. — There is evidence that the excitation pro- cess with its associated electrical change spreads over the auricle from the sinus node, followed at each point by the mechanical change or contrac- tion, in such a manner that the portions nearer the node begin to relax before the more distal units have finished contracting. When a tracing of the approximation of two distant points of the auricle is taken, the total shortening represents the algebraic sum of the contractions and relaxations of all the muscular units between the two points. Fig- 3I5- — Upper curve, pressure in right auricle. Second curve from top, right auricular myogram; the down-stroke corresponds to contraction. Third curve, ventricular sounds. Bottom curve, electrocardiogram, lead II (left hind and right fore leg) (Wiggers). Electrical effects obtained by leading off from the heart in any particu- lar way must also be more or less complex resultants of the changes at different points. Certain general relations, however, have been estab- lished between the mechanical events and the electro-cardiogram. In Fig. 315 are shown simultaneous records of the contraction of the auricle (auricular myogram), the intra-auricular pressure, the heart sounds and the electro-cardiogram (Wiggers). The first electrical variation, commencing at i, is seen to precede the rise of pressure in the auricle, 2, by a definite interval (about 0-02 sec.), and the onset of the mechanical shortening ,3, by a somewhat greater interval (0-03 sec.). The length of these intervals is, of course, not precisely the same in different experiments. The relation of the beginning of the rise of intra-auricular pressure and of the beginning of the mechanical systole of the auricles to the 838 ELECTRO-PHYSIOLOGY P wave of the electro-cardiogram (the wave specially associated with the passage of excitation over the auricle) is also slightly variable. Roughly, the apex of the P wave may be taken as indicating the onset of the auricular systole. The relation between the end of the auricular systole, 4, and the electro-cardiogram, is still more variable, but in general it falls not far from the summit of the R wave (a wave specially associated with the passage of the excitation over the ventricle) . This shows that the auricular systole still continues at a time when the ex- citation process has already made considerable progress in the ventricle. The mechanical contraction of the ventricle, as shown in Fig. 315, by the position of the vibrations corresponding to the first sound, follows very promptly the completion of the auricular systole. Central Nervous System. — It was discovered by du Bois-Reymond that the spinal cord, like a nerve, shows a current of rest between longi- tudinal surface and cross-section, and that a current of action is caused by excitation. Setschenow stated that when the medulla oblongata of the frog was connected with a galvanometer, spontaneous variations occurred which he supposed due to periodic functional changes in its grey matter. Gotch and Horsley have made experiments on the spinal cords of cats and monkeys. Leading off from an isolated portion of the dorsal cord to the capillary electrometer, and stimulating the ' motor ' region of the cortex cerebri, they obtained a persistent nega- tive variation followed by a series of intermittent variations. This agrees remarkably with the muscular contractions in an epileptiform convulsion started by a similar excitation of the cortex, which consist of a tonic spasm followed by clonic or phasic (interrupted) contractions. By means of the galvanometer, the same observers have made in- vestigations on the paths by which impulses set up at different points travel along the cord. To these we shall have to refer again (p. 895). Electrical Phenomena of Glands. — These have been studied with any care chiefly in the submaxillary gland and in the skin, although the liver, kidney, spleen, and other organs also show currents when injured. In the sub- maxillary gland the hilus is galvanometrically positive to any point on the external surface of the gland ; a current passes from hilus to surface through the galvanometer, and from surface to hilus through the gland (Fig. 316). Fig. 316. — Current of Sub- When the chorda tympani is stimulated with maxillary Gland. rapidly - succeeding shocks of moderate strength, there is a positive variation — i.e., the hilus becomes still more positive to the surface. This variation can be abolished by a small dose of atropine. Skin Currents. — So far as has been investigated, the integument of all animals shows a permanent current passing in the skin from the external surface inwards. This is feebler in skin which possesses no glands. In skin containing glands the current is chiefly, but not altogether, secre- tory. As such, it is affected by influences which affect secretion, a positive variation being caused by excitation of secretory nerves — e.g., in the pad of the cat's foot by stimulation of the sciatic. The deflection obtained when a finger of each hand is led off to the galvanometer, which was at one time looked upon as a proof of the existence of currents of rest in intact muscles, is due to a secretion current. Of more doubtful origin is the current of ciliated mucous membrane, which has the same direction as that of the skin of the frog and the mucous membrane of the stomach of the frog and the rabbit — viz., from ciliated to under surface through the tissue, or from ciliated surface to cross-section, if that is the way in which it is led off. The current is ELECTROMOTIVE PHENOMENA OF THE EYE 839 strengthened by induction shocks, by heating, and in general by influ- ences which increase the activity of the cilia. Some circumstances point to the goblet-cells in the membrane as the source of the current; but, on the whole, the balance of evidence is in favour of the cilia being the chief factor (Erigelmann), although the mucin-secreting cells may be concerned, too. Electrical changes associated with secretion have been observed in the frog's tongue on excitation of the glosso-pharyngeal nerve. Eye- Currents. — If two unpolarizable electrodes connected with a galvanometer are placed on the excised eye of a frog or rabbit, one on the cornea and the other on the cut optic nerve, or on the posterior surface of the eyeball, it is found that a current passes in the eye from optic nerve to cornea, the fundus of the eye being therefore negative as regards the cornea (Fig. 317). The current has the same direction if the anterior electrode is placed on the an- terior surface of the retina itself, the front of the eyeball being cut away, or if one electrode is in contact with the anterior and the ether with the posterior surface of the isolated retina. There is nothing of special interest in this ; but the important point is that if light be now allowed to fall upon the eye, or upon the isolated retina, characteristic electrical changes are caused. These are spoken of as the photo- electric reaction, and are best studied by means of the string galvanometer. The features of the curve representing the photo-electric reaction vary with the duration and intensity of the illumination and with the previous condition of the eye as regards illumination. A careful analysis of the curves obtained under different conditions supports the hypothesis that there occur in the eye three separate processes, which may for convenience be con- sidered to depend upon the existence in the retina of three separate photo-chemical substances. When light of moderate intensity is allowed to act upon an eye which has not shortly before been exposed g. 317. — Eye-Current. Fig. 318.— Photo -Electric Reaction of Frog's Eye (Einthoven and Jolly). The duration of the flash (of green light) was o-oi second. The eye had been pre- viously in the dark, i millimetre of the abscissa corresponds to 0-5 second, i millimetre of the ordinate to 10 microvolts. Curve to be read from left to right. to strong light, a form of curve is obtained which seems to represent the combined reaction of the three substances (Einthoven and Jolly) (Fig. 318). After a latent period a small preliminary negative deflec- tion A is observed (downward movement of the string). This is at once 840 ELECTRO-PHYSIOLOGY followed by a much larger upward movement (positive variation) in the same direction as the resting effect, the fundus becoming relatively more negative to the cornea than before. After the peak B has been reached, the curve sinks first rapidly, then more gradually, but soon mounts again, and reaches a second maximum C, vhichis higher than B (second positive variation). Finally, the curve descends to its original level.* The photo-electric reaction is substantially the same in all vertebrate eyes hitherto investigated. In the cephalopcd retina, too, the only important electrical change on illumination is in the same direction as the resting effect. The reaction depends upon the retina alone, and does not occur when it is removed. Bleaching of the visual purple does not much affect it, so that it is not connected with chemical changes in this substance. Its seat must be the layer of rods and cones, since in the Fig. 319. — Diagram showing Direction of Shock in Gymnotus. cephalopods the structure called the retina contains only this layer, the other layers of the vertebrate retina being represented in the optic nerve and ganglion (Beck). Of the spectral colours, yellow light causes the largest variation; blue, the least; but white light is more powerful than either (Dewarand McKendrick). (For 'visual purple, 'see Chap. XVIII.) Electric Fishes. — Except lightning, the shocks of these fishes were probably the first manifestations of electricity observed by man. The Torpedo, or electrical ray, of the coasts of Europe was known to the Greeks and Romans. It is mentioned in the writings of Aristotle and Pliny, and had the honour of being described in verse 1,500 years before Faraday made the first really exact investigation of the shock of the Gymnotus, or electric eel, of South America. Another of the electric fishes, Malapterurus electricus, al- though found in many of the African rivers, the Nile in par- ticular, and known forages, was scarcely investigated till fifty years ago. In all these fishes there is a special bilateral organ immediately under the skin, called the electrical organ. It is in this that the shock is developed. It consists of a series of plates arranged parallel to each other. To one side of each plate a branch of the electrical nerve supplying each lateral half of * In the figure the last portion of the curve while it is still slowly descending has not been reproduced. Fig. 320. — Diagram showing Direction of Shock in Malapterurus. ELECTRIC FISHES 841 the organ is distributed, so that each half of the organ represents a battery of many cells arranged in series. In Gymnotus the plates are vertical, and at right angles to the long axis of the fish, and the nerves are distributed to their posterior surface ; the shock passes in the animal from tail to head. In Malapterurus, although the direction of the plates is the same, and the nerve-supply is also to the posterior surface, the shock passes from head to tail. In Torpedo, the plates or septa dividing the vertical hexagonal prisms of which each lateral half of the organ consists are horizontal ; the nerve- supply is to the lower or ventral surface ; and the shock passes from belly to back through the organ. In all electric fishes the discharge is dis- continuous; an active fish may give as many as 200 shocks per second. The electrical nerve of Malapterurus is peculiar. It consists of a single gigantic nerve-fibre on each side, arising from a giant nerve-cell. The fibre has an enormously thick sheath, the axis-cylinder forming a relatively small part of the whole; and the branches which supply the plates of the organ are divisions of this single axis-cylinder. The electromotive force of the shock of the Gymnotus may be very considerable; and even Torpedo and Malapterurus are quite able to kill other fish, their enemies or their prey. Indeed, Gotch has esti- mated the electromotive force of i cm. of the organ of Torpedo at 5 volts. Schonlein finds that the electromotive force of the whole organ may be equal to that of 31 Daniell cells, or 0-08 volt for each plate, and it is one of the most interesting questions in the whole of electro-physiol'jgy, how they are pro- tected from their own currents . There is no doubt that the current density inside the fish must be at least as great as in any part of the water sur- rounding it, and probably much greater. The central nervous system and the great nerves must be struck by strong shocks, yet the fish itself is Fig. 321. — Diagram showing Direc- not injured; nay, more, the young in tion of Shock in Torpedo, the uterus of the viviparous Torpedo are unharmed. The only explanation seems to be that the tissues of electric fishes are far less excitable to electrical stimuli than the tissues of other animals; and this is found to be the case when their muscles or nerves are tested with galvanic or induction currents. It requires ex- tremely strong currents to stimulate them ; and the electrical nerves are more easily excited mechanically, as by ligaturing or pinching, than elec- trically. In general, too, the shock is morg readily called forth by reflex mechanical stimulation of the skin than by electrical stimulation. But that the organ itself is excitable by electricity has been shown by Gotch. He proved that in Torpedo a current passed in the normal direction of the shock is strengthened, and a current passed in the opposite direction weakened, by the development of an action current in the direction of the shock. And, indeed, a single excitation of the electrical nerve is followed by a series of electrical oscillations in the organ, which gradually die away. The latent period of a single shock is about ^J0 second. The skate must be included in the list of electric fishes. Although its organ is relatively small, and its electromotive force rela- tively feeble, yet it is in all respects a complete electrical organ. It is situated on either side of the vertebral column in the tail. The plates or discs are placed transversely and in vertical planes. The nerves enter their anterior surfaces; the shock passes in the organ from anterior to posterior end. Gotch and Sanderson have estimated the maximum electromotive force of a length of I cm. of the electrical organ of the skate at about half a volt. 842 ELECTRO-PHYSIOLOGY Whether the electrical organ is the homologue of muscle or of nerve- ending, or whether it is related to either, has been much discussed. Our surest guide in a question of this sort is the study of development; and researches along this line have shown that there are two kinds oi electrical organ, one being modified muscle (as in Gymnotus, Torpedo, and the skate) ; the other transformed skin-glands (as in Malapterurus) . The scanty blood-supply of the electrical organs in comparison with that of muscle is noteworthy. In no case do bloodvessels enter the substance of the plates. PRACTICAL EXERCISES ON CHAPTER XV. 1 . Galvani's Experiment. — Pith a frog (brain and cord). Cut through the backbone above the urostyle, and clear away the anterior portion of the body and the viscera. Pass a copper hook beneath the two sciatic plexuses, and hang the legs by the hook on an iron tripod. If the tripod has been painted, the paint must be scraped away where the • hook is in contact with it. Now tilt the tripcd so that the legs come in contact with one of the iron feet. Whenever this happens, the circuit for the current set up by the contact of the copper and iron is completed, the nerves are stimulated, and the muscles contract (p, 822). 2. Make a muscle-nerve preparation from the same frog. Crush the muscle near the tendo Achillis, so as to cause a strong demarcation current. Cut off the end of the sciatic nerve. Then lift the nerve with a small brush or thin glass rod, and let its cross-section fall on or near the injured part of the muscle. Every time the nerve touches the muscle a part of the demarcation current passes through it, stimulates the nerve, and causes contraction of the muscle (p. 822). 3. Secondary Contraction. — Make two muscle-nerve preparations. Lay the cross-section of one of the sciatic nerves on the muscle of the other preparation (Fig. 305, p, 833). Place under the nerve near its cut end a small piece of glazed paper or of glass rod, and let the longi- tudinal surface of the nerve come in contact with the muscle beyond this. Lay the nerve of the other preparation on electrodes connected with an induction machine arranged for single shocks, with a Daniell cell and a spring key in the primary circuit (Fig, 283, p. 808). On closing or opening the key both muscles contract. Arrange the induc- tion machine for an interrupted current. When it is thrown into one nerve, both muscles are tetanized; the nerve lying on the muscle whose nerve is directly stimulated is excited by the action current of the muscle. 4. Demarcation Current and Current of Action with Capillary Elec- trometer. — (a) Study the construction of the capillary electrometer (Fig. 235, p. 729). Raise the glass reservoir by the rack and pinion screw, so as to bring the meniscus of the mercury into the field. Place two moistened ringers on the binding-screws of the electrometer, open the small key connecting them, and notice that the mercury moves, a difference of potential between the two binding-screws being caused by the moistened fingers. (b) Demarcation Current. — Set up a pair of unpolarizable electrodes (Fig. 238, p. 731). Fill the glass tubes about one-third full of kaolin mixed with physiological salt solution till it can be easily moulded. To do this, make a piece of the clay into a little roll, which will slip down the tube. Then with a match push it down until it forms a firm plug. Next put some saturated zinc sulphate solution in the tubes, above the clay, with a fine-pointed pipette. Fasten the tubes in the holder fixed in the moist chamber (Fig. 322). Now amalgamate the small pieces of zinc wire (p. 197) which are to be connected with the binding-screws of the chamber. (Or use Porter's ' boot ' electrodes. These are made of unglazed potter's clay. In use the leg of the boot is half-filled with PRACTICAL EXERCISES 843 saturated zinc sulphate solution, into which dips a thick amalgamated zinc wire, in the toot ot the boot is a hollow (or well) which is filled with physiological salt solution and serves to keep the feet well moist- ened, with the salt solution. The nerve is laid on the feet of the boots. When not in use the boots should be kept in physiological salt solution.) The zincs are now placed in the tubes, dipping into the zinc sulphate. A piece of clay or blotting-paper moistened with physiological salt solu- tion is laid across the electrodes to complete the circuit between their points, and they are connected with the electrometer to test whether they have been properly set up. There ought to be little, if any, move- ment of the mercury on opening the side-key of the electrometer. If the movement is large, the electrodes are 'polarized,' and must be set up again. The second pair of bind- ing - screws in the chamber are con- nected with a pair of platinum-pointed electrodes on the one side, and on the other, through a short-cir- cuiting key, with the secondary coil of an induction machine ar- ranged for tetanus. Next pith a frog (cord and brain), and make a muscle-nerve preparation . Inj ure Q' B Fig. 322. — Moist Chamber. E, unpolarizable electrodes supported in the cork C; M, muscle stretched over the electrodes and kept in position by the pins A, B, stuck in the cork plate P; B, binding-screws connected with galvanometer or capillary electrometer. The other pair of binding-screws serves to connect a pair of stimulating electrodes inside the chamber with the secondary coil of an induction machine. the muscle near the tendo Achillis. Lay the injured part over one unpolarizable electrode, and an un- injured part over the other. Put a wet sponge in the chamber to keep the air moist, and place the glass lid on it. Focus the meniscus of the mercury, and open the key of the electrometer; the mercury will move, perhaps right out of the field. Note the direction of movement, and, remembering that the real direction is the opposite of the apparent direction, and that when the mercury in the capillary tube is connected with a part of the muscle which is relatively positive to that connected with the sulphuric acid, the movement is from capillary to acid, determine which is the galvano- metrically positive and which the negative portion of the muscle (p. 823). (c) Action Current. — Now, without disturbing its position on the electrodes, fasten the muscle to the cork or paraffin plate in the moist chamber by pins thrust through the lower end of the femur and the tendo Achillis. Lay the nerve on the platinum electrodes. Open the key of the electrometer, and let the meniscus come to rest. This happens very quickly, as the capillary electrometer has but little inertia. If the meniscus has shot out of the field, it must be brought back by raising or lowering the reservoir. Stimulate the nerve by opening the key in the secondary circuit ; the meniscus moves in the direction oppo- site to its former movement. (d) Repeat (b) and (c) with the nerve alone, laying an injured part (crushed, cut, or overheated) on one electrode, and an uninjured part on the other. Of course, the nerve does not need to be pinned. 844 ELECTRO- PH YSIOLOG Y Clean the unpolarizable electrodes, and he sure to lower the reservoir of the electrometer ; otherwise the mercury may reach the point of the capillary tube and run out. In 4 a galvanometer (p. 726) may be used with advantage by students, if one is available, instead of the electrometer, the un- polarizable electrodes being connected to it through a short-circuit- ing key. The spot of light is brought to the middle of the scale by moving the control-magnet; or if a telescope-reading is being used, the zero of the scale is brought by the same means to coincide with the vertical hair-line of the telescope. The short-circuiting key is then opened. 5. Action Current of Heart. — Pith a frog (brain and cord). Excise the heart, and lay the base on one unpolarizable electrode, and the apex on the other, having a sufficiently large pad of clay on the tips of the electrodes to insure contact during the movements of the heart, or having little cups hollowed in the clay and filled with physiological salt solution, into which the organ dips. Connect the electrodes with the capillary electrometer and open its key. At each beat of the heart the mercury will move (p. 833). 6. Electrotonus. — Set up two pairs of unpolarizable electrodes in the moist chamber. Connect two of them with a capillary electrometer (or galvanometer), and two with a battery of three or four small Daniell cells, as in Fig, 304. Lay a frog's nerve on the electrodes. When the key in the battery circuit is closed, the mercury (or the needle of the galvanometer) moves in such a direction as to indicate that in the extra- polar regions parts of the nerve nearer to the anode are relatively positive to parts more re- mote, and parts nearer to the kathode are rela- tively negative to parts more remote. The direction of movement of the mercury (or gal- vanometer needle) must be made out first for one direction of the polarizing current. Then the latter must be reversed, and the movement of the mercury (or needle) on closing it again noted (P Sao/ 7. Paradoxical Contraction. — Pith a frog (brain and cord) . Dissect out the sciatic nerve down to the point where it splits into two divisions, one for the gastrocnemius b, and the other for the peroneal muscles a. Divide the peroneal branch as low down as possible, and make a muscle-nerve preparation in the usual way. Lay the central 323- — Paradoxical end of the peroneal nerve on electrodes con- Contraction, nected through a simple key with a battery of two Daniell cells. When the peroneal nerve is stimu- lated, the gastrocnemius muscle contracts. This result is not due to the current of action, for it is not obtained with mechanical stimulation of the nerve. But it is not the result of an escape of current, for if the peroneal nerve be ligatured between the point of stimulation and the bifurcation, no contraction is obtained. The contraction is really due to a part of the electrotonic current set up in the peroneal nerve passing through the fibres for the gastrocnemius, where they lie side by side in the trunk of the sciatic. 8. Alterations in Excitability (and Conductivity) produced in Nerve by the Passage of a Voltaic Current through it. — Set up two pairs of unpolarizable electrodes in the moist chamber. Connect a battery of two or three Daniell cells, arranged in series through a simple key PRACTICAL EXERCISES 84$ with the side-cups of a Pohl's commutator with cross-wires in. Con- nect the commutator to one pair of the unpolarizable electrodes (' the polarizing electrodes '), as in Fig. 324. The other pair of unpolarizable electrodes (' the stimulating electrodes ') are to be connected through a short-circuiting key with the secondary of an induction machine arranged for tetanus. A single Daniell is put in the primary coil. Pith a frog (brain and cord), make a muscle-nerve preparation, pin the lower end of the femur to the cork plate in the moist chamber, attach the thread on the tendo Achillis to the lever connected with the chamber through the hole in the glass provided for this purpose, and arrange the nerve on the electrodes so that the stimulating pair is between the muscle and the polarizing pair. By moving the secondary, seek out such a strength of stimulus as just suffices to cause a weak tetanus when the polarizing current is not closed. Set the drum off (slow speed), and take a tracing of the contraction. Then close the polarizing current with a Pohl's commutator so arranged that the anode is next the stimulating electrodes — i.e., the current ascending in the nerve. Again open the short-circuiting key in the secondary; the contraction will now be weaker than before, or no contraction at all may be obtained. Allow the preparation two minutes to recover, Fig. 3 H. — Arrangement for showing Changes of Excitability produced by the Voltaic Current. M, muscle; N, nerve; Ej, Eg, electrodes connected with secondary coilS; E3, E4, unpolarizable electrodes connected with Pohl's commutator (with cross-wires) C; B', 'polarizing' battery; B, 'stimulating' battery in primary circuit P; K, K", simple keys; K', short-circuiting key. then stimulate again, as a control, without closing the polarizing current. If the contraction is of the same height as at first, close the polarizing current with the bridge of the commutator reversed, so that the kathode is now next the stimulating electrodes. On stimu- lating, the contraction will now be increased in height. (See Figs. 273. 274, P- 786.) 9. Pfliiger's Formula of Contraction (p. 788). — To demonstrate this, connect two unpolarizable electrodes, through a spring key and a commutator, with a simple rheocord (Fig. 286, p. 810), so as to lead off a twig of a current from a Daniell cell. The unpolarizable elec- trodes are placed in a moist chamber. A muscle-nerve preparation is arranged with the nerve on the electrodes and the muscle attached to a lever. The effects of make and break of a weak current, ascending and descending, can be worked out with the simple rheocord. The effects of a medium current will probably be obtained with a single Daniell connected directly with the electrodes through a key. The effects of a strong current will be got when three or four Daniells are connected with the electrodes. Care must be taken to keep the prepara- tion in a moist atmosphere, and more than one preparation may be needed to verify the whole formula. 846 ELECTRO-PHYSIOLOG Y 10. Formula of Contraction for (Human) Nerves in Situ. — Connect eight or ten dry cells in series.* Connect one terminal of the battery to a large plate electrode, and the other to a small electrode, both covered with cotton, flannel, or sponge, moistened with salt solution. Include in the circuit a simple key for making or breaking the current, and a commutator for changing its direction at will. Leave the key open. Place the large electrode behind the shoulder (or on the back of the neck), and the small electrode over the ulnar nerve at the elbow between the internal cbndyle and the olecranon. Arrange the com- mutator so that the small electrode shall be the kathode. Close, and then open the key. If no contraction occurs at closing, the battery is too weak, and more cells must be added. If contraction occurs at closing, but not at opening, reverse the commutator, making the small electrode the anode, and observe whether contraction now occurs at closing, at opening, or at both. Note also the relative strength of the various contractions. If the current is ' weak/ the only contraction will be a closing one when the kathode is over the nerve. If the current is of ' medium ' strength, a closing kathodic contraction and both opening and closing anodic contractions will be obtained. With ' strong ' currents contractions will occur at closing and at opening, whether the kathode or the anode is over the nerve. The contractions will vary in strength, as described on p. 789. To work out the different cases of the formula summarized in the table, the number of cells must be increased or diminished. Weak Currents. Medium Currents. Strong Currents. KCC KCC KCC — ACC ACC — AOC AOC — — KOC The abbreviations KCC, ACC, are used respectively for kathodic closing contraction and anodic closing contraction; KOC, AOC, for kathodic opening contraction and anodic opening contraction. KCC is stronger than KOC, and ACC than AOC. KCC is stronger than ACC, and AOC than KOC. Therefore, as the strength of the current is increased, in the case of normal tissues, KCC is first obtained, then ACC, then AOC, and finally KOC. n. Ritter's Tetanus. — Lay the nerve of a muscle-nerve preparation on a pair of unpolarizable electrodes connected through a simple key with a battery of three or four small Daniells. Connect the muscle with a lever. Pass an ascending current (anode next the muscle) for a few minutes through the nerve, and let the writing-point trace on a slowly-moving drum. When the current is closed there may be a single momentary twitch, or the muscle may remain somewhat con- tracted (galvanotonus) as long as the current is allowed to pass, or it may continue to contract spasmodically (' closing tetanus '). When the current is opened the muscle will contract once, and then immedi- ately relax, or there may be a more or less continued tetanus (Ritter's qr ' opening tetanus '). If opening tetanus is obtained, divide the nerve between the electrodes: the tetanus continues. Divide it be- tween the anode and the muscle: the tetanus at once disappears. This shows that the seat of the excitation which causes the tetanus is in the neighbourhood of the anode (p. 831). * If the laboratory possesses a battery (with rheostat), such as is used by neurologists, the experiment is more conveniently performed with this. CHAPTER XVI THE CENTRAL NERVOUS SYSTEM IN other divisions of our subject we have been able to follow to a greater or less extent the processes which take place in the organs described. The chemistry and the physics of these processes have bulked more largely in our pages than the anatomy and histology of the tissues themselves. In dealing with the central nervous system, we must adopt a method the very reverse of this. Its ana- tomical arrangement is excessively intricate. The events which take place in that tangle of fibre, cell, and fibril are, on the other hand, almost unknown. So that in the description of the physiology of the central nervous system we can as yet do little more than trace the paths by which impulses may pass between one portion of the system and another, and from the anatomical connections deduce, with more or less probability, the nature of the physiological nexus which its parts form with each other and the rest of the body. And here it may be well to remark that, although for convenience of treatment we have considered the general properties of nerves in a separate chapter, there is not only no fundamental distinction between the central nervous system and the outrunners which connect it with the periphery, but obviously a central nervous system would be meaningless and useless without afferent nerves to carry information to it from the outside, and efferent nerves along which its commands may be conducted to the peripheral organs. SECTION I. — STRUCTURE OF THE CENTRAL NERVOUS SYSTEM — HlSTOLOGICAL ELEMENTS. In unravelling the complex structure of the central nervous system, we avail ourselves of information derived (i) from its gross anatomy ; (2) from its microscopical anatomy ; (3) from its develop- ment ; (4) from what we may call, although the term is open to the criticism of cross-division, its physiological and pathological anatomy. Certain tracts of white or grey matter are differentiated from each other by the size of their fibres or ceils. For example, the postero- median column of the spinal cord has small fibres, the direct cere- 847 848 THE CENTRAL NERVOUS SYSTEM bellar tract large fibres; the large pyramidal cells (giant cells or cells of Betz), in what we shall afterwards have to distinguish as the ' motor area ' (p. 950) of the cerebral cortex, are the cells of origin of fibres of the pyramidal tract subserving the volitional movements of the limbs and trunk. The pyramidal cells of the ' face area ' are comparatively small. In general, an efferent or motor nerve-cell is larger the longer its axon is — e.g., the largest of all the pyramidal cells in the ' motor ' region are found in the portion known as the ' leg area,' from which the pyramidal fibres have to pass all the way down the cord to the segments from which the spinal nerves going to the lower limbs arise. The recent work of Brodman and of Campbell has shown that the cerebral cortex may be histologically differentiated into regions which correspond to a great extent to the various functional regions mapped out by physiological methods (p. 958). The study of development enables us not only to determine the homology, the morphological rank, of the various parts of the brain and cord, but also, by comparison of animals of different grades of organization, sometimes to decide the probable function and physio- logical importance of a strand of nerve-fibres or a column of nerve- cells. It is of special value in helping us to differentiate the various areas of grey matter on the surface of the brain, and to trace the various tracts or paths into which the white matter of the central nervous system may be divided. For the medullary sheath is not developed at the same time in all the tracts, and a strand of nerve-fibres in which it is wanting — e.g., the pyramidal tract (p. 878), which is the last of the spinal tracts to become myelinated — is readily distinguished under the microscope. Then, again — and this is what we propose to include under the fourth head — experimental physiology and clinical and pathological observation throw light not only on the functions, but also on the structure, of the central nervous system. For instance, complete or partial sectioH, or destruction by disease, of the white fibres of the cord or brain, or of the nerve-roots, or removal of portions of the grey matter, is followed by degeneration in definite tracts. And since, as we have already seen, degeneration of a nerve-fibre is caused when it is cut off from the cell of which it is a process, the amount and dis- tribution of such degeneration teaches us the extent and position of the central connections of the given tract. Conversely, the cells in which a tract of nerve-fibres arises may sometimes be identified by the alterations in the chromatin (p. 859) and other changes which occur in them after section of their axons. Particularly in young animals, removal of a peripheral organ — an eye or a limb — or section of its nerves, may be followed by atrophy of portions of the central nervous system immediately related to it. ' Softening ' of a definite portion of the white or grey matter may also in certain cases be caused by depriving it of its blood-supply by the injection of artificial emboli, and the resulting degenerations may then be studied. For instance, fine particles like lycopodium spores are injected into the abdominal aorta between the origins of the renal and inferior mesenteric arteries. They are prevented by clamps from entering these vessels, and, passing through the lumbar arteries, stick in the branches of the anterior spinal artery, and cause softening mainly of the grey matter of the lumbar portion of the cord. When the abdominal aorta of a rabbit is temporarily compressed (for about an hour) below the origin of the renal arteries, the grey matter of the corresponding portion of the cord is so seriously injured that it and the fibres that arise from it degenerate, while the fibres whose cells of 325- — Formation of the Neural Canal at an Early Stage (Beard). origin are not situated in this part of the grey matter are not affected, or at least completely recover. Certain tracts may also be marked out by means of the electrical variation, which gives token of the passage of nervous impulses along them when portions of the central nervous system or peri- pheral nerves are stimulated (Horsley and Gotch) . Development of the Central Nervous System. — Very early in development (Fig, 325) the keel of the vertebrate embryo is laid down as a groove or gutter in the ectoderm of the blastodermic area (Chap. XIX.). The walls of this ' medullary ' or ' neural ' groove grow inwards, and at length there is formed, by their coalescence, the ' neural canal ' (Fig. 326), which expands at its anterior end to form four cerebral vesicles (Fig. 327). Thus there is a continuous tunnel from end to end of the primary cerebro-spinal axis; and this persists as the central canal of the spinal cord and the ven- tricles of the brain, whose ciliated epithelium represents the ectodermic lining of the primitive neural canal. In the adult portions of the canal may become obliterated from an overgrowth of the lining cells, and the cilia are, if present at all, less distinct than in the child, and far less distinct than in the lower animals. From the wall of this canal is formed the cerebro-spinal axis, in which developing nerve-cells or neuro- B Fig. 327. — Diagram to illustrate the Formation of the Cerebral Vesicles. A. i indicates the cavity of the secondary fore- brain, which eventually becomes the lateral ventricles. In B the secondary fore-brain has grown backwards so as to overlap the other vesicles. I, first cerebral vesicle (primary fore-brain or 'tween brain) ; II, second cerebral vesicle (mid-brain); III, third cerebral vesicle (hind-brain) ; IV, fourth cerebral vesicle (after- brain). Fig. 326. — Neural Canal at a Later Stage (Beard). C, neural canal; G, posterior spinal ganglion. blasts soon become differentiated from the supporting cells or spongio- blasts, and wander outwards from the neighbourhood of the central canal (Fig. 338) till their further progress is checked by the barrier of the mar- 54 85o THE CENTRAL NERVOUS SYSTEM ginal veil, a closely-woven network or thicket, into which the processes of the spongioblasts break up at the outside of the primitive cerebro-spinal axis. Although the neuroblasts themselves are unable to penetrate the marginal veil, the axis-cylinder processes of some of them do so, and form the motor roots of the spinal nerves. The neuroblasts from which the fibres of the white columns of the cord are developed are apparently unable to send their axons through the marginal veil. They are accordingly forced to assume a longitudinal direction, and in this way the central grey matter becomes covered with a sheath of longitudinal white fibres. For a time only motor nerve-cells and the fibres connected with them are developed in the cerebro-spinal axis. The ganglia on the posterior roots arise from a series of ectodermic thickenings or sprouts from the neural crest which runs along the dorsal aspect of the neural canal. These sprouts contain the neuroblasts which develop into the spinal ganglion cells with the posterior root- fibres. From each pole of each neuroblast a process grows out, one towards the periphery, which forms a peripheral nerve-fibre, the other centrally to connect the cell with the cord. From the after-brain (or myelencephalon) is developed the medulla oblongata or spinal bulb, from the hind-brain (or metencephalon) the cerebellum and pons, from the mid-brain (or mesencephalon) the corpora quadrigemina and crura cerebri. The fore-brain, or primary fore-brain (thalamencepha- lon), gives rise of itself only to the third ventricle and optic thalamus; but a secondary fore-brain (telencephalon) buds off from it and soon divides into two chambers, from the roof of which the cerebral hemi- spheres, and from the floor the corpora striata, are derived. Their cavities persist as the lateral ventricles, which communicate with the third ventricle by the foramen of Monro. The olfactory tracts are formed as buds from the secondary fore-brain. To complete the story of the development of the brain, it may be added that the retina is really an expansion of its nervous substance. A hollow process, the optic vesicle, buds out on each side from the primary fore-brain. A button of ectoderm, which afterwards becomes the lens, grows against the vesicle and indents it so that it becomes cup-shaped, the inner concave surface of the cup representing the retina proper, the outer convex surface the choroidal epithelium. The stalk becomes the optic nerve. Histological Elements of the Central Nervous System. — The central nervous system is built up (i) of true nervous elements, (2) of sup- porting tissue. The nervous elements have usually been described as consisting of nerve-fibres and nerve-cells, but the antithesis of a time- honoured distinction must not lead us to forget that the essential part of a nerve-fibre, the axis-cylinder, is a process of a nerve-cell, and the medullary sheath a structure whose integrity is intimately related to that of the axis-cylinder.* In strictness, the term ' nerve- cell ' ought to include not only the cell-body, but all its processes, out to their last ramifications. But the habit of speaking of the position of the ccll-bodyf as that of the nerve-cell is so ingrained, that it seems better to continue the use of the latter term in its old signification, and to speak of the cell and branches together as a neuron (also spelled neurone). * While the medullary sheath, like the axis-cylinder, seems to be as regards its nutrition under the control of the nerve-cell, and must therefore be looked upon as an integral portion of the neuron, although not essential for its de- velopment, the neurilemma in respect both of its nutrition and its develop- ment appears to be an independent structure. f Foster and Sherrington call the cell-body the perikaryon. HISTOLOGICAL ELEMENTS 851 The Neurons. — A typical nerve-cell (Figs. 328, 330, 332)15 a knot of granular protoplasm, containing a large nucleus, inside of which lies a highly refractive nucleolus. A centrosome and attraction sphere (p. 5) have also been found in some nerve-cells, though not as yet demonstrated in all. Pigment may also be present, especially in old age. By certain methods of staining it may be shown that fibrils (neuro-fibrils) run through the protoplasm of the cell, forming a felt- work in it, and entering the dendrites on the one hand and the axis- cylinder process on the other (Figs. 328, 331, 335). In the axis- cylinders of nerve-fibres the fibrils (Fig. 329) appear to preserve their identity down to the distribution of the fibre. In the ground substance between the fibrils lie round, an- gular, or spindle-shaped bodies (Nissl's bodies) which stain with basic dyes (Fig. 340) . * These bodies vary in appearance in different kinds of nerve-cells, and in the same nerve-cell under different con- ditions. According to Macallum, they contain organically combined iron. In a multipolar cell, like those Fig. 328. — Anterior Horn Cell from Man showing Fibrils (Bethe). Fig- 329. — Medullated Nerve- Fibre showing Fibrils of Axis- Cylinder (Bethe). The fibrils are seen passing, without interrup- tion, across a node of Ranvier. in the anterior horn of the spinal cord, several processes — it may be five or six, or even more — pass off from the cell-body (Frontispiece). The most complete pictures of them are given by preparations impregnated according to the method of Golgif (Figs. 330, 333). One of the pro- * In Nissl's method the sections are stained in a solution of methylene blue, and decolourized in anilin-alcohol. f The method depends upon the deposition of mercury, or silver, in or around the cell-bodies and their processes in tissues which have been hardened in bichromate of potassium and then soaked in a solution of mercuric chloride or silver nitrate. In Pal's improvement of Golgi's method a solution of sodic sulphide follows the mercuric chloride. 852 THE CENTRAL NERVOUS SYSTEM cesses of most nerve-cells is distinguished from the rest by the fact that it maintains its original diameter for a comparatively great distance from the cell, and gives off comparatively few branches. This process, which in favourable preparations can be traced on till it becomes the axis-cylinder of a nerve-fibre, is called the axis-cylinder process, or more shortly the axon. The few slender brai ches that come off from it, usually at right angles, are called collaterals. The collaterals consist essentially of one or more fibrils of the axon. Both the main thread of the axon and the collaterals end by breaking up into an arborescent system of fibrils or telodendrion. The telodendrions vary greatly in appearance from simple end -brushes to far-branching thickets, or such special end-organs as motor plates (Fig. 335) or muscular spindles. The rest of the processes of the cell, which are termed dendrites or protoplasmic processes, very rapidly diminish in diameter, as they pass away from the cell by breaking up into Fig- 33O. — Multipolar Nerve-Cell: Golgi Preparation (Barker, after Kolliker). », axon; c, collaterals. fibrils like the branches of a tree. The Nissl bodies extend for some distance into the dendrites, but not into the axon. The dendrites of some cells, especially the pyramidal cells of the cerebral, and the Purkinje's cells of the cerebellar cortex, have small swellings, the so-called lateral buds or gemmules, on their course. Their signifi- cance is unknown. The dendrites terminate at a little distance from the cell, where they come into relation with the end -arborizations of the axons of other neurons. In this way two or more neurons are linked together to form a nervous path. According to the view most commonly held (neuron hypothesis), the relation is not one of actual anatomical continuity, but the processes come so close together that nerve impulses are able to pass across from the terminal brush of the axon of one nervous element to the dendrites or cell-body of another. This kind of junction is called a synapse. It has been suggested that the contact may be rendered more or less close through amoeboid movements of the dendrites, and that in this HISTOLOGICAL ELEMENTS way the nervous impulse may be switched like a railway-train from one path to another. But there is no experimental basis for this some- what crude, if fascinating, hypothesis. Sherrington has suggested that the presence of a ' membrane ' at the synapse may limit the con- duction and determine its direction. Some membranes, such as frog's skin, are known to possess a so-called irreciprocal permeability for certain substances, permitting them to pass more easily in one direc- tion than the other, and it is conceivable that a membrane at the synapse might have a similar action in respect to the movement of ions concerned in the propagation of the nervous excitation. Whatever A. Fig. 33I-— Nerve-Cells of Hirudo (Schafer, after Apathy). A, unipolar motor cell; a, network of neuro-fibrils near the sur- face of the cell ; b, near the nucleus n ; c, afferent, d, efferent neuro-fibril. B, bi- polar sensory cell a with its nucleus n ; cu, cuticle; ep, epidermis cells between which a neuro-fibril passes up from its branched ending near the surface of the skin to the nerve-cell, where it forms a network, which gives off a fibril passing towards the central nervous system. Fig. 332. — Large Pyramidal Cell of Cerebral Cortex (Barker, after Bech- terew). a, axon; b, dendrite. the nature of the relation between two superposed neurons may be, it does not permit the conduction of nerve-impulses indiscriminately in both directions. For instance, stimulation of the central end of the posterior root of a spinal nerve causes an electrical response (p. 824) in the anterior root of the same segment, while no electrical change is produced in the posterior root by stimulation of the anterior. We shall see later on (p. 872) that some of the fibres of the posterior root and their collaterals end by arborizing around the dendrites of the cells ot the anterior horn. The excitation is, therefore, able to pass from the 854 THE CENTRAL NERVOUS SYSTEM telodendrions of the posterior root-fibres through the dendrites of the anterior horn cells towards their cell-bodies, but not in the opposite Fig. 333. — * — e shows the development of the pyramidal nerve-cells of the cerebral cortex in a typical mammal; a, neuroblast with commencing ax on; b, dendrites appearing; d, commencing collaterals. A — D shows the different degree of com- plexity in the fully-developed pyramidal cells in different vertebrates: A, frog; B, lizard; C, rat; D, man (Donaldson, after Ram6n y Cajal). direction, and in general the direction of conduction is from the den- drites towards the cell-body. Some investigators believe that the fibrils already spoken of as forming a felt-work in the protoplasm of the nerve-cell may run right Fig. 334. — Cells from the Gasserian Ganglion of a Developing Guinea-Pig. originally bipolar cells are seen changing into cells apparently unipolar, same process occurs in the cells of the spinal ganglia (Van Gehuchten). The The through from one cell to another, thus constituting an actual anatomical connection between the neurons, and that such a connection may be established also by fibrils which do not enter the cells at all, but run in the intercellular substance of the grey matter. Such a continuity of HISTOLOGICAL ELEMENTS fibrils from cell to cell has been demonstrated in some of the invertebrates — e.g., in annelids (Fig. 331) — where previously the best examples of strictly iso- lated neurons were sup- posed to be found (Apathy) . The supporters of the theory of continuity look upon the cell-body as merely necessary for the nutrition of the nerve-net, but deny that it is neces- sary for the conduction of nerve-impulses. If this is the case, it is obvious that the neurons can no longer be considered as functional units in which the law of isolated conduction of nerve-impulses (p. 793) holds good. Nor is it by any means so easy to un- derstand as on the neuron hypothesis such facts as the strict limitation of Wal- lerian degeneration to the boundaries of the neurons directly affected, or the strict limitation of the silver reduction in Golgi preparations to single neu- rons. It is, of course, true that the simplicity and order introduced by the neuron hypothesis into our conceptions of the nervous conduction paths by no means prove its accuracy. Yet they are reasons for not lightly abandoning it, and it has recently been corroborated by important new evidence on the growth of nerve-cells on artificial media outside of the body (p. 802; Fig. 337, p. 857). Varieties of Neurons. — Nearly all the nerve-cells of the cerebrb-spinal axis agree with the cells of the anterior horn in the posses- sion of an axon and one or more dendrites, although sometimes the dendrites are scanty in number and Fig. 335. — Scheme of Lower Motor Neuron (Barker), a, h, axon-hillock (the portion of the cell from which the axon comes off), containing no Nissl bodies, and showing fibrillation; cue, axis-cylinder or axon; m, medullary sheath, outside of which is the neurilemma; c, cell- substance (cytoplasm), showing Nissl bodies in a lighter ground substance; d, protoplasmic processes or dendrites containing Nissl bodies; n, nucleus; »', nucleolus; n, R, node of Ranvier; s,f, side fibril; n of n, nucleus of the neurilemma; tel., motor end-plate; m', striped muscle-fibre; s, L, incisure. 856 THE CENTRAL NERVOUS SYSTEM insignificant in size. In the cerebral cortex the typical cells are oi pyramidal shape. From the base comes off the axon, and from the angles dendritic processes, a particularly massive dendrite proceeding from the apex of the pyramid towards the surface of the brain. Sometimes an axon, instead of ending in an arborization which comes into relation with the dendrites of another nerve-cell, or, as is more frequently the case, with the dendrites of more than one cell, breaks up into a sort of basket-work of fibrils surrounding the cell- body. The cells of Purkinje, for instance, in the cerebellum are sur- rounded by such pericellular baskets (Fig., 336). The cells of the spinal ganglia have two axons, which in the embryo arise one from each end of the bipolar cell, but in the adult, in all vertebrates except some fishes, are connected to the cell by a single process (Fig. 334)- It has been commonly held that the unipolar cell with a single T-shaped process is developed from a bipolar cell, which grows towards one side, so that the two processes come together and fuse. Such observations as that of Harrison on the bifurcation of the growing end of the main process of isolated nerve-cells culti- vated in vitro suggest an alternative and a simpler explanation — viz., that the T-shaped process is derived from the splitting of a single chief process. If this be the case, one of the original processes at the poles probably undergoes a retarded development or disappears, since the great majority of the spinal ganglion cells with the T-shaped process appear to have no dendrites. Another kind of cell which seems undoubtedly to be of nervous nature is the ' granule-cell.' Granule-cells are much smaller than the nerve-cells we have been describing. Their processes are much less easily followed, but all appear to give off an axon and several dendrites. They contain a relatively large nucleus (5 to 8 /j. in diameter), with only a mere fringe of cell-substance. The nucleus, unlike that of a large nerve-cell, stains deeply with hsematoxylin. Some parts of the grey matter are crowded with these granule-cells — e.g., the nuclear layer of the cerebellum and the substantia gelatinosa, or substance of Rolando, which caps the posterior horn in the cord. In other parts they are more thinly scattered, but probably they are as widely diffused as the large nerve-cells proper, and no extensive area of the grey matter is wholly without them. Although there are several varieties of granules (Hill), they all agree in this, that their axons run a comparatively short course, and never, or rarely, pass beyond the grey matter. Another kind of neuron which is also confined to the grey matter, and is typically seen in the cortex of the cerebrum and cerebellum, presents the peculiarity of an axon which branches into an intricate network immediately after coming off from the cell (cell of Golgi's second type). Unlike the long axon of the typical large nerve-cell, the axis-cylinder process of this Golgi cell remains unmedullated. The -sympathetic ganglion cells are developed from immature neuro- Fig. 336. — Pericellular Baskets (Schafer, after Cajal). Two cells of Purkinje from the cerebellum are seen sur- rounded by end ramifications forming a basket-work, b, de- rived from the branching of axons of small nerve-cells in the molecular layer; a, axon. HISTOLOGICAL ELEMENTS 857 blasts that migrate, in the course of development, from the rudiments of the spinal ganglia, and gathering in clumps form the ganglia of the sympathetic chain (His). They agree in general with the cells of the cerebro-spinal axis in possessing an axon and one or more, commonly several, dendrites, although a few of them are devoid of dendrites. The great majority of the axons remain unmedullated, but a few acquire a very fine medullary sheath. The epithelium lining the central canal of the cord and the ventricles of the brain has also been considered by some as of nervous nature. The fact that the deep ends of the cells are continued into processes which pierce far into the grey substance has been supposed to lend weight to this opinion, but there is no good ground for it. Growth of Neurons. — The growth of a neuron from origin to com- pletion is a comparatively slow process in the higher animals. Early in foetal life (about the third or fourth week in man) certain round germinal cells make their appearance amid the columnar ectodermic cells surrounding the neural canal. From their division are formed, in the first months of embryonic life, the primitive nerve-cells or neuroblasts. These soon elongate and push out processes, first the Fig. 33,7.— Isolated nerve-cells from the spinal cord of a tadpole growing in clotted Ivmph. A, B, C, are cells in different stages of growth. The lower view of C was drawn under the microscope 4$ hours later than the upper (Harrison). axon or axons, and then the dendrites (Fig. 333). The formation of the axons from the nerve-cell is most clearly followed in isolated cultures (Fig. 337). As development goes on, the cell-body grows larger, and the processes longer and more richly branched. The axon and its collaterals, when it has any, in the case of the great majority of the nervous elements of the brain and cord, ultimately acquire a medullary sheath, although, as we have said, the time at which medul- lation is completed varies in different groups of elements, and in some nervous tracts it is even wanting at birth. At birth, too, the branches of many of the cells are less numerous, and the connections between different nervous elements therefore less intimate than they will after- wards become. For many years the processes, and particularly the axons, continue not only to grow longer, but to grow thicker as well. The cell-body also enlarges, and the quantity of material in it that stains with basic dyes increases. In the growing (lumbar) spinal ganglia of the white rat the increase in volume of the largest cell- bodies is very closely correlated with the increase in area of the cross- section of the nerve-fibres growing out of them. The cross-section of the axis-cylinder is, and remains, almost exactly equal to the area 858 THE CENTRAL NERVOUS SYSTEM of the medullary sheath (Donaldson). Even after puberty is reached the anatomical organization of the nervous system may still continue to advance, although at an ever-slackening rate, and the finishing touches may only be given to its architecture in adult life. In old age the nervous elements decay as the body does. The cell- body diminishes in size; the stainable material lessens in amount; vacuoles form in the protoplasm and pigment accumulates ; the nucleus shrinks; the nucleolus is obscured or may disappear altogether. At the same time the processes of the cell, and especially the dendrites, tend to atrophy (Fig. 339). Nutrition of the Neuron. — We have already seen that when an axon is cut off from its cell-body, it and its medullary sheath, when it possesses one, undergo a rapid degeneration. It was long supposed Fig. 338. — Section through Half of Neural Tube (Barker, after His). The pear- shaped neuroblasts are seen migrating out- wards. The axons of some of them are seen pushing their way out through the marginal veil as the anterior root of a spinal nerve. Fig. 339. — i, spinal ganglion cells of a still-born male child; 2, of a man ninety-two years old ( x 250) — N, nuclei; 3, nerve-cells from the antennary ganglion of a honey-bee just emerged in the per- fect form; 4, of an old honey-bee. The nucleus is black in the figure. In 3 it is very large, in 4 it is shrunken and the cell-substance contains vacuoles (Hodge). that no change took place in the nerve-cell. The researches of recent years have shown that not only does loss of the specific function and trophic influence of the cell-body affect the nutrition of the axon, but loss of function of the axon reacts on the cell-body. In many cases at least, when a nerve-fibre is divided from its cell, characteristic changes are produced in the latter and in its dendritic processes, and they are scarcely less rapid, although usually less profound, and far more transient than the degeneration in the peripheral portion of the nerve-fibre. The cell-body and the nucleus swell. Many of the Nissl bodies (Fig. 340) disintegrate, and are reduced to a finely granular condition. After a time much of the disintegrated chromatic sub- stance disappears altogether. The nucleus may be displaced to one HISTOLOGICAL ELEMENTS 859 side of the cell. Certain changes in the neurofibrils of the cell may accompany the changes in the chromatin. In rabbits after division of the facial nerve the alterations in its nucleus of origin have been found to reach a maximum in about three weeks, after which there is a tendency to recovery on the part of the majority of the cells, even when regeneration of the nerve has been prevented by cutting out a portion of it. Some of the cells may completely atrophy and disappear. Similar changes have been found by Warrington in the motor cells ol the anterior horn after section of the posterior (dorsal) spinal roots. Since in this case no anatomical injury has been inflicted on the motor neurons, it has been surmised that the cause of the alterations is the loss of impulses which normally reach them along their dendrites. In short, we may say, with Marinesco, that the functional and anatomical integrity of the neuron depends on the integrity of all its constituent parts, and of the neurons which carry to it functional excitations — i.e., excitations connected with its proper physiological work. The neuron, in fact, lives by its function, or, in common language, by doing its work. Yet the anatomical tokens of mere disuse, as in the motor cells Fig. 340. — Cells from the Nuclei of the Oculo-Motor Nerves of the Cat Thirteen Days alter Division of the Root-Fibres on one Side: Nissl's Stain (Barker, after Flatau). a, normal cell from side on which the roots were not cut ; b, cell from side operated upon. Only a few Nissl bodies are present in b, and the nucleus is displaced to one side of the cell. of the anterior horn after division of the cord at a higher level, are less distinct than those which follow section of the axon. Therefore it must be concluded that the latter, although not indispensable for the nutrition of the cell as the cell is for the axon, exerts an influence upon it. Similar changes in the chromatin may also be produced in nerve- cells by a period of anaemia, in extensive superficial burns, in tetanus caused by the injection of bacterial cultures, in acute alcoholic poisoning, in fatigue, and in other ways. According to Wright, the inhalation of ether or chloroform (in dogs) so alters the chromatic substance that it loses its affinity for aniline dyes. In long-continued anaesthesia the nucleus is also affected, while the nucleolus is the last part of the cell to suffer. A greater alteration occurs in the cells in the three hours between the sixth and ninth hours of anaesthesia than in the five hours between the first and sixth. Although the changes are transitory, the cells, after a narcosis of nine hours, being practically normal in forty- eight hours, they indicate that the duration of safe surgical anaesthesia has a limit measured by hours. It is probable that the alterations in the chromatic substance should THE CENTRAL NERVOUS SYSTEM not be looked upon as the token of any specific lesion; they are the common structural response of the cell to injurious influences of the most varied nature. Grey and White Matter. — Nerve-cells are the most distinctive his- tological feature of the grey nervous substance. Sown thickly in the cerebral cortex, the basal ganglia, the floor of the fourth ventricle, and the cervical and lumbar enlargements of the cord, they are scattered more sparingly wherever the grey matter extends. They also occur in the spinal ganglia, and their cerebral homologues (such as the Gas- serian ganglion), in the ganglia of the sympathetic system, and the sporadic ganglia in general. But wide as is their distribution, and great as is the size of the individual cells, some of which have a diameter of 140 //, or even more, they yet make up but a small portion of the whole of the central nervous substance, the total weight of the 9,000 millions of nerve-cell bodies in the human brain being less than 27 grammes (Donaldson). And although it is not to be wondered at that objects so notable when viewed under the microscope should have struck the imagination of physiologists, it is probable that the very high powers which it is so common to attribute exclusively to them are, in part at least, shared with the network or feltwork formed by their processes. The grey matter, in addition to this ex- ceedingly delicate feltwork of non-medullated fibres and filaments representing the den- drites and such axons and collaterals as terminate within itself, contains also, as may be seen in preparations stained by Weigert's method,* great numbers of exceedingly fine medullated fibres, many of which are the collaterals of fibres that are passing out to the white matter. Only medullated nerve-fibres are met with in the white matter of the cerebro-spinal axis. They are commonly stated to be devoid of a neurilemma (or neurolemma), and in the sense that there is no continuous separate mem- branous sheath corresponding to the sheath of Schwann of the peripheral medullated fibres this is correct. Sheath cells, however, are present, and form a reticulum around each fibre in the meshes of which myelin is contained. In diameter the medullated fibres of the white matter vary from 2 /* to 20 /*. In Malapterurus electricus the fibre in the cord which supplies the electrical organ is of immense size; and in the anterior column of many fishes may also be seen a single gigantic fibre on each side with a diameter of nearly 100 ft. It cannot be said that any relation between the functions of neurons and the calibre of their axons has been definitely established. Many afferent fibres, it is true, are small — this is notably the case with the fibres of the posterior column, and many motor fibres are large. But the distinction can by no means be generalized, for the fibres of the direct cerebellar tract (p. 866), which certainly are afferent, are amongst the largest in the spinal cord ; and the vaso-motor fibres, which pass from the cord by the anterior (ventral) roots (Fig. 341) into the sympathetic, are smaller than the fibres of the posterior column. Even the motor * Weigert's is a special method of staining the medullary sheath with haematoxylin. Fig. 341. — Transverse Section of a Bundle of Nerve- Fibres from the Anterior (Ventral) Root of the First Coccygeal Nerve of the Cat (Dale). The great differ- ence in the diameter of the fibres is well shown. The small fibres are vaso-motor. 86 1 nerve-fibres of striated muscles vary considerably in diameter, those of the tongue, e.g., being smaller than those of the muscles of the limbs. Further, the medullated fibres of the brain are, without reference to function, in general finer than the fibres of the cord. As a rule the fibres whose course is the longest are the thickest, but the rule is often broken. For example, the average diameter of the fibres going to the thigh of the frog is greater than that of the fibres going to the lower part of the limb (Dunn). The cause of these differences in the size of nerve-fibres is quite unknown. It is more likely to be morphological than physiological. Supporting Tissue. — The protective membranes of the central nervous system consist of ordinary connective tissue derived from the meso- derm. The supporting framework which interpenetrates the nervous substance consists of a peculiar form of tissue derived from the ecto- derm, and called neuroglia. The whole cerebro-spinal axis is wrapped in four concentric sheaths. Next the walls of the bony hollow in which it lies is the dura mater. Next the nervous substance itself, following the convolutions of the brain and the fissures of the cord, and giving off bloodvessels to both, is the pia mater. Between the dura and the pia, separated from the latter by a jacket of cerebro-spinal fluid, is the double layer of the arachnoid. The comparatively coarse septa that run into the nervous substance as if coming off from the pia mater are the main beams in the scaffolding of non-nervous material with which that substance is interwoven, and by which it is supported. The interstices are filled in by a thick-set f eltwork of interlacing neurog- lia fibres, which lie close against the small glia cells. In preparations impregnated by the Golgi method many of the neuroglia fibres appear to be processes running out from the attenuated cell-body like the arms of a microscopic crab or spider. But this is a deceptive appear- ance, as has been shown by means of special methods in which the neuroglia fibres are alone stained (Weigert, Huber, etc.). They generally lie in close contact with, or embedded in, the protoplasm of the neuroglia cells, from which they have become differentiated structurally and chemically, but sometimes they may detach themselves entirely from the cells and lie free in the intervening tissue. The neuroglia is present in greatest abundance in the grey matter immedi- ately surrounding the central canal of the cord and the ventricles of the brain (the ependyma, as it is called), from which long neuroglia fibres pass out radially, giving off branches on their course, and ending in little knobs or enlargements attached to the pia mater. SECTION II.— GENERAL ARRANGEMENT OF THE GREY AND WHITE MATTER IN THE CENTRAL NERVOUS SYSTEM. (1) Around the central canal, as we have seen, a tube of grey matter sheathed with white fibres is developed. This tube, from optic thalamus to conus medullaris, may be conveniently referred to as the central grey axis or stem, which, in the lowest vertebrates — e.g., fishes — is much the most important part of the central nervous system. (2) On the outer surface of the anterior portion of the neural axis, but not in the part corresponding to the spinal cord, is laid down a second sheet or mantle of cortical grey matter. Between 862 THE CENTRAL NERVOUS SYSTEM this and the primitive grey stem are interposed (a) the sheath of white fibres that clothes the latter, and connects its various parts, and (b) a new development of white matter (corona radiata, cere- bellar peduncles), which serves to bring the cortex into relation with the primitive axis, and through it with the rest of the body. Although there are histological and developmental differences between the cerebral and the cerebellar cortex, we may, for some purposes, classify them together as cortical formations. And we may also include under this head the corpora striata, which, although for descriptive purposes generally grouped with the optic thalami and the other clumps of grey matter at the base of the brain, as the basal ganglia, are to be regarded as cortical in character. As we mount in the vertebrate scale the cortex formation of the secondary fore-brain and hind-brain acquires prominence. In other words, the grey matter developed in the roof of the cerebral vesicles I. and III. (Fig. 327) (the grey matter of the cerebral and cere- bellar cortex) comes to overshadow the superficial grey matter hitherto present only in the roof of vesicle II. (in the corpora bigemina). And this cortex formation becomes larger in amount, and, in the case of the cerebral grey matter, more richly convoluted, the higher we ascend, until it reaches its culmination in man. As the anterior cerebral vesicles develop, they spread continually backward, until at length the cerebral hemispheres cover over, and almost completely surround, the primary fore-brain and the mid- and hind-brains, so that the anterior portion of the primitive stem comes, as it were, to be invaginated into the second wider tube of cortical grey matter. This development of the cortical grey substance is accompanied with a corresponding development of nerve-fibres, for an isolated nerve-cell (apart, of course, from possible embryonic rudiments which have not undergone com- plete development) is no more conceivable than a railway-station the track from which leads nowhere in particular, or a harbour on the top of a hill. But it is to be particularly observed that the new formation does not supplant the old, but works through and directs it. The neuro- blasts of the cortex do not throw out their axons to make direct junc- tion with muscles and sensory surfaces. Such junction the cortex finds already established between the primitive cerebro -spinal axis and the periphery. It joints itself on by nerve-fibres to the cells of the central stem; and we have reason to believe that no single axon in an ordinary spinal or cranial nerve* runs all the way from the periphery to the cortex, and no axon of a cortical nerve-cell all the way from the cortex to the periphery, but that the connection is made by a chain of at least two neurons, the cell-body of one of which is situate in this primitive grey tube. The fibres from the cortex of each cerebral hemisphere (corona radiata), radiating out like a fan below the grey matter, are gathered together into a compact leash as they sweep down through the isthmus of the brain in the internal capsule, to join the crura cerebri. The cortex of each cerebellar hemisphere, and the ribbed pouch of grey * The olfactory and possibly to some extent the optic nerves are exceptions to this statement. Their relation to the cortex, as is easily understood from the manner of their development (p. 850), is different from that of the other nerves. GREY AND WHITE MATTER IN THE SPINAL CORD 863 matter, known as the corpus dentatum, which is buried in its white core, are also connected by strands of fibres with the central stem and the cerebral mantle. The restiform body or inferior peduncle brings the cerebellum into communication with the spinal cord. The superior peduncle by one path, and the middle peduncle by another, connect it with the cerebral cortex. A great transverse commissure, the corpus callosum, unites the cerebral hemispheres across the middle line, while transverse fibres, that break through the middle lobe or worm, form a similar though far less massive junction between the two hemispheres of the cerebellum. The fibres of the nervous system may be divided into (i) fibres connecting the peripheral organs with nerve-cells in the central grey axis; (2) fibres connecting nerve-cells in this central axis with cells in the external or cortical grey tube; and (3) fibres linking cortex with cortex, or central ganglia with each other. In the third group are included (a) fibres which connect portions of the cortex on the same side (association fibres) ; (b) fibres which connect portions on opposite sides of the middle line (commissural fibres) ; (c) fibres which connect the central grey matter at different levels — e.g., the proprio- spinal or endogenous fibres of the cord. Our first task is, therefore, to trace the peripheral nerves to their cells of origin or centres of reception* in the nervous stem. And although there is reason to believe that the whole of the peripheral nerves, cerebral and spinal (with the exception of the olfactory and optic, which are rather portions of the brain than true peripheral nerves), form a morpho- logical series, it will be well to begin with the spinal nerves, since their motor and sensory fibres are gathered into different and definite roots, whose course within the cord is, in general, more easily traced than the course of the cerebral root-bundles within the brain. SECTION III. — ARRANGEMENT OF THE GREY AND WHITE MATTER IN THE SPINAL CORD. The grey matter of the spinal cord is arranged on each side in a great unbroken column of roughly crescentic section, joined with its fellow across the middle line by a grey bar or bridge, which springs from the convexity of the crescent, and is pierced from end to end by the central canal. The anterior horn of the crescent, although it varies in shape at different levels of the cord, is, in general, broad and massive, in comparison with the slender and tapering posterior horn. In the lower cervical and upper dorsal region a moulding or projection, forming a lateral horn, springs from the fluted outer side of the grey substance. Within the grey matter nerve-cells are found, sometimes so regularly arranged that they form veritable cellular or vesicular strands. Of these the best * The centre or nucleus of reception of a nerve contains the nerve-cells around which its axons terminate; the nucleus of origin of a nerve contains the cells from which its axons arise. 864 THE CENTRAL NERVOUS SYSTEM marked are — (i) The tract or tracts made up by the cells of the anterior horn (Fig. 342), which practically run from end to end of the cord, swell out in the cervical and lumbar enlargements, where the cells are very numerous and of great size (70 jj, to 140 JJL in diameter), and contract to a thin thread in the thoracic region, where they are relatively few, scattered, and small. In the enlargements there are several groups of these cells corresponding with the segments of the limbs, the movements of the hand, forearm, and upper arm being each repre- sented by a group in the StillinGS Cervical cerv^ca^ an<^ those of the foot, Nucleus leg> and tnigh by groups in CeruLCaL *he lumbar swelling. In the Enlargement rll- Lateral &ft- column (.column of- jfe infer- txxt) rest of the well-marked are present in horn, a cord only groups of two cells the anterior mesial and a lateral. Stilling's dorsal nucleus or Clark's Column Cefls of 1tx. anterior Cornu -Scattered cells cf intermedia- tract (2) Clarke's column, whose cells, mostly of good size and somewhat rounded in outline, are situated at the inner side of the root of the posterior horn just where it joins on to the grey cross-bar. It gradu- ally increases in size from above downwards, usually appearing first at the level of the seventh or eighth cervical nerve, attaining its maximum development at the eleventh or twelfth dorsal and dis- appearing altogether, as a continuous strand, at the level of the second or third lumbar nerves. Scattered nerve-cells, however, constituting the so- called cervical and sacral nuclei of Stilling, are frequently found occupying the same position towards the upper and lower ends of the cord, and may be looked upon as isolated portions of Clarke's column. (3) A tract of small cells calle4 the intermedia- lateral tract, lateral cell column, or lateral Jiqrn* situated at the outer edge of the grey matter, about midway between the anterior and pos- terior horns. It is best marked in the thoracic region, up to about the second thoracic segment, although jii the corresponding situa- tion there are scattered cells in the lumbar swelling and the cervical Sacral Fig. 342. — Diagram of Grey Tracts of Cord. GREY AND WHITE MATTER IN THE SPINAL CORD 863 cord. There is reason to believe that the axons of cells of the inter- medio-lateral tract, which pass out as small medullated fibres in the anterior roots, form the preganglionic segments of the efferent vascular and visceral nerves (p. 185)- (4) The cells of the posterior horn, which, although numerous, are smaller than those of the anterior horn. Throughout the whole cord, however, two small groups of cells may be distinguished, one on the lateral side of the horn, about its middle, and the other on the mesial side, a little in front of — i.e., ventral to — the edges of the substance of Rolando. Both of these groups are broken up by the passage through them of bundles of fibres which form a network, and they are therefore called respectively the group of the lateral and the group of the posterior reticular formation. The white matter of the cord is anatomically divided by the position of the nerve-roots and the anterior and posterior fissures Fig. 343. — Diagrammatic Section of the Spinal Cord in the Cervical and Lumbar Enlargements, to show Tracts of Fibres (Starr), i, antero-median column; 2, antero-lateral column; 3, ascending antero-lateral or Gowers' tract; 4, mar- ginal tract (ground bundle, consisting of short endogenous fibres); 5, lateral or crossed pyramidal tract; 6, direct cerebellar tract; 7, tract of Lissauer; 8, ex- ternal portion, and 9, root zone, of Burdach's column; 10, comma tract; u, pos- terior commissural tract; 12, Coil's column; 13, septo-marginal tract. into three columns on each side: the anterior, lateral, and posterior columns. The first two, since they are not separated by a perfectly definite boundary, are often grouped together as the antero-lateral column. In the cervical region it may be seen with the microscope that the posterior white column is almost bisected by a septum running in from the pia mater towards the grey commissure. The inner half is called the postero-median column, or column of Goll; the outer half the postero-external column, or column of Burdach (Fig. 343). No localization of any of the other conducting paths in the cord is possible by gross anatomical examination; but by means of the developmental method and the method of degenera- tion the columns of Goll and Burdach can be followed throughout the cord, and several similar areas can be mapped out. We shall only mention those that are physiologically the most important. 55 866 THE CENTRAL NERVOUS SYSTEM When the spinal cord is divided, and the animal allowed to survive for a time, certain tracts are picked out by the degeneration of their fibres, although in every degenerated tract some fibres remain un- affected. We may distinguish the tracts that degenerate above the lesion (ascending degeneration) from those that degenerate below the lesion (descending degeneration). Ascending Tracts. — Above the lesion degeneration is found both in the posterior and the antero-lateral columns. Immediately above the section nearly the whole of the posterior column is in- volved. Higher up the degeneration clears away from Burdach's tract, and, shifting inwards, comes to occupy a position in the column of Goll. In the antero-lateral column two degenerated regions are seen, both at the surface of the cord, one a compact, sickle-shaped area extending forwards from the neighbourhood of the line of entrance of the posterior roots, and the other an area of scattered degeneration, embracing many intact fibres, and complet- ing the outer boundary of the column almost to the anterior median fissure. The compact area is called the dorsal or direct cerebellar tract, or tract ofFlechsig (or the fasciculus cerebello-spinalis),the diffuse area the antero-lateral ascending tract, or tract of Gowers, or ventral cerebellar tract (or the fasciculus antero-lateralis superficialis) .* The dorsal cerebellar tract is distinguished by the large size of its fibres. It is only distinct in the dorsal and cervical regions of the cord. The tract of Lissauer, or posterior marginal zone, is another small ascend- ing tract at the outer side of the tip of the posterior horn. It is made up of fine fibres from the posterior roots which soon pass into the posterior column. Descending Tracts. — When the cord is divided, say, in the upper dorsal or cervical region, the following tracts degenerate below the lesion : (1) A small group of fibres close to the antero-median fissure, which has received the name of the direct pyramidal tract — pyramidal because higher up in the medulla oblongata it forms part of the pyramid ; direct, because it does not cross over at the decussation of the pyramids, but continues down on the same side. In the stan- dard anatomical nomenclature it is termed the fasciculus cerebro- spinalis anterior. The direct pyramidal tract is only present in man and the higher apes. (2) A tract of degenerated fibres in the posterior part of the lateral column. This is the lateral or crossed Pyramidal tract (or the fasciculus cerebro-spinalis later alis),a.nd is muchlarger thanthedirect. In the medulla it also lies within the pyramid, but, unlike the direct pyramidal tract, it crosses to the opposite side of the cord at the decussation. The pyramidal tracts are also called cortico- spinal to indicate their origin and termination. * Some writers employ the terms dorsal and ventral spino cerebellar tracts. GREY AND WHITE MATTER IN THE SPINAL CORD 867 D. C (3) A tract of scattered degeneration lying along the margin of the cord in the anterior portion of the antero-lateral column, and partly overlapping the tract of Cowers. It is called the antero-lateral descending tract, or tract of Loewenthal, or the vestibule- spinal tract. (4) The prepyramidal (or rubro-spinal) tract, or Monakow's tract (also called the fasciculus intermedio-lateralis) , lying immediately in front of the crossed pyramidal tract. (5) A small, comma-shaped island of degeneration (comma tract] can be followed downwards for a short distance in the middle of Burdach's column. It is only seen in the cervical and upper thoracic regions. Less well known descending tracts are — (6) The olivo - spinal and thalamico - spinal tracts (or Helweg's bundle) in the anterc- lateral column opposite the head of the anterior horn. This tract does not pass down beyond the lower cervical re- gion. The olivo-spinal tract appears to consist of fibres running down from the olivary body into the cord, while the thalamico-spinal tract is made up of descending fibres origina- t/ng in the oftic thala'mu, This is an important tract in the lower vertebrates, but not in man. (7) The tract of Marie in the anterior column is chiefly a continuation into the cord of the posterior longitudinal bundle, one of the conspicuous tracts of the brain-stem or upper portion of the cerebro-spinal axis (p. 885). It contains both ascending and descending fibres. When we have deducted the long ascending and descending tracts which have been described, there still remains in the antero-lateral column a balance of white matter unaccounted for. This white substance, which does not degenerate for any great distance either above or below a lesion, is called the antero-lateral ground-bundle, and lies chiefly in the form of an incomplete ring around the grey matter. For descriptive purposes it is sometimes distinguished as the anterior ground-bundle (or fasciculus anterior proprius) in the anterior column, and the lateral ground-bundle (or fasciculus later alts proprius) in the lateral column. It is believed to consist of fibres (endogenous or proprio-spinal fibres) which run only a comparatively short course in the cord, and serve to connect nerve-cells at different levels. Some of these endogenous fibres are ascending, others V.B Fi£- 344.-Scheme of Cross-Sectton of Spinal ord (Donaldson, after Lenhossek). On the right side DR> posterior (dorsal) root; VR anterior (ventral) root; C.P., crossed pyramidal fibres; C., direct cerebellar tract; A.L., antero-lateral tract; B.C., posterior columns. 868 THE CENTRAL NERVOUS SYSTEM descending. The septo-marginal bundle consists also largely of fibres which begin and end in the cord (proprio-spinal fibres) . Some endog- Fig. 345 . — Medulla Oblongata, Pons and Corpora Quadrigemina (Dorsal or Pos- terior View) (Sappey). i, corpora quadrigemina ; 2, nates; 3, testes; 4, anterior brachium uniting the nates to the lateral geniculate body; 5, pos- terior brachium uniting the testes to the internal geniculate body 6 ; 7, pos- terior commissure; 8, pineal gland pulled forward to show nates; 9, su- perior peduncle of the cerebellum ; 10, ii, 12, valve of Vieussens; 13, troch- lear nerve; 14, lateral sulcus; 15, fillet; 16, superior, 17, middle, and 18', in- ferior, peduncle of the cerebellum; 19, floor of fourth ventricle; 20, audi- tory nerve ; 21, spinal cord; 22, postero- median column, continued hi the me- dulla as the funiculus gracilis; 23, the clava, the continuation of the funiculus gracilis. Fig. 346. — Medulla Oblongata, Pons and Crura Cerebri (Ventral or An- terior View). i, infundibulum ; 2. tuber cinereum; 3, corpus mam- inillare; 4, cerebral peduncle or crus cerebri; 5, pons; 6, middle peduncle of cerebellum; 7, pyramid; 8, decus- sation of pyramids; 9, olive; 10, tu- bercle of Rolando ; n, external arcuate fibres ; 12, upper end of cord ; 13, ligamentum denticulatum ; 14, dura mater of cord; 15, optic tract; 16, chiasma; 17, third nerve; 18, fourth nerve; 19, fifth nerve; 20, sixth nerve; 21, seventh. nerve; 22. eighth nerve; 23, nerve of Wrisberg (portio intermedia), which unites with the facial; 24, glosso-pharyn- geal nerve; 25, vagus; 26, spinal accessory; 27, hypoglossal; 28, 29, 30, first, second, and third pairs of cervical spinal nerves. enous fibres may also be intermingled with the fibres of certain of the long tracts, both in the antero-lateraland posterior columns, and GREY AND WHITE MATTER IN CEREBRO-SPINAL AXIS 869 Sherrington has shown (in the dog) that long proprio-spinal fibres passing down in the lateral column connect the upper with the lower parts of the cord (p. 908). The next question which arises is: How are the long tracts con- nected below — i.e., with the periphery — and above — i.e., with the higher parts of the central nervous system ? The answer to this question, partly derived from clinical records and partly from experimental results, is in the case of some of the tracts unexpectedly full and minute, though meagre in regard to others. But to render it intelligible it is necessary, first of all, to describe briefly — SECTION IV. — ARRANGEMENT OF GREY AND WHITE MATTER IN THE UPPER PORTION OF THE CEREBRO-SPINAL Axis. In the medulla oblongata the grey and white matter of the spinal cord is rearranged, and, in addition, new strands of fibres and new nuclei of grey substance make their appearance. Of these nuclei the most conspicuous is the dentate nucleus of the in- ferior olive, whi ch , covered by a crust of white fibres, appears as a projection on j the antero -lateral surface of the medulla. In front of the olive, between it and the continuation of the anterior median fis- sure, is another projec- tion, the pyramid, which looks like a prolongation of the anterior column of the cord, but is made up of very different consti- tuents. Dorsal to the olive is the restiform body or inferior peduncle of the cerebellum, and behind the restiform body lie two Fig. 347 —Medulla Oblongata and Cerebellum, with thin columns, the fumcu- Fourth Ventricle (Hirschfeld). i, mesial groove his cuneatus, which con- of floor of ventricle running down to the calamus tinues the postero-exter- scriptorius; 2, striae acusticae; 3, inferior peduncle nal column of the cord, of the cerebellum; 4, clava; 5, superior peduncle and thefuniculus gracilis, crossing the inferior and passing to its internal which continues the pos- side; 7, 7. lateral sulcus; 8, corpora quadrigemina. tero-internal column. In these f uniculi are contained collections of small or medium-sized nerve- cells termed respectively the nucleus cuneatus and the nucleus gracilis. The rearrangement of the constituents of the cord is due mainly to two causes: (i) The opening up of the central canal to form the fourth ventricle, and the folding out, on either side, of the grey matter which lies posterior to it in the cord; (2) the breaking up of the grey matter of the anterior horn by strands of fibres as they sweep through it from the lateral pyramidal tract to take up a position in the pyramid of the opposite side (decussation of the pyramids), and a little higher up by 870 THE CENTRAL NERVOUS SYSTEM fibres passing across the middle line from the gracile and cuneate nuclei (sensory decussation or decussation of the fillet). The mosaic of grey and white matter formed in the medulla by the interlacing of longi- tudinal and transverse fibres with each other and with the relics of the anterior horn, is called the reticular formation (formatio reticularis). It occupies the anterior and lateral portions of the bulb behind the pyramids and olivary bodies, and is continued upwards in the dorsal portion of the pons and crura cerebri, and downwards for a little way into the upper part of the cervical cord. The cerebro-spinal axis passes up from the medulla through the pons, encircled and traversed by the transverse pontine fibres derived from the middle cerebellar peduncle or commissure, which enclose every- where between them numerous collections of nerve-cells (nuclei pontis). Enlarged by the accession of many of these fibres which come from the cortex of the cerebellum on the opposite side, as well as of fibres from the nuclei of the cranial nerves that take origin in this neighbourhood (fifth and eighth), the central nervous stem bifurcates above the pons into the two divergent crura cerebri. From each crus a great sheet of fibres passes up between the optic thalamus and the caudate nucleus of the corpus striatum on the one hand, and the globus pallidus of the lenticular nucleus on the other, as the internal capsule, from which they are dis- persed, in the corona radiata, to the cerebral cortex. Both in the upper part of the pons and in the crus a ventral portion, or crusta, containing the fibres of the pyramidal tract, and a dorsal portion, or tegmentum, can be distinguished, the line of separation being marked in the crus by a collection of grey matter, called^ from its usual, though not invariable, colour the substantia nigra (Fig. 352). A portion of the tegmentum is continued below the optic thalamus. SECTION V. — CONNECTIONS OF THE LONG PATHS OF THE CORD. Coming back now to our question as to the connections of the long tracts of the cord, let us consider, first of all, The Connections of the Postero-Median and Postero- External Columns. — When a single posterior root is divided, say, in the dorsal region, between the cord and the ganglion, its fibres, as we have already seen (p. 797), degenerate above the section. Since the cell- bodies of these neurons lie in the ganglion, if a series of microscopic sections of the spinal cord be made, well-marked degeneration will be found at the level of entrance of the root on the same side of the cord, while below that level there will be only a few degenerated fibres in the comma tract. Immediately above the plane of the divided root the degeneration will be confined to Burdach's column and to its external border. Higher up it will be found in the internal portion of Burdach's and the external rim of Coil's column. Still higher up the degenerated fibres will be confined to the postero- median column ; the postero-external will be free from degeneration. When a number of consecutive posterior roots are cut, the whole of the postero-external column in the sections immediately above the highest of the divided roots will be found occupied by degene- rated fibres, while Coil's column may be free from degeneration, or CONNECTIONS OF THE LONG PATHS OF THE CORD 871 degenerated only at its outer border. Higher up degeneration will be found to have involved the whole of the postero-median column, and to have cleared away altogether from the postero-external. The degeneration in the column of Goll may be traced along the whole length of the cord to the medulla, although the number of degenerated fibres diminishes as we pass upward. The explanation of these appearances 'is as follows : It may be seen in preparations of the cord impregnated by Golgi's method that the fibres of the posterior roots soon after their entrance into the cord divide into two processes, one of which runs up and the other down in the posterior column, or in the adjoining portion of the posterior horn. From both of these collaterals are given off at intervals to the grey matter. The descending branches run downwards only for a short dis- tance, and the degeneration in the comma tract seen after section of the cord is due to the division of these branches. Many of the as- cending branches pass up for a short distance in the postero-external column, sweeping obliquely through it to gain the tract of Goll. In this tract some of them run right on to the medulla ob- longata, to end by arborizing among the cells of the nucleus gracilis. Other fibres, both of Coil's and of Burdach's tract, end at various levels in the cord, their collaterals, and ultimately the main branches themselves, coming into relation with nerve-cells in the grey matter. When the cervical posterior roots are cut, many of the degenerated fibres remain in Burdach's column up to the medulla, where they terminate Fig. 348. — Diagrams of Degeneration at Different Levels in the Cord after Sec- tion of a Number of Posterior Roots of Nerves forming the Lumbo-Sacral Plex- us (Mott). Fig. 349. — Branching of Posterior Root-Fibres in Cord (Donald- son, after Cajal). Collaterals, Col, are seen coming off from the two main branches of the root-fibres, DR, and ending in arborizations. CC, cells in the grey matter of the cord, whose axons also give off collaterals. 872 THE CENTRAL NERVOUS SYSTEM in the nucleus cuneatus. In the posterior column, then, the numerous fibres of the posterior roots which do not end in the spinal cord are arranged in layers, the fibres from the lower roots being nearest the median fissure (in the postero-median column), and those from the higher roots farthest away from it (in the postero-external column. Thus, in a section through the upper cervical region Goll's column is almost entirely composed of fibres from the posterior limb, while the column of Burdach consists of fibres from the anterior limb. Other collaterals from the posterior root-fibres, and many of the main root-fibres themselves, run into the anterior horn and terminate in arborizations around its cells; some pass into the posterior horn, and doubtless come into relation with its scattered cells and, in the dorsal region, with the cells of Clarke's column. Some of the posterior root-fibres and their collaterals also form synapses with the cells of the intermedio-lateral tract. Other collaterals and probably some axons cross the middle line in the anterior and posterior commissures and end.in the grey matter of the opposite side. Connections of the Direct or Dorsal Cerebellar Tract. — Since the dorsal or direct cerebellar tract does not degenerate after section of the posterior nerve-roots, but does degenerate above the level of the lesion after section of the spinal cord, the nerve-cells from which its axons arise must be situated somewhere or other in the cord. Now, it has been observed that the vesicular column of Clarke first becomes prominent in the lower dorsal region, and that in this same region the direct cerebellar tract begins. Atrophy of the cells of Clarke's column has sometimes in disease been shown to accompany de- generation of the direct cerebellar fibres. After an experimental lesion of these fibres in animals, some of the cells of the vesicular column show the changes in the Nissl bodies and the other changes which we have already described as occurring in nerve-cells whose axons have been cut. After two or three months these cells may be found almost completely atrophied (Schafer). Finally, axis- cylinder processes have been seen sweeping out from Clarke's column into the direct cerebellar tract (Mott). The evidence, then, is complete that the cells of origin of this tract are in Clarke's column. Clarke's cells are surrounded by arborizations, some of which, as previously stated, represent the terminations of posterior root-fibres and of their collaterals. The neurons whose axons run in the dorsal cerebellar tract are therefore the second link in an afferent path. The direct cerebellar tract runs right up to the cerebellum through the restiform body, without crossing and without being further interrupted by nerve-cells. The restiform body ends partly in the dentate nucleus of the cerebellum, partly in the vermis, and among the fibres which end in the vermis are those of the direct cerebellar tract. In the dorsal cerebellar tract there is a definite stratification CONNECTIONS OF THE LONG PATHS OF THE CORD 8/3 of the fibres: the fibres from the lowest segments of the cord lie outermost ; beneath these come fibres from the lowest thoracic, seg- ments, then fibres from the higher thoracic segments; and, internal to all, fibres from the topmost thoracic and lowest cervical segments. Connections of the Antero-Lateral Ascending Tract. — According to Schafer, the axons of this tract are probably connected "with cells situated in the middle and posterior parts of the grey crescent, mainly nX' .p.m.f. ff ».y f.c. n.ZT. c.c. u.m.f. Fig. 350. — Transverse Section of Medulla Oblongata at the Level of the Decussa- tion of the Fillet (Halliburton, after Schwalbe). a.m.f, anterior, and p.m.f, posterior, median fissure; f.a and /.a2, external arcuate fibres; f.a', internal arcuate fibres becoming external; n.a.r, nuclei of arcuate fibres; py, pyramid; o, o', lower end of nucleus of olive ; f.r, formatio reticularis; n.l, lateral nucleus; n.g, nucleus gracilis; f.g, funiculus gra- cilis; n.c, nucleus cuneatus; n.c', external cuneate nucleus ; f.c, funiculus cuneatus; g, substance of Rolando; c.c, central canal surrounded by grey matter ; n. XI, nucleus of spinal accessory; n.XII, of hypoglos- sal; a.V, ascending root of fifth nerve; s.d, the decussation of the fillet, or superior decussation. i§>- 35i — Transverse Section of Medulla Oblongata at about the Middle of the Olive (Schwalbe). f.l.a, anterior median fissure; n.a.r, arcuate nucleus; £., pyra- mid; n.XII, hypoglossal nucleus; XII, root bundle of hypoglossal nerve coming off from the surface ; at b it runs between the pyramid and the dentate nucleus of the olive, o; f.a.e, external arcuate fibres; n.l, lateral nucleus; a, arcuate fibres going to restiform body c.r, partly through the substantia gelatinosa g, partly superficial to the ascending root of the fifth nerve a.V; X, root-bundle of vagus; h.X, n.X', two portions of vagus nucleus; f.r, for- matio reticularis; n.g, nucleus gracilis; n.c, nucleus cuneatus; n.t, nucleus of the funiculus ter'es;'».am, nucleus ambiguus; r, rap he; o', o", accessory olivary nucleus; p.o.l, peduncle of the olive. on the opposite side of the cord, although also on the same side. None of the fibres of the tract can come directly from the posterior nerve-roots, since no degeneration is seen in it on section of the roots alone. The antero-lateral ascending tract passes up through the medulla, where some of its fibres perhaps form synapses with the cells of the 874 THE CENTRAL NERVOUS SYSTEM lateral nucleus, a collection of grey matter in the lateral portion of the spinal bulb. But its main strand runs on unbroken through the medulla, in front of the restiform body, and behind the olive, and after reaching the upper part of the pons bends back over and in company with the superior peduncle as the ventral spino-cerebellar bundle, to end in the worm of the cerebellum (Fig. 363). A few fibres of Gowers' tract may pass by the middle peduncle to the opposite cerebellar hemisphere. Some of its fibres do not go to the cerebellum at all. One group can be followed to the corpora quadri- gemina (spino-tectal fibres], and another by way of the tegmentumof the cms cerebrt to the optic thalamus (spino-thalamic fibres) . Through the relay of the gracile and cuneate nuclei, the postero- internal and postero-external columns of the cord are further con- nected on the one hand with the cerebrum, and on the other with the cerebellum. The cells of the nuclei give off fibres (internal arcuate fibres) which, sweeping in wide arches across the mesial raphe to the opposite side, take up a position behind the pyramid in the tract of the fillet, a bundle of fibres which becomes more compact, and therefore more distinct, as it passes brainwards. Receiving fibres from other sources on its way, and also giving off fibres, it runs up- wards through the dor- sal or tegmental portion of the pons. In the mid- brain it divides into two portions, the lateral fillet, also called the lower fillet or fillet of Reil, and the intermediate, also called the upper fillet. The lateral fillet contains mainly fibres arising in the cochlear nucleus of the auditory never, and ends in grey matter of the pos- terior corpus quadrigeminum, and partly in the mesial geniculate body. It appears to be a path for the conduction of auditory impulses. The intermediate fillet contains chiefly the fibres that come off from the gracile and cuneate nuclei, but is enlarged by the accession of fibres from the sensory nuclei of the cranial nerves. It terminates in the lateral nucleus of the optic thalamus by forming synapses with nerve-cells, whose axons, passing through the posterior limb of the internal capsule and the corona radiata, continue the afferent path to the cerebral cortex. Not all of the axons from the cells of the cranial sensory nuclei run Fig. 352. — Diagrammatic Transverse Section of Crura Cerebri and Aqueduct of Sylvius, a, an- terior corpora quadrigemina 6, aqueduct; c, red nucleus; d, fillet; e, substantia nigra;/, pyramidal tract in the crusta of the crura cerebri; g, fibres from frontal lobe of cerebrum; h, fibres from tern- poro-occipital lobe ; i, posterior longitudinal bundle. CONNECTIONS OF THE LONG PATHS OF THE CORD 875 in the fillet. Many of them occupy a position in the reticular forma- tion of the tegmentum dorsal to the fillet as they pass through the pons and mid-brain to end in the thalamus and the region below it (sub-thalamic region). From the sensory nucleus of the fifth nerve a separate bundle of fibres ascends to the thalamus, in the tegmen- tum of the mid-brain lateral to the posterior longitudinal bundle. Connections of the Pyramidal Tracts. — When the cortex in and in front of the fissure of Rolando is destroyed by disease in man, or removed by operation in animals, it is found that in a short time degeneration has taken place in the fibres of the corona radiata which pass off from this area. The degeneration can be followed down through the genu and the anterior two-thirds of the posterior limb of the internal capsule (Fig- 353) and the crusta of the cerebral peduncle of the corre- sponding side into the medulla oblongata. Below the decussation of the pyramids it is found that the degene- ration has involved the two pyramidal tracts, V ™- ^V/ )&\a£f and only these — the INTERNAL crossed pyramidal tract on the side oppo- site the cortical lesion, the direct pyramidal tract on the same side — and that the cross- section of the two de- generated tracts goes on continually dimin- ishing as we pass down the cord. (We overlook for the moment, in the interest of simplicity of statement, the fact that some degenerated fibres are found in the crossed pyramidal tract on the same side as the lesion.) This is proof positive that the cell-bodies of the neurons whose axons run in these tracts are situated in the cerebral cortex. They have indeed been identified with certain of the large pyramidal cells (the so-called giant cells or cells of Betz) in the cortex of the ' motor ' region in front of the Rolandic fissure (p 950). For after division of the motor pyramidal fibres in the upper cervical region of the cord (in monkeys) changes in the chromatin (so-called chromatolysis) and atrophy of these large cells occur. The same has been found to be true in man in cases where the cord was injured by fracture of the spine in such a way as to interrupt the tract (as well as other tracts) MID. BRAIN Fig. 353. — Pyramidal Path (after Cowers). Degenera- tion after destruction of the ' motor ' area of the right cerebral hemisphere. The degenerated areas are indicated by the shading. 876 THE CENTRAL NERVOUS SYSTEM completely and permanently, without entailing death for a consider- able time (Holmes and May). The fact that after destruction of the cortex or the path in its course the degeneration below the lesion does not spread to the anterior roots shows that at least one relay of nerve-cells intervenes between the pyramidal fibres and the root- fibres. The results both of normal and morbid histology enable us to identify the cells of the anterior horn as the cells of origin of the axons of the anterior root-fibres. For (i) Axis-cylinder processes have been actually observed passing out from certain of the so-called, motor cells of the anterior horn to become the axis-cylinders of the anterior root. (2). In the pathological condition known as anterior poliomyelitis, the cells of the anterior horn degenerate, and so do the anterior roots of the affected region, the motor fibres of the spinal nerves, and the muscles supplied by them. (3) As already mentioned (p. 858), comparatively transient but decided changes occur in the anterior horn cells on section of the corre- sponding anterior roots. (4) An enumeration* has been made in a small animal (frog) of the cells of the anterior horn and of the anterior root-fibres, and it has been found that the numbers agree in a remarkable manner. From all this it cannot be doubted that most, at any rate, of the cells of the anterior horn are connected with fibres of the anterior root. But since the number of fibres in the pyramidal tracts (about 80,006 in each half of the human cord) falls far short of the number of fibres in the anterior roots (not less than 200,000 in man on each side), it is necessary to suppose either that one pyramidal fibre may be connected with several cells or that all the anterior root-fibres are not in functional connection with the pyramidal tract. There is no reason to assume any such connection in the case of the fine medullated root-fibres arising in the lateral liorn and going to the visceral and vascular muscles. While there is no doubt that anterior root-fibres and pyramidal fibres of the brain and cord form segments of the same nervous path, the connection between the pyramidal fibres and the cells of the anterior horn has not yet been anatomically demonstrated. Many of the pyramidal fibres pass into the grey matter between the anterior and posterior horns or near the base of the posterior horn. The anterior horn cells are surrounded by arborizations. Some of these are probably the terminations of axons whose cell-bodies are situated in the posterior horn, others the terminations of posterior root-fibres or their collaterals. Many of them very likely represent the end arborizations of pyramidal fibres or their collaterals. Some ob- servers, however, suppose that the pyramidal fibres do not come into immediate relation with the anterior horn cells, but that another neuron is intercalated between them and the cells. The pyramidal fibres are unquestionably paths for voluntary motor impulses passing down from the cortex to the cord. But * Such enumerations can be made with great accuracy from photographs of sections of the nerves (Hardesty, Dale). (See Fig. 341, p. 860.) CONNECTIONS OF THE LONG PATHS OF THE CORD 877 they are not the only cortico-spinal efferent paths, and in many animals they are not even the most important paths for voluntary movements. It is the more skilled and delicate movements which the pyramidal tract subserves in man, and it is' these movements which are permanently lost when the tract is destroyed. The size of the path is proportioned to the degree of development of the brain. Thus, it is larger in the monkey than in the dog, larger in the anthropoid apes than in the lower monkeys, and larger in man than Fig. 354. — Paths from Cortex in Corona Radiata (Starr). A, tract from frontal con volutions to nuclei of pons and so to cerebellum; B, motor pyramidal tract; C, afferent tract for tactile sensations (represented in the diagram as separated from B by an interval for the sake of clearness); D, visual tract; E, auditory tract; F, G, H, superior, middle, and inferior cerebellar peduncles; J, fibres from the auditory nucleus to the posterior corpus quadrigeminum ; K, decussa- tion of the pyramids in the bulb; FV, fourth ventricle. The roman numerals indicate the cranial nerves. in even the highest of the apes. In the lower mammals it is exceed- ingly small. While in man the pyramidal tracts constitute nearly 12 per cent, of the total cross-section of the cord, they make up little more than i per cent, in the mouse, 3 per cent, in the guinea-pig, 5 per cent, in the rabbit, and nearly 8 per cent, in the cat. In some mammals, as the rat, mouse, guinea-pig, and squirrel, the pyramidal tracts lie, not in the antero-lateral, but in the posterior columns. In vertebrates below the mammals the pyramidal system does not exist as a collection of neurons which send their axons with- 878 THE CENTRAL NERVOUS SYSTEM out interruption down from the cortex to the cord. In birds, e.g., after the removal of a hemisphere, the degeneration does not extend below the mid-brain (Boyce). SECTION VI. — PATHS FROM AND TO THE CORTEX. Thus far we have been able to map out two great paths from the cerebral cortex to the periphery — one efferent, the other afferent, (i) The great efferent or motor pyramidal path, which, starting in the cortex in front of the fissure of Rolando, where its axons give off numerous collaterals to the grey matter soon after emerging from the cells, and sweeping down the broad fan of the corona radiata, passes through the narrow isthmus of the internal cap- sule into the crusta of the crus cerebri, and thence into the pons (Figs. 354, 355). At this level, the fibres destined to make connection with the motor nuclei of the cranial nerves in the grey matter underlying the aqueduct of Sylvius and the fourth ven- tricle terminate. Most of these fibres decussate to make physiological connection with nuclei on the opposite side, but some join nuclei on the same side. The question whether they arborize di- rectly around the cells of the motor nuclei or make junc- Fig. 355. — Motor Pyramidal Tracts (Diagram- matic) (Halliburton, after Gowers). The convolutions are supposed to be cut in vertical transverse section, the internal capsule, I, C, and the crus in horizontal section. O, TH, optic thalamus; CN, cau- date nucleus; L2 and Lj, middle and ex- ternal portions of lenticular nucleus; /, a, I, fibres from the face, arm, and leg areas of the cortex respectively; E, S, Sylvian fis- sure. The genu or knee of the internal capsule is indicated by the asterisk. tion with them through another intercalated neuron is precisely in the same position as the corresponding question for the spinal pyra- midal path (p. 876). On their way through the pons they send off collaterals to the nuclei pontis, as they do higher up to the grey matter of the basal ganglia of the cerebrum and the substantia nigra, and the path may be continued to the motor nuclei by axons arising here. There is no proof, however, that this is the case. The rest of the pyramidal fibres run on into the pyramid of the bulb, where the greater part (usually about 90 per cent.) of the fibres decussate, appearing in the cervical cord as the massive crossed PATHS. FROM AND TO THE CORTEX 879 pyramidal tract of the opposite side. A few (usually about 10 per cent.) remain on the same side as the slender direct pyramidal tract. The size of this tract varies much in different individuals, and it is occasionally absent. Its breadth constantly diminishes as it proceeds down the cord, and it disappears before the middle of the thoracic region is reached, its fibres continually decussating across the anterior white commissure and plunging into the opposite anterior horn. They either end among its cells, or, passing through it, reinforce the crossed pyramidal tract. The fibresof thiscrossedtract A _ '-F IS tf^E are, in their turn, con- tinually passing off into the grey matter to make connection (p. 876) with the cells of the anterior horn, whose axis-cylin- der processes enter the anterior roots of the spinal nerves. The losses which it suffers as it descends the cord may be in some slight degree compensated by the bi- furcation of some of its fibres (geminal fibres), but ultimately the whole tract forms synapses with cells in the grey matter, and dwindles awav as the lumbar re- FiS- 356.— Horizontal Section through the Right J. , , /T^. Hemisphere to show the Constituents of the glonisreached(Flg-343). internal Capsule (von Monakow). A, knee of A certain number of the corpus callosum ; B, anterior, B', posterior, horn of Calc. T' pyramidal fibres do not decussate either in the bulb or in the cord. These are called homo- lateral fibres. They run down in the lateral py- ramidal tract, and are represented by the fibres that degenerate in that tract after a lesion in the lateral ventricle; C, knee of internal capsule; S, sensory fibres; V, visual tract; AH, Amraon's horn; Calc, calcarine fissure; T, first, T', second, temporal convolution; OR, optic radiation; Aud, auditory tract ; D, retrolenticular region of internal capsule; lo, lenticulo-optic division of internal capsule; Cl, claustrum; op., operculum; I, island of Reil ; E, external capsule ; Is, lenticulo-striate division of internal capsule ; F, fibres from frontal lobe; F', inferior part of third frontal convolution; Th, optic thalamus; Put, putamen. motor ' area of the same side (p. 875). This would explain the escape in hemiplegia (paralysis of one side 88o THE CENTRAL NERVOUS SYSTEM Thalamo-Corfital FibrfS /sense of position \ \ • •• movement/ 5(>ino-Cerebe!Ur Tracts /Co- ordination &•] \ Muscular Tone I .Crossed Sensory Fibres /PW Hci,t& Cold\ llouch & Pressure/ ioAl Nerve -'^ Fig. 357. — Ascending Nerve Tracts (after Holmes). PATHS FkOM AND TO THE CORTEX 881 of the body) of those muscles which are accustomed to work with the corresponding muscles on the opposite side — e.g., the respiratory muscles, these being innervated to some extent from both cerebral hemispheres. (2) A great afferent or sensory path by which some at least of the impulses carried up through the posterior roots of the spinal nerves, after passing through various relays of nerve-cells, reach the cortex of the cerebellum ; or the upper portions of the central grey tube, the corpora quadrigemina and optic thalamus; or, finally (through the tegmentum and the posterior limb of the internal capsule behind the motor fibres), the cerebral cortex itself. The efferent pyramidal path from the cortex to the periphery is broken by at most two relays of nerve-cells — those intercalated cells to which reference has already been made (p. 876), if they really exist, and the motor cells of the anterior horn. The afferent path to the cerebral cortex is interrupted by at least three relays with axons of considerable length. One of the cells is situated in the ganglion on the posterior root, another in the medulla oblongata, a third in the optic thalamus ; and on some of the routes another, or even more than one, is intercalated between the medulla and the cortex (Fig. 357). The Internal Capsule. — We have already recognized the pyramidal tract and the afferent tegmental path as constituents of the internal capsule. The cranial fibres of the pyramidal tract occupy mainly the genu or knee, the spinal fibres the posterior limb as far back as the posterior border of the lenticular nucleus (Fig. 356). The fibres from the various motor areas are to a certain extent arranged in order in the capsule, those for the eyes and head lying farthest forward, those for the leg farthest back, while the fibres going to the face, arm and trunk occupy intermediate positions. The separation, however, is far from complete, the fibres of neigh- bouring regions being considerably intermixed (Hoche). As the tracts pass downwards the intermingling becomes continually greater (Simpson and Jolly) (Figs. 358, 359)- The afferent fibres from the thalamus to the cortex, which we have described as the last segment of the afferent tegmental path, lie in the posterior part of the posterior limb. But here again there is no absolutely sharp line of demarcation. Some motor fibres are intermingled with the sensory in the posterior part of the capsule, for lesions of this region produce a certain degree of paralysis as well as anaesthesia on the opposite side of the body. A pure capsular hemianaesthesia — that is, a loss of sensation on the opposite side due to a lesion in the internal capsule and unaccompanied by motor defect — does not appear to exist. Accordingly the common statement that the efferent (motor) path occupies the anterior two-thirds, and the afferent (sensory) path the posterior third of the posterior limb of the internal capsule, while true in a general sense, is not strictly correct, 56 882 THE CENTRAL NERVOUS SYSTEM The destination of the afferent fibres of the internal capsule has not been definitely settled. There is no doubt that they pass up to the convolutions around the fissure of Rolando (central convolu- tions), and there is reason to believe that some of them terminate in the ' motor ' region in front of that fissure, although many of the fibres concerned in tactile sensations seem to end in the ascending parietal convolution. But we have not yet exhausted the constituents of the internal capsule. Two great cones of fibres sweep down into it, one from the frontal, the other from the occipital and temporal portions of the cerebral cortex. The first passes through its anterior limb, the second behind the sensory path in its posterior limb. The cells of origin of the frontal fibres are known, and those of the occipital and temporal fibres are supposed, to be situated in the cortex. They are therefore efferent fibres as regards the cortex (cortifugal). Running on through the crusta of the cerebral peduncle (Fig. 352), the frontal tract Fig. 358.— Pyramidal Tract in In- ternal Capsule (Simpson and Jolly). Horizontal section through right cerebral hemi- sphere, cutting fibres of internal capsule transversely at an upper level a little below the upper surface of the lenticular nucleus. The extent of the degeneration following destruction of the whole of the right ' motor ' cor- tex, except the ' head and eyes ' area (in one of the lower mon- keys), is shown. Note over- lapping of fibres from face, arm, and leg areas, as shown by ex- periments in which one or other of these areas was alone re- moved. Fig- 359- — Pyramidal Tract in Internal Capsule at Lower Level (Simpson and Jolly). CN, head of caudate nucleus; OT, optic thalamus; Cl, claustrum. internal, the occipito-temporal external, they end in the grey matter of the pons, and serve as one segment of an extensive connection between the cerebral and the cerebellar cortex of the opposite side, the other segment being formed by neurons whose cell-bodies are situated in the pons, and whose axons, crossing PATHS FROM AND TO THE CORTEX 883 the middle line, pursue their course through the middle cerebellar peduncle, to terminate in the superficial grey matter of the cere- bellum. It is evident that the junction of the cerebral cortex with this pontine grey matter, through and into which so many nerve- tracts pass, multiplies the number of possible routes by which impulses may travel between one part of the brain and another. The corpus callosum forms a mighty link between the two cerebral hemispheres. And intertwined in the corona radiata with the callosal fibres are other systems, of which it is especially necessary to mention the afferent (cortipetal) fibres that join the optic thalamus with nearly every part of the cerebral cortex. Such fibres pass from the cells of the grey matter of the thalamus to the frontal and parietal Fig. 360. — Association Fibres (after Starr). Cerebral hemisphere seen from the side A, A, association fibres between adjacent convolutions; B, between frontal and occipital lobes; C, cingulum, connecting frontal and temporo-sphenoidal lobes; D, uncinate fasciculus between frontal and temporal regions; E, inferior longi- tudinal bundle between occipital and temporo-sphenoidal lobes; O.T., optic thalamus; C.N., caudate nucleus. regions through the anterior border of the internal capsule in front of the frontal fibres previously described as running in the anterior limb of the capsule to the pons; and from the thalamus to the occi- pital region through the extreme posterior border of the internal capsule, behind the occipital fibres that proceed to the pons. The fibres that connect the thalamus with the occipital cortex are spoken of as the optic radiation. Some of the fibres of the optic radiation, however, proceed, not from the thalamus, but from the anterior corpus quadrigeminum and the lateral geniculate body. The thalamus is also connected with the cortex of the temporal lobe, with the cerebellum, and through the fillet with the posterior part of the tegmental system, the medulla cblongata and the spinal cord (p. 874). Fibres also pass from the inner and deeper part of the thalamus to the lenticular nucleus of the corpus striatum. The 884 THE CENTRAL NERVOUS SYSTEM thalamus mast be regarded as a great sensory centre through which afferent impulses stream on their way to all parts of the cortex. The fibres which connect the cerebral cortex with lower levels of the central nervous system are sometimes grouped together as projection fibres in contradistinction to the commissural and associa- tion fibres. A large proportion of the fibres of the corona radiata are projection fibres, including efferent groups (pyramidal tract, fronto- pontine fibres, temporo- pontine fibres) and afferent groups (the fillet system, thalamo-cortical fibres from the grey matter of the thalamus to the cortex, in- cluding the optic radiation). SECTION VII. — CONNECTIONS OF BRAIN STEM WITH CORD — CONNECTIONS OF CERE- BELLUM. Connections of the Vestibulo- Spinal or Antero-Lateral De- scending Tract. — The main origin of these fibres is the nucleus of Deiters, a collection of large multipolar nerve-cells in the floor of the fourth ven- tricle near the inner auditory nucleus. This nucleus consti- tutes an important intermedi- ate station between the cere- bellum and the cord. Its cells give off axons which pass into the posterior longitudinal bundle of the bulb and pons, mostly to the bundle of the same side, but partly into that of the opposite side. Here the fibres bifurcate into an ascend- ing branch, which passes up to the oculo-motor nucleus, and a descending (vcstibulo - spinal) branch, which passes down- wards to the spinal cord and enters the antero-lateral de- Fig. 361. — Fibres connecting Frontal and Temporo-Occipital Lobes with Cerebellum, etc. (Diagram) (after Gowers). Fr, frontal, Oc, occipital lobe ; interrupted lines indicate fibres (TOC) connecting cerebellum and temporo-occipital cortex, and fronto-cere- bellar fibres (FC). On left side the position of these two groups of fibres and of motor (pyramidal) tract, PY, in the crus, is indi- cated by letters. The pyramidal tract is seen on the right passing down from the Rolandic area through posterior limb of internal capsule 1C (the genu or knee of which is indicated by asterisk) to decussate in the bulb. AF, ascending frontal convolu- tion; AP, ascending parietal convolution; FR, fissure of Rolando; IPF, intraparietal fissure; PCF, precentral fissure; Ipt, crossed pyramidal, apt, direct pyramidal tract. scending tract. The fibres of Ihis tract ultimately pass into the anterior horn, where most of them end by arborizing amongst the cells of the horn. Higher up corre- sponding fibres from the posterior longitudinal bundle arborize in the cranial motor nuclei. Connections of the Rubro-Spinal Tract. — These fibres, as the name given to the bundle implies, originate in the red nucleus and run down into the cord. A little distance from their cells of origin they cross the CONNECTIONS OF CEREBELLUM 885 median line, and then pass down first in the tegmentum, and below that in the lateral column of the bulb and cord. On their way they come into relation with the motor nuclei of the cranial nerves, and in the cord with the cells of the anterior horn. It is obvious that in contrast to the projection fibres the fibres of the rubro-spinal, vestibulo-spinal, olivo-spinal, and thalamico -spinal tracts (p. 867) and the posterior longitudinal bundle connect the brain stem or cerebral axis with the cord, or different levels of the brain stem with each other. Connections of the Grey Matter of the Cerebellum with the Periphery and other Parts of the Central Nervous System. — Numerous as are the nervous ties of the cerebral cortex, those of the grey matter of the cerebellum are, in proportion to its mass, still more extensive, particu- larly as regards afferent fibres, and perhaps not less important. Speaking broadly, we may say that the restiform body or inferior peduncle connects chiefly the dentate nucleus and the grey matter of the worm with the spinal cord and medulla oblongata, and through them with the periphery. The fibres which it receives from the direct Fig. 362. — Direct Sensory Cerebellar Path of Edinger. D, Deiters' nucleus; v, median nucleus of auditory nerve; t, nucleus of the roof; g, nucleus globosus. . e Fig- 363- — Diagram of Dorsal and Ventral Spino Cerebellar Tracts entering Cerebellum (Mott). P.C.Q., posterior corpora quadrigemina; s.v., superior vermis of cerebellum; d.a.c., v.a.c., dorsal and ventral ascending Cerebellar tracts. cerebellar tract (dorsal spino-cerebellar tract) of its own side it carries to the worm. These fibres occupy the outer portion of the peduncle. The fibres which reach the restiform body from the olivary nucleus of the opposite, and ako in smaller numbers from that of the same side, run mainly to the hemisphere. All these fibres are afferent in relation to the cerebellum (cerebello-petal) . An uncrossed afferent connection also exists between the cerebellum and the vestibular branch of the auditory nerve, through certain of its nuclei of reception, and also between it and the nuclei of other cranial nerves, such at> the trigeminus and the vagus. The fibres pass up in the inner portion of the inferior peduncle (direct sensory cerebellar path of Edinger, Fig. 362) to the nucleus of the roof (nucleus tecti) and nucleus globosus. Some efferent fibres (cerebellofugal) also run down from the cerebellum in the inferior peduncle, including fibres from the nucleus tecti of the opposite side which are on their way to the medulla oblongata. 886 THE CENTRAL NERVOUS SYSTEM The middle peduncle is in the main a link between the cerebellar cortex and the cerebral cortex of the opposite side, through the relay of the pontine grey matter. Most of the fibres in it are afferent in rela- tion to the cerebellum, their cells of origin being situated in the nuclei of the pons, and sending their axons across the middle line to end in the cerebellar cortex. The superior peduncle connects chiefly the dentate nucleus of one side with the cortex of the opposite cerebral hemisphere through the red nucleus of the tegmentum of the crus cerebri and the optic thalamus on the opposite side. The great majority, or perhaps all, of its fibres are efferent fibres as regards the cerebellum — i.e., their cells of origin lie in the dentate nucleus. Running upwards and forwards in the superior peduncle towards the mid- brain they cross the middle line below the corpora quadrigemina, and then bifurcate into ascending and descending branches. The ascending branches end mainly in connection with cells in the red nucleus, but some of them pass on to the optic thalamus, with which cells of the red nucleus are also connected. The thalamus, as we have seen, is in its turn exten- sively connected with the cerebral cortex, and the red nucleus (by the efferent tract of Monakow) with the grey matter of the cord. Fig- 364. — Paths of Middle Cerebellar Peduncle (Mingazzini). The scheme indicates the afferent and efferent paths which run through the middle cerebellar peduncle, connecting cerebellum with opposite side of cerebrum, a, fibre coming from a cell in the nuclei pontis and going to the cere- bellar cortex; b, fibre from a cell in cortex of The descending branches of the fibres of the superior peduncle, entering the re- tic ular formation of the pons, pass down, it is said, to make connection with the motor nuclei of the cranial and spinal nerves. The tract of Gowers, as previously stated, comes into relation with the su- opposite cerebral hemisphere making connection in the pons with a (a and b together constitute an afferent path to the cerebellum); c, a fibre springing from a Purkinje's cell in the cerebellar cortex and making connection in the pons with a cell d, which sends its axon to the cerebral cortex of the opposite side, c and d constitute an efferent path from cerebellum to opposite cerebral hemisphere; e, f, represent a path , coming from the cerebellar cortex, which crosses the middle line in the pons, and then ascends till it loses itself in the formatio reticularis. perior peduncle, passing backwards along its mesial border to the worm. Since the cortex of the cerebellum is linked to the dentate nucleus, the superior peduncle affords an indirect connection between it and the cerebral cortex. Through the restiform body afferent impulses pass up to the cerebellum. From the cerebellum they may proceed to the cerebrum. So that the path by the restiform body, dentate nucleus, and superior peduncle may form an alternative route for afferent impressions ascending from the periphery to the great brain — a path broken by at least four relays of nerve-cells. The cerebellar hemisphere may be connected by an efferent path through the nucleus of Deiters and the descending antero-lateral tract with the .motor roots of the same side. Another efferent path (from the dentate nucleus) may be constituted by the fibres of the superior peduncle and Monakow' s bundle. FUNCTIONS OF CENTRAL NERVOUS SYSTEiM 887 We have purposely omitted to enumerate other paths by which the various tracts of grey matter in the brain are brought into com- munication with each other, and our knowledge of such connections is no doubt far from complete. When we add that not only are the cerebral hemispheres united by many ties to the subordinate portions of the cerebro-spinal axis and to each other, but that cortical areas of one and the same hemisphere are in communication by short connecting loops of ' association ' fibres (Fig. 360), it will be seen that the linkage of the various parts of the central nervous system is extremely complex; that an excitation, blocked out from one path, may have the choice of many alternative routes; and that the ap- parent simplicity and isolation of the pyramidal tracts must not be allowed too far to govern our views of the possibilities open to a nervous impulse once it has been set going in the labyrinth of the nervous network. Nor is it only by the main channel of the axis- cylinder that nervous impulses can be conducted, they can also pass along the collaterals. And the actual route taken by a given impulse is determined not only by anatomical relations, but also by molec- ular conditions, particularly in the terminal fibrils of the axons, collaterals and dendrites, and in the substance, if such a substance there be, which intervenes between the end arborizations of a neuron and the dendrites or cell-bodies of the neurons with which they lie in contact. So that a road open at one moment may be closed at another. (See p. 902.) We may suppose that the greater the number of connections between the cells of the central nervous system, the greater is the complexity of the processes which may be carried on within it. And, indeed, comparison of the brains of different animals shows that it is not so much by excess in the number of nerve-cells as by the increased complexity of linkage, that a highly-developed brain differs from a brain of lower type; the higher the brain, the more richly branched are the dendrites and the terminations of the axons and their collaterals, and, therefore, the greater is the numbei of possible paths between one nerve-cell and another. SECTION VIII. — FUNCTIONS OF THE CENTRAL NERVOUS SYSTEM — (i) THE SPINAL CORD (INCLUDING THE MEDULLA OBLONGATA). Much of our knowledge of the functions of the central nervous system and of its divisions has been gained by the removal or destruction of more or less extensive tracts of nervous substance, or the cutting off of connection between one part and another. But it is well to warn the reader at the very outset that in no other part of physiology is such caution required in making deductions as to the working of the intact mechanism from the phenomena which mani- fest themselves after such lesions. 888 THE CENTRAL NERVOUS SYSTEM In the first place, every operation of any magnitude on the brain or cord is immediately followed by a depression of the functional power of the nervous tissue distal to the lesion, a depression which may extend far from the actual seat of injury and manifest itself by various phe- nomena, which are grouped together under the name of ' shock,' better termed spinal or cerebro-spinal shock, to distinguish it from the cardio- vascular or surgical shock already mentioned (p. 192). Thus, when the spinal cord of a dog is divided, e.g., in the dorsal region, all power, all vitality, one might almost say, seems to be for ever gone from the portion of the body below the level of the section. The legs hang limp and useless. Pinching or tickling them calls forth no reflex movements. The vaso-motor tone is destroyed, and the vessels gorged with blood. The urine accumulates, overfills the paralyzed bladder, and continually dribbles away from it. The sphincter of the anus has lost its tone, and the faeces escape involuntarily. And if we were to continue our observa- tions only for a short time, a few hours or days, we should be apt to appraise at a very low value the functions of that part of the cord which still remains in connection with the paralyzed extremities. But these symptoms are essentially temporary. They are the immediate results of the section; they are not permanent ' deficiency ' phenomena. And if we wait for a time, we shall find that this torpor of the lower dorsal and lumbar cord is far from giving a true picture of its poten- tialities; that, cut off as it is from the influence of the brain, it is still endowed with marvellous powers. If we wait long enough, we shall see that, although voluntary motion never returns, reflex movements of the hind-limbs, complex and co-ordinated to a high degree, are readily induced. A few months after transection of the cord it is easy to show that, although the dog cannot use the hind limbs for continuous pro- gression, the machinery for executing the appropriate movements exists in the cord. When, for example, the animal is held in the proper position and given a slight push forwards, it may take two or three steps before its legs give way. The regulation of the movements necessary for the maintenance of equilibrium cannot be achieved when the control of the higher centres is eliminated. In water, where the problem of the maintenance of the normal position is solved by the mechanical properties of the medium, the dog can use the hind-legs in swimming. The tone of the resting muscles below the lesion is even somewhat greater than normal. Vaso-motor tone also comes back. The functions of defaecation and micturition are normally performed. Erection of the penis and ejaculation of the semen take place in a dog. A man with complete paralysis below the loins and destitute of all sensation in the paralyzed region has been known to become a father (Brachet). Pregnancy carried on to labour at full term has been observed in a bitch whose cord was completely divided above the lumbar enlargement. The severity and duration of spinal shock are greater in the monkey than in the dog, in man than in the monkey, and in the whole mammalian group than in the lower vertebrates. The mechanism of its production has been much discussed, and will be referred to on another page (p. 912). We cannot doubt that. the spinal cord takes an important share in the recovery of function after shock. But here again it would be erroneous to conclude that everything is due to the cord. For Goltz and Ewald have been able to keep dogs alive for long periods after preliminary section of the cord in the cervical region and subsequent removal of large portions of it. They find tha.t even after destruction of the lumbar and sacral regions of the cord the external sphincter of the anus, striped and even voluntary muscle though it be, regains its FUNCTIONS OF THE SPINAL CORD 889 tone, although it is temporarily lost after the first cervical section. The bladder ultimately recovers the power of emptying itself spon- taneously and at regular intervals. A pregnant bitch in which the lumbar enlargement and the whole cord below it to the cauda equina had been removed, and therefore all the nerve-roots supplying fibres to the uterus cut, whelped in a normal manner, and the corresponding mammary glands behaved exactly as the rest. Digestion went on as usual when practically nothing of the cord except its cervical portion was left. Certain vaso-motor phenomena were also observed which suggest that the sympathetic system, independently of the cerebro- spinal system, is itself possessed of regulative powers (p. 183). Secondly, we must not run into the opposite error, and assume, without proof, that all the functions which the brain or cord is capable of manifesting under abnormal circumstances are actually exercised by either when, under ordinary conditions, it is working along with and guiding, or being guided by, the other. For example, in many animals certain of the reflex powers of the cord are, if not increased, at all events more freely exercised when the controlling influence of the higher centres has been cut off than when the central nervous system is intact. Thirdly, there is another class of phenomena which we must make allowance for, and perhaps more frequently in the case of pathological lesions in man than in experimental lesions in the lower animals. This is the class of ' irritative ' phenomena. The irritation set up by a blood- clot or a collection of pus, or in any other way, in a wound of the grey or white matter, may cause a stimulation of nervous tracts by which, for a time, the ' deficiency ' effects of the lesion may be masked. In the fourth place, we must not hastily conclude that, when no obvious deficiency seems to follow the removal of a portion of the central nervous system, the function of that portion must necessarily be of such a nature as to give rise to no objective signs. For there is reason to believe that, to a certain extent, the function of one part may, in its absence, be vicariously performed by another. Bearing in mind the cautions we have just been emphasizing, we may broadly distinguish between the functions of the cord (including the bulb) and those of the brain proper by saying that the cord is essentially the seat of reflex actions, the brain the seat of automatic actions and conscious processes. But neither of these conceptions is entirely correct. Both err by defect and by excess. The brain, it is true, is pre-eminently automatic. The movements which are started in the grey matter of the cerebral cortex are pre-eminently voluntary (p. 946), but we cannot deny to the brain the possession of reflex powers as well. The movements in which the only nerve centres concerned are those of the spinal cord are above all reflex (p. 897)- But some of its centres, and especially those lying in the medulla oblongata — e.g., the respiratory centre — are, much as they are influenced by afferent impulses, capable of discharging auto- matic impulses too. And while consciousness is certainly bound up with integrity of the brain, and in all the higher mammals is asso- ciated with cerebral activity alone, it has been plausibly maintained that the spinal cord, even of such an animal as the frog, may also be endowed with something which might be called a kind of hushed consciousness. Whether this be so or not for the frog, with its THE CENTRAL NERVOUS SYSTEM distinct though relatively ill-developed cerebral hemispheres, it would seem by no means unlikely in the case of fishes and animals below them, which are practically devoid of a cerebral cortex altogether. The functions of the spinal cord may be classified thus: 1. The conduction of impulses set up elsewhere — either in the brain or at the periphery. 2. The modification of impulses set up elsewhere (reflex action). 3. The origination of impulses (?). Cerebral Fi.bre.for Jfead J fiecussation cf Pyramids ofDi.rect. Pyramidal tract cfAnt%m !5T -TniTxjl Arborisation of a Pyramids I Fibre, around, cell of ' ' ffow Anterior Root Fibres Fig- 365. — Some Possible Paths of Efferent Impulses in the Central Nervous System (Schematic). Details are omitted from the scheme. For instance, each pyra- midal fibre is represented as arborizing around one anterior cornual cell only, and no collaterals are shown. The hypothetical intercalated neurons between the pyramidal fibres and the anterior horn cells (p. 876) are not shown. I. Conduction of Nervous Impulses by the Cord. — The old con- troversy as to whether the white fibres of the spinal cord are directly excitable has long been settled in the affirmative. The inquiry was complicated by the presence of the spinal roots, which, since the experiments of Charles Bell, have been known to be capable of excitation by artificial stimuli. But at length the difficulty was overcome in this way: The posterior (dorsal) portion of several FUNCTIONS OF THE SPINAL CORD 891 segments of the cord with the attached posterior roots and the grey matter was excised. Long strands of the white matter of the anterior (ventral) portion of the cord were isolated, and laid on electrodes, and contractions of muscles were seen to follow stimulation, even when the anterior roots nearest the stimulating electrodes had been cut, and every precaution taken to avoid escape of current on to the distant anterior roots of the nerves supplying the muscles. Indeed, apart from direct experimental evidence, the fact that the white fibres of the brain are universally admitted to be excitable by artificial means would be of itself almost sufficient to decide the question, for we know of no essential difference between the cerebral and the spinal fibres. But the con- ditions must rarely occur under which direct stimulation of white fibres in their course is possible in the intact body ; and the only impulses with which we need concern ourselves here are those that reach the con- ducting paths from grey matter in the cord itself or in the brain, or from the peripheral organs. What sort of impulses do the various tracts of the spinal cord conduct ? For the dorsal or posterior roots this question was first fully answered by Magendie; for the venlial or anterior roots, although with a certain degree of ambiguity, by Sir Charles Bell. Bell observed that when, in an animal just killed, he mechanically stimulated the anterior roots, muscular contractions were obtained at each touch of the forceps. He concluded that the anterior roots are motor^and sensory, while the posterior roots are ' vegetative ' — i.e., connected ^wrfh" the functions of the viscera, the so-called ' vegetative ' organs. But although he is often credited with the discovery of the functions of the posterior roots as well, he was not the first to make the decisive experiment necessary to show that they are the conductors of sensory impulses. It was after Magendie's discovery that only a portion of the nerves are sensitive, and that there are nerves ' which are like tendons, aponeuroses, or cartilages in insensibility/ that Bell formulated the law that the anterior roots are purely motor, the posj£»or purely sensory. This law, often termed Bell's Law, is more correctly denominated the Magendie- Bell Law. When the posterior roots are divided, loss of sensation occurs in the region to .which they are distributed. If only one root is cut, the loss of sensation is never complete in any part of the skin ; and Sherrington has found that the cutaneous areas of distribution of consecutive nerve-roots are not perfectly independent, but to some extent overlap. Stimulation of the peripheral end of the divided posterior root has no effect. Stimulation of the central end gives rise, if the animal be conscious, to evidences of pain, and other signs of the passage of afferent impulses — e.g., a rise in blood-pressure. The latter may also be observed when the animal is anaesthetized. Referred Pain. — The posterior roots contain sensory fibres not only for the skin, but also for the deeper structures and the viscera. The afferent fibres reach the viscera by the sympathetic, the vagus, and the pelvic nerves or nervi erigentes. Clinical observations have thrown 892 THE CENTRAL NERVOUS SYSTEM much light upon the distribution of the visceral fibres and their rela- tion to the cutaneous sensory nerves. It has long been known that in disease of an internal organ the pain is often referred to some super- ficial part. It has now been demonstrated that each organ is related to a more or lees definite region, or more than one region, of the skin. In disease of the organ there is in this area increased excitability (hyperalgesia) or tenderness to slight mechanical stimuli, and often also increased excitability for heat or cold, and the reflexes elicited by stimulation are exaggerated (Head, Dana). The bond of connection appears to be the origin from the same spinal segments of the autonomic* sensory fibres of. any viscus and the sensory supply of the corresponding cutaneous area. The common anatomical origin seems to carry with it a physiological correlation, either because the irritation of the visceral fibres spreads in the cord to the somatic afferent fibres which enter the corresponding segments, or because of some action higher up in the cerebral centres, the nature of which will be best considered along with the general topic of the localization of sensory impressions (Chapter XVIII.) . Recurrent Sensibility. — Although muscular contraction is the most conspicuous event that follows stimulation of the peripheral end of an anterior nerve-root, it is by no means the only one. It is frequently observed, though not' in all kinds of animals, that here, too, pain is caused. That this pain is not due to the muscular con- traction is proved by the fact that it can still be elicited when the nerve-trunk is divided between the junction of the roofs and the periphery. The real explanation of the phenomenon is that certain fibres from the posterior roots (' recurrent fibres/ see footnote on p. 797) bend up for some distance into the anterior roots, and then turn round again and pursue their course to their peripheral dis- tribution in the mixed nerve, or run on in the motor roots to supply the sheath surrounding them (nervi nervorum), and even the mem- branes of the spinal cord. The afferent impulses that enter the cord along the posterior roots have the choice of many paths by which they may reach the brain. The following are a few of the routes which they may follow : (1) They may pass directly up through the postero-median column. If they take this route, their course will be first interrupted by nerve- cells in the gracile or cuneate nuclei in the medulla oblongata. Thence they may find their way across the middle line by the arcuate fibres of the upper or sensory decussation, and, sweeping along the fillet and the sensory path in the hinder part of the posterior limb of the internal capsule, finally arrive at the cerebral cortex. Between the gracile and cuneate nuclei and the cortex they pass through nerve-cells in the optic thalamus. (2) They may pass up by the direct cerebellar tract and restiform body to the grey matter of the cerebellar worm. If they take this route their course will be interrupted very soon after their entrance into * Langley uses the term autonomic nervous system to include the nerve supply of heart muscle, all unstriated muscle, and all gland cells in, the body. It embraces, in addition to the sympathetic, cranial autonomic fibres in several of the cranial nerves and sacral autonomic fibres in the nervi erigentes (see Chapter XVII.). FUNCTIONS OF THE SPINAL CORD 893 the cord in the cells of Clarke's column. Since the superficial grey matter of the vermis is connected by association fibres with the dentate nucleus, and the dentate nucleus by the superior peduncle with the oppo- site cerebral hemisphere, this is also a possible path to the great brain. (3) They may reach the antero-lateral ascending tract of the same side through its cells of origin in the spinal grey matter, and, passing through the medulla and pons to the superior peduncle of the cere- bellum, enter the grey matter of the superior worm. (4) They may cross the middle line, after entering the cord, through axons or collaterals (p. 872) which run in the anterior and also in the posterior commissure, enter one of the ascending tracts on the other side — e.g., the tract of Gowers — and continue without further decus- sation up to their central destination. (5) They may spread from neuron to neuron in the tangle of the grey matter itself, and pass out again at a different level into one of the white tracts on the same or on the opposite side of the cord. Efferent impulses from the brain may travel — (1) Through the direct or crossed pyramidal tract. (2) From one side of the cerebral cortex to the other, and then down the pyramidal tracts corresponding to that side (?). (3) From the frontal part of the cerebral cortex, through the anterior limb of the internal capsule to the grey matter in the pons, and thence to the cerebellum by its middle peduncle. (4) From the occipital or temporal cortex, in the hinder rim of the internal capsule, to the pontine grey matter, and through the middle peduncle to the cerebellum. From the cerebellum they may possibly pass down to the nucleus of Deiters and thence along the antero-lateral descending tract to the anterior horn of the cord, and indirectly to the periphery. All the paths enumerated, as well as others to which it would be tedious to formally refer, and which the ingenuity of the reader may profitably be employed in constructing for himself, from the data already given, are to be looked upon as possible channels for the passage of impulses between the brain and the periphery. But it must be distinctly pointed out that what is certain is in this case much more limited than what is possible. Among the efferent paths it is certain that the pyramidal tracts are conductors of voluntary motor impulses, and that in most individuals the great majority of such impulses decussate in the medulla oblongata, only a small minority in the cord. For a lesion involving the pyramidal tract above the decussation of the pyramids causes paralysis of the oppo- site side of the body, a lesion below the decussation paralysis of the same side. It is certain that when one pyramidal tract has been destroyed, in many animals at least, the resulting paralysis is soon recovered from, at any rate to a great extent, and it is possible that in this case the motor cortex on the side of the lesion has placed itself again in communication with the paralyzed muscles through its commissural connections with the opposite hemisphere. This, however, is not the only alternative, for, as already pointed out, the pyramidal tracts are not the only cortico-spinal paths which can 894 THE CENTRAL NERVOUS SYSTEM subserve volitional movements, and division of the anterior portion of the antero-lateral column may cause deeper and more permanent paralysis than division of the pyramidal tract. In the dog total section of the pyramids is not followed by com- plete paralysis of voluntary movements, and stimulation of the cortical motor areas can still elicit characteristic movements. It is obvious that impulses emanating from the cortex can reach the motor nuclei of the cord by other routes than the long pyramidal fibres, possibly by paths with several segments, of which, for example, the rubro-spinal tract (p. 867) may be one. Just as an important business house may find it useful or indispensable to supplement or replace the common telegraph service by private wires in the interest of more prompt and satisfactory communication with its principal correspondents, while still utilizing the ordinary channels to some extent, so the higher brains may be supposed to have developed more and more the direct service of the pyramidal tract to tighten the grip of the cortex upon the motor nuclei of the cerebro- spinal axis, while still availing themselves, although in diminishing degree as their evolution proceeds, of the more primitive indirect paths. Decussation of the Sensory Paths. — On the other hand, it is certain that pathological or traumatic lesions, apparently involving the destruction of one lateral half of the cord in man and experimental jemisections in some mammals, are followed by symptoms which suggest that some kinds of sensory impulses decussate chiefly in the spinal cord — viz., diminution or loss of sensibility to pain and to changes of temperature on the opposite side below the level of the lesion, with little or no impairment, and often increase of sensibility (hypersesthesia) on the same side. Tactile sensibility is lost on the side of the lesion, and likewise the muscular sense. The first general description of this symptom-complex was given by Brown-Sequard. On the basis of clinical observations in man, he came to the conclusion that unilateral lesions of the cord, equivalent approximately to a semisection, are associated with muscular paralysis below the level of the lesion on the same side, and loss of cutaneous sensibility on the opposite side, while on the side of the lesions there may be an augmentation of sensibility. He interpreted these facts as meaning that the sensory path decussates soon after its entrance into the cord. The sensory path from the left side is therefore spared by a lesion of the left side of the cord, but interrupted by a lesion of the right side of the cord. The left and right motor paths, having already decussated in the bulb, are cut by lesions in the left and right halves of the cord respectively. Long afterwards Brown-Sequard saw cause to retract this interpretation of the facts observed by him, but the majority of subsequent observers have considered his original hypothesis more satisfactory than his later ones. While it may be true that in man it has not been rigidly demonstrated that the symptoms are associated with a clean-out lesion precisely limited to one-half of the cord, clinical observation lias on the whole tended to confirm the view that an FUNCTIONS OF THE SPINAL CORD 895 important portion of the sensory path decussates in the cord. But it is a curious circumstance that experimental physiologists have for the most part obtained contradictory results. Thus Mott, working with monkeys, found that the different kinds of sensation, far from being abolished, are as a rule impaired in a smaller degree on the side opposite to the semisection than on the same side, while Ferrier and Turner obtained on the whole a contrary result, and one that corresponded closely with Brown-Sequard's original description. The discovery that no ascending degeneration, or only a trifling amount, is to be found on the opposite side of the cord, either after semisection or after division of posterior roots, does not of itself enable us to decide the question. For while this latter fact shows that few or none of the afferent fibres cross the middle line to enter the long conducting paths before being interrupted by nerve-cells, it by no means proves that afferent impulses do not decussate in the cord. The long paths of the posterior column, indeed, do not decussate below the level of the bulb. The dorsal and ventral spino-cerebellar tracts are also, in the main at least, uncrossed spinal paths. A portion of the afferent impulses must therefore be carried up to the cerebrum and the cerebellum without decussating in the cord. But nobody can tell how massive a link between the two halves of the cord may be formed by the grey matter and the endogenous fibres of the white columns and their collaterals. We know that some afferent impulses do decussate far below the level of the medulla. For, (i) A part of the action current (p. 838) crosses the middle line and ascends in the opposite half of the cord when the central end of one sciatic is stimulated (Gotch and Horsley). (2) Crossed reflex move- ments are possible ; and when excitation of the central end of the sciatic is followed by contraction of the muscles of the opposite fore-limb, the afferent impulses must either decussate in the lumbar cord, and then run up on the opposite side to the level of the brachial plexus, or must ascend on the same side and cross over somewhere between the plane of the sciatic and the brachial nerve-roots. The only other hypothesis on which crossed reflex action can be explained — but a hypothesis for which there is not a tittle of evidence — is that the afferent impulse always acts on the few motor cells whose axis-cylinder processes pass over to the opposite side, and there enter anterior nerve-roots. But while, for these reasons, it cannot be denied that some afferent impulses decussate in the cord, it would be an error to conclude that all do so in any animal, or that all animals are in this respect alike. It is indeed extremely probable that in different species of animals, and even in individuals of the same species, there are considerable differences in the extent of the sensory decussation in the cord, just as there are in the extent of the motor decussation in the bulb. In some animals the greater part of the sensory path may decussate in the cord ; in others the greater part may decussate in the bulb, or higher up. The lack of agreement in the experimental results may be due partly to this cause. When it is further remembered how difficult it sometimes is to interpret the account which a man gives of his sensations, and to recognize precisely the degree and nature of sensory defects produced by disease in the human subject, it will not be thought surprising that experi- ments on animals, from the time of Galen onwards, should have yielded evidence which, although perhaps now at length tending to a definite result, is still unfinished and in part conflicting. If, leaving them out of account, not as valueless but as still difficult of interpretation, we attempt to draw any general conclusion 896 THE CENTRAL NERVOUS SYSTEM from the clinical observations which, however imperfect, are in such questions our surest guide, it can only be this, that in man some of the sensory impulses, and particularly those connected with the cutaneous sensations of pain and temperature, decussate, in part at least, in the cord. But there is also evidence that tactile afferent im- pulses, including those coming from the muscles and related to the muscular sense, and, it may be, some of the impulses associated with pain, decussate, not in the cord, but in the bulb. The Paths for Different Kinds of Sensory Impressions. — If this is the state of our knowledge where the problem is merely to determine the crossing-place of afferent impulses which are certainly known to cross, it is only to be expected that we should be still more in the dark as regards the routes by which different kinds of afferent impulses thread their way through the maze of conducting paths in the neural axis to reach their planes of decussation and gain the ' sensory crossway ' in the internal capsule. Some authors have indeed cut the Gordian knot by assuming that any kind of sensory impression may travel up any afferent path. Direct stimulation of a naked nerve-trunk, it has been argued in favour of this view, gives rise to a sensation of pain; stimula- tion of the skin in which the end-organs of the nerve lie gives rise to a sensation of touch or a sensation of temperature, according as the stimulus is a mild mechanical or a thermal one, the contact of a feather or of a hot test-tube. Why, it has been asked, should we imagine that the difference in the result of stimulation depends on a difference in the nerve-fibres excited, and not on a difference in the kind of impulses set up in the same nerve-fibres ? This is a question which we shall have again to discuss. But apropos of our present problem, we may say that there is very clear proof from the pathological side that a limited lesion in the conducting paths of the central nervous system may be associated with defect or total loss of one kind of sensation, while all the other kinds remain intact. And there seems no other tenable hypothesis than that in such cases the pathological change has picked out a particular group of fibres, either collected into a single strand or scattered among unaltered fibres of different function. For example, in syringo-myelia, a condition in which cavities are formed in the grey matter of the cord secondary to a new growth of the neuroglia surround- ing the central canal, a frequent symptom is the loss in a certain region of sensibility to pain and to changes of temperature, while tactile sensibility is unaffected (dissociation of sensations). Again, in loco- motor ataxia, a disease in which inco-ordination of movement and derangement of the mechanism of equilibration are prominent symptoms, degeneration in the posterior column of the cord is a most constant lesion. And there is strong evidence that afferent impulses from muscles and tendons, which either give rise to impressions belonging to the group of tactile sensations, or produce no effect in consciousness, and which, according to the most widely accepted doctrine, serve as the basis of the muscular sense, and play an important part in the main- tenance of equilibrium (p. 941), pass up in the posterior column. It may also conduct tactile impressions from the skin. A case has been observed where a man received a stab which divided the whole of one side of the cord and the posterior column of the other side. Sensibility to touch was lost on both sides of the body below the level of the injury, sensibility to pain only on the side opposite to the main lesion. In another case, in which some small syphilitic tumours (gummata) in FUNCTIONS OF THE SPINAL CORb $97 the lateral column on the left side caused marked degeneration in the left direct cerebellar tract, the tract of Cowers, and the crossed pyram- idal tract, without affecting the posterior columns, tactile sensibility was only slightly impaired in the opposite leg, while the sensibility for pain and temperature was much enfeebled. In the left leg, which was paralyzed, there was slight hyperaesthesia. These observations indicate that impressions of pain and temperature pass up in the antero-lateral column, either in the tract of Cowers, or in the direct cerebellar tract, or in both (Dejerine and Thomas). But it does not follow that they cannot ascend by other paths as well. It appears, indeed, that the grey matter of the cord, or, rather, short endogenous fibres arranged in series in the antero-lateral column so as to connect the grey matter at different levels, constitute such a path, and that impulses which give rise to pain can be propagated along a cord in which hardly a vestige of white substance remains uncut. In man the path for pain and temperature impressions along these short endogenous fibres seems to be mainly or entirely a crossed path. The afferent paths for such vaso-motor reflexes as are elicited by stimulation of the central end of the sciatic ascend in the lateral column, and the impulses largely cross the middle line in the cord. The posterior columns have nothing to do with the conduction of painful impressions, for division of them causes not anaesthesia, but rather hyperaesthesia, while if they are left intact, and the rest of the cord, including the grey matter, divided, the animal is insensitive to pain below the level of the lesion. Just as man differs from lower animals in the completeness with which certain of the sensory impressions decussate in the cord, so differences exist in the degree of localization of the different kinds of impressions in particular tracts. One of the outstanding differences is that in animals it seems to be easier for a still intact path to be substi- tuted for a severed path as a conductor of impulses which normally traverse the latter. The rapidity with which sensation is restored below the lesion after semisection of the cord in animals is an illustration of this. Another difference, which can be explained in the same way, is that a sharply-marked dissociation of sensations — retention of tactile sensibility, for example, with loss of sensibility to pain or to pain and temperature changes — either cannot be produced experimentally in animals, or is very difficult to realize. The impulses which descend the cord give token of their arrival at the periphery by causing either contraction of voluntary muscles, or con- traction of the smooth muscular fibres of arteries, or secretion in glands. They all pass down in the antero-lateral column, but the path of the voluntary impulses in the pyramidal tracts is the best known and most sharply defined. 2. Modification of Impulses set up elsewhere (Reflex Action). — The spinal cord, although it is a conductor of nervous impulses originating elsewhere, is by no means a mere conductor. Many of the impulses which fall into the cord are interrupted in its grey matter. Some of the efferent impulses proceeding from the brain are perhaps modified in the cord, and then transmit ted to the muscles. Some of the afferent impulses are modified, and then transmitted to the brain ; some are modified, and deflected altogether into an efferent path. These last are the impulses which give rise to reflex effects. A reflex action has sometimes been defined as an action carried out in the absence of consciousness: not necessarily, however, in the 57 898 THE CENTRAL NERVOUS SYSTEM absence of general consciousness, but in the absence of consciousness of the particular act itself. But the term is now more correctly used so as to embrace all kinds of actions which are not directly voluntary, whether the individual is conscious of them or not. For example, when the sole of rhe foot is tickled, the leg is irresistibly and involuntarily drawn up by reflex contraction of its muscles; yet the person is perfectly cognizant both of the movement and of the sensation which accompanies the afferent impulse. Many reflex actions usually associated with sensations proceed normally when consciousness is entirely in abeyance; during sleep most of the ordinary reflexes can be elicited. Anatomical Basis of Reflex Action. — Since the essence of reflex action is that the arrival of afferent impulses in the spinal cord or brain causes the discharge of efferent impulses, there must be some connection between the incoming and the outgoing nerve-fibres. In unicellular animals, such as amoeba, there is no differentiation of any special nervous or conducting path. A stimulus applied at one point may cause contraction anywhere. Even in the lowest multicellular animals or metazoa — e.g., in the sponges — there is no special nervous tissue. In some species of hydra, however, many of the surface or ectodermic cells (p. 6) possess deeply-placed contractile or muscular processes, and stimulation of the surface cells is followed by contraction of these processes. We may imagine that the first beginnings of an actual nervous system may have arisen by a further differentiation of such an ectodermic cell into a receptive portion at the surface, a deeper con- tractile portion, and an intermediate strand of protoplasm connecting the two, and capable of conducting the excitation from surface to muscular process. In such a reflex arc the nervous link would consist only of the conducting strand analogous to the nerve fibre joining the receptive or sensory element to the contractile element, and the dis- tinction between afferent and efferent fibre would not exist. When development has gone a step further, and the neuro-muscular process is interrupted by a second epithelial cell transformed into a nerve-cell, the afferent fibre enters one pole and the efferent fibre leaves the other pole of the same cell. It is this condition which we actually find when the nervous system first emerges in the animal scale as an unmistakably differentiated structure — namely, in the Ccelenterates in such forms as the jellyfishes. Here the three types of cell, receptive or sensory cell, reactive or central cell, and motor or contractile cell, are connected together by conducting paths or nerve-fibres. In a simple reflex action three events can be distinguished : stimulation of a receiving mechanism, conduction of the excitation, and the consequent reaction or end-effect. The receiving mechanism or receptor may consist of ordinary sensory nerve-endings in the skin, or of special sense-endings, as in the retina or internal ear. The conducting mechanism or conductor in all except the very simplest nervous systems is made up of at least two neurons, one the afferent portion of the reflex arc connected with the receptor, the other the efferent portion of the arc, connected with the organ, sometimes termed the effector organ — a muscle, e.g., or a gland — which accomplishes the end-effect. The transference of the excitation from the afferent to the efferent neuron takes place across the intervening synapce. The simple isolated reflex arc, as thus described, although a convenient abstraction, corresponds but little to anything which actually exists in one of the higher animals. With increasing com- FUNCTIONS OF THE SPINAL CORD 899 plexity of organization the nervous impulse passing up an afferent fibre is in general offered, instead of a single efferent path, a choice of many potential routes when it reaches the spinal cord. We have previously (p. 872) described the course taken by the fibres of the posterior roots on entering the cord. It is obvious that through the main fibres and their collaterals an extensive connection — partly direct, partly by the link of intermediate neurons — is established with the motor cells on both sides of the cord. But the facts of physiology demonstrate an even ampler connection than the mere anatomical study of the distribution of the root-fibres would suggest. Indeed, the phenomena of strychnine-poisoning seem to show that every afferent fibre is potentially connected with the motor mechanisms of the whole cord, or at least with a very large proportion of them. For in a frog under the influence of this drug, stimulation of the smallest portion of the skin will cause violent and general convulsions, which are unaffected by de- struction of the brain, but cease at once on destruction of the cord (p. 904) . In an unpoisoned reflex frog — that is, a frog in which interference with the single spinal reflexes has been prevented by section of the bulb or destruction of the brain — the movements resulting from stimulation of a given receptive area are by contrast surprisingly limited, localized, and constant. Thus, a harmful stimulus of a certain intensity applied to a toe will elicit time after time a raising of the leg — in other words, an excitation of muscles whose motor nerves arise from cells in the same region of the cord into which the afferent fibres from the receptive skin area enter. The localization of the reflex is in this case without, doubt dependent upon the fact that the connections of the afferent fibres with the group of efferent neurons in question are more direct and more intimate than with any other group. This anatomical isolation of a given reflex arc is never complete, but so far as it goes it may be assumed to be constant and incapable of variation. Under normal conditions the anatomical isolation is always supplemented by a physiological isolation, which is susceptible of variation in the direction either of increase or of diminution. It is therefore a question of great interest how the isolated con- duction of the impulses in a given reflex arc, in so far as it depends upon the physiological condition of the arc and of its connections, is normally achieved. The best answer which can at present be given is that it is not equally easy for a reflex excitation to pass across all the synapses which, are potentially open to it, and that a lowering of the resistance of the synapses in the favoured path is probably quite as important a factor in the isolation as an increase of the resistance in those which are to be barred. Following the path of least resistance, the excitation traverses the synapse or synapses which it is easiest for it to break through. What property of the synapse is associated with resistance to the passage of the impulse is unknown. But it is a variable property, and when a general reduction in the resistance is produced, as by strych- nine or tetanus toxin, an excitation impressed upon a single afferent path may force a great many synapses normally impervious to it. While it is convenient in a preliminary survey to speak of the resist- ance to spreading of the excitation in the cord being diminished by strychnine or by tetanus toxin, we shall see presently that more than this is involved (p. 903). Principle of the Common Path. — In considering the architecture of the cerebro-spinal nervous system as a basis of reflex action, one feature is of such importance as to deserve special mention. The QOO THE CENTRAL NERVOUS SYSTEM afferent neurons, running from the receptive surfaces to the centres, constitute each for its own receptive point a ' private ' path which can only be used by impulses arising at that point, and not by impulses arising at any other point. Through its central connec- tions an afferent neuron from a single point may be put into com- munication with numerous efferent neurons, and thus with numerous and distant effector organs (muscles or glands). Conversely, the efferent portion of a single reflex arc can convey reflex excitations originating in numerous and distant receptive fields. It is the sole path which all efferent impulses — let them originate where they may — must use to reach the end-organ in question. It is therefore not a private but a public path, and may be termed in this relation the final common path (Sherrington). The existence of the common path is of great importance in under- standing the manner in which reflexes are compounded together, a problem absolutely fundamental in nervous co-ordination. One consequence of the existence of a common path is that when, among the receptors which may use it, two are simultaneously stimulated which, when separately excited, produce opposite effects upon the sffector organ, only one of the effects is produced. In other words, impulses which produce the two opposed effects can be successively, but cannot be simultaneously, sent along the common path. Thus, ' excitation of the central end of the afferent root of the eighth or seventh cervical nerve of the monkey evokes reflexly in the same individual animal sometimes flexion at the elbow, sometimes ex- tension. If the excitation be preceded by excitation of the first thoracic root, the result is usually extension; if by excitation of the sixth cervical root, it is usually flexion. Yet though the same root may thus be made to evoke reflex contraction of the flexors or of the extensors, it does not evoke contraction in both flexors and extensors in the same reflex response. . . . The flexor-reflex, when it occurs, seems, therefore, to exclude the extensor- re flex, and vice versa. Either the one or the other results, but not the two together ' (Sherrington). It is obvious that this is an advantageous arrange- ment. An algebraical summation of the opposed effects by the common path would result in a useless action which was neither effective flexion nor effective extension, a compromise and not a co-ordination. The conditions which determine which of two or more competing reflexes shall obtain possession of the final comi path are considered on p. 902. Role of the Receptor in Reflex Action. — The role of the the reflex arc is above all to sift out from the various kinds of im- pressions impinging upon the receiving surface the particular kind to which the appropriate response is the reflex action in question. As will be pointed out in greater detail in the study of the special senses, each kind of afferent end-organ has become adapted to a FUNCTIONS OF THE SPINAL CORD 90! special, or, as it is termed, an ' adequate ' stimulus, so that it is easily affected by this, and with difficulty or not at all by other modes of stimulation. Thus, light is the adequate stimulus of the end-organ of the optic nerve, heat that of the end-organs of the nerves by which we perceive the sensation of warmth, mechanical pressure that of the nerves by which we perceive the sensation of pressure. Other kinds of stimuli are either entirely inactive or much less effective in evoking the particular sensory response. There is every reason to believe that the receptor in the reflex arc occupies -the same position in regard to adequate stimuli as it does when it func- tions as a sense-organ. $herrington has shown that the different kinds of nerve-endings in one and the same area of the skin (in the dog) must be assumed to possess totally different spinal connections, since the movements elicited by stimuli suitable for one form of nerve-ending are quite different from those elicited by stimuli suitable for another. The ' extensor-thrust ' is a reflex obtained in the hind-leg of the dog, and characterized by a brief, strong extension at the hip, knee, and ankle. . It is only elicited by a certain kind of mechanical stimu- lation, best in the spinal dog — i.e., in a dog whose brain has been destroyed or severed from the cord some time before — by pushing the tip of the finger between the plantar cushion and the pads of the toes. The stimulus is similar to that which normally liberates the reaction — namely, the pressure of the ground on the sole of the foot during locomotion. The reflex cannot be obtained by electrical stimula- tion or by any kind of direct stimulation of afferent nerve-trunks. The same is true of the pinna-reflex in the cat — i.e., the backward crumpling of the ear elicited by squeezing or tickling its tip. The scratch-reflex,* a scratching movement of the hind-foot, is much more easily elicited in the spinal dog by mechanical stimulation (rubbing, tickling, or tapping) applied to the skin of the back behind the shoulder than by electrical stimulation, which often fails to evoke it at all. The puzzling fact that, according to surgical ex- perience, many of the internal organs — e.g., the ureters and bile- ducts — can be handled, cut, and sutured without pain, while the passage of a renal calculus or a gall-stone may cause excruciating agony, becomes explicable in view of the apparently slight difference which sometimes distinguishes an adequate from an inadequate stimulus. Thus Sherrington has shown that very distinct reflex effects — e.g., a rise of blood-pressure — can be obtained by sudden distension of the bile-duct by the injection of salt solution into its lumen. Distension is here the adequate form of mechanical stimu- lation, and it is the form induced by the passage of a calculus, while nerve-cutting, although a mechanical stimulus, is not an adequate one. * The scratch-reflex is very easily obtained in. cats during resuscitation after a period of cerebral anaemia. 902 THE CENTRAL NERVOUS SYSTEM Characteristic Properties of the Reflex Arc. — Conduction in reflex arcs shows certain peculiarities when compared with the conduction in nerve-trunks already studied (p. 781) : (i) The direction of the reflex conduction cannot be reversed. There is an absolute block on the passage of impulses backwards through a synapse. (2) The velocity of conduction over the whole reflex arc is much smaller than over a nerve-trunk of equal length. Both of these differences depend mainly on the fact that the impulses must be transmitted from one neuron to another, and very likely on a fundamental property of the synapse. The delay or ' lost time ' in the discharge of the efferent impulses which constitute the reflex response to the excitation of an afferent path increases with the complexity of the response — that is, with the number of neurons and therefore of synapses involved in it. (3) The reflex arc is easily fatigued, easily affected by deprivation of oxygen and by drugs, in comparison with the nerve-trunk. This difference is due to the portion of the arc in the grey matter, including the synapse or synapses. Fatigue expresses itself by an increase in the degree of block or resistance to the passage of impulses along the arc. (4) The reflex end-effect may much outlast the stimulus — in other words, "a marked ' after-dis- charge ' is characteristic of reflexes. The more intense the stimulus which liberates the end-effect, the greater is the duration of the after- discharge. For example, the ' crossed extension reflex ' (extension at the knee, ankle, and hip, produced in the spinal dog by stimula- tion of the skin of the opposite or contralateral hind-limb), when provoked by a stimulus of more than a certain intensity, may outlast the stimulation by ten or fifteen seconds, and the after-discharge may be stronger than any other part of the reflex (Sherrington). (5) A succession of impulses may easily passj.lojjg a reflex arc when one of the series would fail to pass (temporal summation). This does not occur in a nerve-trunk. The first stimulus, though itself unable to produce the reflex effect, facilitates the action of succeeding stimuli, so that summation of the impulses occurs in the cord (Stirling). A stimulus — e.g., a make-induction shock, far too weak to produce the scratch-reflex when applied once only to a point of that area of skin from which the reflex is normally elicited — has been seen to cause the reflex after more than forty shocks had been delivered at the rate of eighteen per second. The facilitation of the passage of an impulse by the previous passage of impulses along the same reflex path recalls a somewhat similar phenomenon already alluded to in connection with the conduction of the propagated disturbance in nerve- fibres (p. 791), although in the case of the reflex arc the effect, it may be supposed, is exerted upon the fields of conjunction, including the synapses, between the different neurons. There is reason to believe that summation in the reflex arc is mainly achieved by the removal of block or resistance. The phenomenon of FUNCTIONS OF THE SPINAL CORD 903 facilitation is probably of great importance in the acquirement of new reactions and in rendering these acquisitions stable. It is prob- ably one of the main physiological foundations of habit, and there- fore of education. In this connection it is important to note that the very same repetition of stimuli which leads to facilitation leads to fatigue when the stimuli are applied in too rapid succession. (6) The rhythm and intensity of the reflex end-effect correspond much less closely with the rhythm and intensity of the stimulus than in nerve-trunks. (7) The phenomena of refractory period (p. 155), inhibition and ' shock,' are much more conspicuous in the reflex arc than in nerve-trunks. Inhibition in Reflex Action. — Special emphasis must be laid upon the part played by inhibition in reflex actions. For the proper carrying out of many reflex movements it is necessary not only that the appropriate effector organ, the appropriate muscle, or group of muscles, should be caused to contract at the proper time, but that their contraction, or that of other muscles, should be diminished or abolished by inhibition, or even rendered for a certain period im- practicable by the establishment somewhere in the reflex arc of a refractory state, which is itself a phenomenon of inhibition. There is good evidence that this is a central inhibition — i.e., it depends on some process occurring in the spinal portion of the reflex arc. As an example of the numerous class of reflexes in which the excitation of certain muscles is accompanied by the inhibition of their antagonists (reciprocal inhibition), we may take the ' flexion reflex,' the flexion at the knee, hip, and ankle of the hind-limb readily elicited in the spinal dog by ' nocuous ' or harmfm stimuli (such as a prick, a strong squeeze, chemical agents, or excessive heat), or by electrical stimuli applied to the skin of the limb or of any afferent nerve of the limb. Sherrington has shown that when the legs of the animal are so prepared that only the flexors can act on one knee, and only the extensors on the other, stimulation of symmetrical points on the two sides in the area of skin (receptive field) from which the flexion reflex can be evoked causes contraction (excitation) of the flexors and simultaneous relaxation (inhibition) of the tone of the extensors. The same is true when corresponding afferent nerve-twigs are stimulated on the two sides. From this it is inferred that each of the nerve-fibres from the receptive field of the reflex divides in the cord into two sets of end-branches (e.g., collaterals) — a set which produces excitation when it is stimulated, and another set which produces inhibition. Reversal of Reflexes. — The difference in action is specific in the sense that no mere change in the kind or intensity of stimulation affects it. Yet there are facts which show that the specificity is not absolutely immutable, and that a change of conditions in the spinal cord may permit excitation of a given group of muscles to be pro- 9o4 THE CENTRAL NERVOUS SYSTEM duced by the stimulation of an afferent path which is primarily inhibitory for them. One of the most striking illustrations of this possibility is seen in the action of strychnine. Stimulation of the internal saphenous nerve below the knee — say in a dog after removal of the cerebrum — is known always to produce inhibition of tl.e portion of the quadriceps extensor whose contraction causes the knee-jerk. If now the animal be poisoned by a small dose of strychnine, stimulation of the nerve causes no longer reflex relaxation, but reflex contraction of the muscle. This fact indicates that the essential action of strychnine is something different from a mere reduction of the resistance to the spread of impulses in the cord (Sherrington). Tetanus toxin produces a similar effect, though more slowly. The reversal of the depressor reflex on the blood- pressure has been previously alluded to (p. 188). A different type of reversal, and one of most interest in connection with the co-ordination of reflexes, is illustrated by such observations as the following: The extensor thrust is only obtained by the adequate form of stimulation de- scribed on p. 901, when the hind-limb is in a condition of flexion. When the leg is passively extended at the time when the stimulus is applied, the response is not the extensor thrust, but flexion of the leg and thigh (direct flexion reflex). The passive assumption of a condition of flexion at the knee and thigh appears, accordingly, to favour the extensor reaction (Sherrington). The observations of Magnus have shown that such relations are general ; for example, the reaction is usually extension when the opposite posterior limb is flexed at the time of stimulation, and flexion when the opposite leg is extended. In the spinal cat, stimulation applied to the tail, especially near the root, elicits always a stroke towards the side on which at the time of stimulation the muscles are extended. The phenomenon is dependent upon the integrity of the afferent nerves of the passively extended or flexed muscles whose position influences the reflex, and of the afferent nerves of the tendons and fascia re- lated to them. The condition of the reflex centres is in some way influenced by impulses conducted along those afferent paths. Not only is the tone of the extensors diminished or abolished during the activity of the flexors, but the contraction of the knee extensors evolved by striking the patellar tendon, which is called the knee-jerk, either fails to appear, or appears but feebly, when the flexion reflex is simultaneously elicited, even when the mechanical antagonism of the flexor contraction has been eliminated by pre- viously detaching the flexors from the knee. The Knee-jerk. — This is sometimes termed a pseudo-reflex. For certain authorities believe that the mechanism by which it is pro- duced is different from that concerned in the reflex blinking of the eyelid, or the reflex retraction of the testicle, or the drawing-up of FUNCTIONS OF THE SPINAL CORD 9<>5 the foot when the sole is tickled. The knee-jerk is obtained in undiminished strength when the nerves of the ligamentum patellae have been divided. It is therefore not a reflex movement caused by stimulation of afferent nerves coming from the tendon, and the name ' tendon-reflex ' is clearly a misnomer. But that it is related in some way or other to afferent impulses is certain, for division of the posterior roots that enter into the anterior crural nerve abolishes the knee-jerk. The phenomenon, according to these authors who deny that it is a true reflex, comes under the head of what is called myotatic irritability — that is, it depends on mechanical stimulation of the slightly-stretched muscle by the pull of the tendon when it is struck. It is necessary for this stimulation that the muscle should be to a certain extent tonically contracted. So that when the afferent fibres are interrupted, or the grey matter of the cord dis- organized, and the reflex tone abolished, the knee-jerk disappears. The strongest objection to considering it an ordinary reflex is the shortness of the interval which elapses between the tap and the jerk, which, according to some observers, is not much greater than the latent period of the quadriceps muscle for direct electrical stimula tion, as measured under the ordinary conditions of its contraction. There is no doubt that the interval is very brief, although somewhat conflicting results have been obtained by different observers for the corrected latency — that is, the period between stimulus and response minus the latent period of the muscle. In man this period seems to be about 0-02 second. In a dog with divided spinal cord the interval was found to be 0-014 t° 0-02 second (Applegarth) ; in the rabbit only 0-008 to 0-005 second (Waller and Gotch). Recent observations in which the electrical response of the muscle as re- corded by the string galvanometer was employed instead of the contraction have yielded results not very different from those obtained by the older methods, o-on to 0-015 second according to Synder. Taking account of the newer observations on the velocity of the nervous impulse (p. 793), it would appear that the interval is not really too short to prevent the knee-jerk from being classified as a true reflex contraction, although a very brief one.* The rein- forcement of the knee-jerk is referred to under another heading (p. 911). It is admitted that, in addition to the direct stimu- lation of the muscle on the same side, the tendon-tap may cause also a true reflex knee-jerk on the opposite side, the interval between tap and contraction being about £ second. Spread or Irradiation of Reflex Action. — As the strength of the stimulus which has been evoking a given reflex movement is in- creased, the reflex effect becomes more and more extensive, spreading out or irradiating in various directions. If, for example, the reflex * There is really now no good reason for regarding the knee-jerk as anything else than a true reflex. The term pseudo-reflex and other terms implying that the knee-jerk is not a reflex should therefore be dropped. go6 in question is the flexion reflex elicited by stimulation of the plantar surface of the hind-foot in the spinal animal, increase of the stimulus will cause, in addition to flexion of the same hind-foot, extension of the opposite hind-limb, then in the homonymous fore-limb (i.e., the limb on the same side) extension at the elbow and retraction at the shoulder, then certain definite movements, the details of which need not detain us here, in the opposite fore-limb, and ultimately also definite movements of the head and tail (Sherrington). Obviously there is a certain orderliness in the spread of the reflexes; they follow a certain regular march; the irradiation in the tangle of the spinal paths is not an indiscriminate one. The same fact emerges quite as clearly when other reflexes are studied in a similar way; and certain laws or rules which define the spread of the impulses in spinal reflexes have been deduced. For descriptive purposes, in dealing with reflex action, it is convenient to consider each lateral half of the cord as divisible into regions each related on the one hand to a certain area of the receptive surface (skin), and on the other to certain muscles. Such regions are those of the neck, including the pinna (cervical), the fore-limb (brachial), the trunk (thoracic), the hind-limb (crural), and the tail (caudal). According to their rela- tion to these regions the spinal reflexes can be classified as ' short ' or ' long.' The short spinal reflexes are those in which the muscular response takes place in the same region as the application of the stimulus. The long reflexes are those evoked when the stimulus is applied to the receptive field of one region, and the response occurs in the musculature of another region. For the short reflexes Sherrington has given a number of rules, which may be stated as follows: (i) The closer together their spinal segments, the easier is it for stimulation of a given efferent root to excite reflex contractions of muscles supplied by a given afferent root. (2) For each afferent root there exists in its own spinal segment (of course, on its own side of the cord) a reflex motor path of as low a threshold (i.e., as easily set into action) and of as high potency (i.e., producing as great a reflex effect) as any open to it anywhere. It has been shown that the afferent nerves of a skeletal muscle are derived from the spinal ganglion corresponding to the segment of the cord containing its motor cells. (3) Motor mechanisms for the skeletal musculature lying in the same region of the cord, and in the selfsame spinal segment, show markedly unequal accessibility to the local afferent channels as judged by the reflex contractions produced. For example, the reflex contraction of the flexors of the knee on the stimulated side, and of the extensors of the opposite knee, is in many animals much more easily elicited than contraction of the extensors of the homonymous and the flexors of the contralateral (i.e., opposite) side. This, however, is not because the last-named extensors and flexors are really incapable of being reflexly affected through the afferent fibres of the corresponding spinal segments, but FUNCTIONS OF THE SPINAL CORD 907 because the reflex effect produced by them is in this case not con- traction, but inhibition. (4) The groups of motor cells contempor- aneously discharged by spinal reflex action innervate synergic muscles (muscles which act in the same direction in effecting a harmonious movement), and not antergic muscles (which antagonize each other). This disproves the old idea that the movements, caused by ex- citation of an efferent spinal root, are co-ordinated synergic move- ments, since at many joints the flexors and extensors both receive motor fibres from one and the same root, and stimulation of the root must simultaneously excite antagonistic muscles. ' The collection of fibres in a motor spinal root does not represent a reflex figure — i.e., a number of simple reflexes occurring simultaneously — nor does the receptive field of a reflex correspond with the distribu- tion of an afferent root.' (5) It follows from (i), (2), and (4) that the spinal reflex move- ment which can be elicited in and from any one spinal region will exhibit much uniformity even when the exciting stimulus is applied at different and distant points within the receptive field. The flexion reflex of the hind-limb, e.g., will have the same general char- acter— i.e., flexion of each of the three main joints — whatever part of the surface of the limb is stimulated. Yet the flexion movement will be strongest at the joint whose flexors are innervated by motor cells situated in a spinal segment near the entrance of the afferent fibres from the stimulated skin area. For the long spinal reflexes it is less easy to deduce definite rules, for they can be less easily and constantly evoked than the short reflexes. The so-called laws of spread formulated by Pfliiger for the long spinal reflexes, and based mainly on observations made in the brainless frog and on clinical records in cases of spinal lesion in man, need not be stated here. For Sherrington has shown that they require serious modification. Especially is this true of Pfliiger's fourth law, that the reflex irradiation spreads always more easily up in the direction of the medulla oblongata, so that stimulation of a fore-limb does not cause reflex contraction of a hind-limb, although excitation of a hind-limb may cause movement of one or both fore- limbs. This law does not hold in the mammal. As a rule, indeed, irradiation takes place more easily down than up the cord. Excita- tion of the skin of the pinna easily causes reflex movements of the limbs, while the reverse is rare. Reflex movements of the hind- limb in the spinal animal are more easily evoked by stimulation of the fore-limb than movements of the fore-limb by stimulation of the hind. It is easier for the irradiation to cross the cord from hind-limb to hind-limb than to pass up from hind- to fore-limb; but it is often easier for irradiation to occur down the cord from fore- to hind-limb than across the cord from one fore-limb to the other. Afferent channels from the skin of the shoulder, through 90S THE CENTRAL NERVOUS SYSTEM which the scratch- reflex is discharged (in the dog), are freely con- nected with efferent paths to the muscles of the hip, knee, and ankle by an uncrossed path descending the lateral column (Sherrington), In cats, after temporary occlusion of the cerebral circulation, which throws the brain out of gear, it is easy to elicit movements of the hind-legs by pinching the fore-paws or the skin of the upper part of the body. The scratch-reflex can also be very readily evoked, and in great intensity, by stimulating the pinna, and is not confined to the side stimulated. In anaemia of the brain and (cervical) cord and subsequent resuscitation, homolateral reflexes (i.e., on the same side as the stimulus) are submerged later and emerge sooner than contralateral reflexes whose centres lie in the area which was rendered anaemic (Pike, Guthrie, and Stewart). Co- Ordination of Reflexes. — The co-ordination or orderly combina- tion of muscular actions for the production of appropriate and har- monious movements is one of the most important functions of the central nervous system. Both the brain and the cord take a share in this co-ordination. The role of the brain will be considered later on, but it is essential to recognize now that many of the movements which the brain directs represent spinal reflexes already synthesized, compounded, or co-ordinated in a very high degree. This is the reason why, in the spinal animal, the inexperienced observer may sometimes be startled by the apparently ' purposive character ' of a reflex movement — the scratch-reflex in the dog or cat, e.g., or the extensive reflex movements of the hind-legs of a brainless frog when the skin is pinched or painted with dilute acid, so plainly directed to the seat of irritation. When a drop of acid is applied to the flank of such a frog, it will attempt to wipe it off with the foot which is situated most conveniently. If this foot be held, it will use the other. These reactions are necessarily purposive in character, since they have been evolved with reference to the ad- vantage of the organism as a whole. They are the sort of complex reactions which the intact animal would have had to improvise by the combination of a considerable number of simple movements when it was executing such defensive reactions, with the conscious purpose of escaping from the irritant, were they not already present as purposive reflexes in the ready-made condition. In the combining of reflexes we may distinguish between simul- taneous combination — i.e., the combination of reflex actions taking place at the same time — and successive combination — i.e., the combination of reflexes in such a way that they follow each other in an orderly sequence. The facts already mentioned in speaking of irradiation afford a partial explanation of the co-ordination of reflexes by simultaneous combination. The movements are orderly and harmonious because the spread of the reflexes is not indis- criminate, but follows a definite ' march,' determined partly by the FUNCTIONS Of THE SPINAL CORD 909 anatomical relations of afferent and efferent paths, partly by the varying resistance of the synapses or other structures whose proper- ties fix the threshold value of the excitation by which an arc can be forced. In general it is not enough that the channel of the final common paths (p. 899) to the muscles whose contraction produces the reflex movement should be thus open to the afferent arcs that elicit the movement; they must be closed to other afferent arcs which might disturb the reflex. Not only so: there is evidence that very frequently the final common paths are, so to say, more widely opened to the afferent arcs in question by the ' reinforcing ' or ' facilitating ' influence of allied, though it may be distant, afferent arcs, which are simultaneously excited (p. 911). Further, the final common paths to antagonistic muscles must also be temporarily closed. The closing of these central connections, or rather the raising of their threshold sufficiently to bar the impulses from passing through the door, is an inhibitory phenomenon. Ex- citation of the desired movements and inhibition of antagonistic movements go hand-in-hand in the simultaneous combination of reflexes. It is obvious that if a movement is to be efficiently exe- cuted, it cannot be the result of a compromise between competing reflexes. A segment of a limb can be either flexed or extended, but cannot at the same time undergo both flexion and extension. In the interests of an effective movement, one or the other must give way utterly. The reflex which eventuaUy prevails makes a clear field for itself by inhibiting all other reflexes which do not co-operate with it. For example, if the receptive area of skin from which the scratch reflex is elicited be stimulated and a painful stimulation be at the same time applied to the foot, we do not obtain a mixed scratch and flexion reflex, which would result in a confused and ineffective combination of movements, but a pure scratch or flexion reflex, as a rule the latter. The successive combination of reflexes is well illustrated by the contraction of the oesophagus in deglutition. First one portion of the tube and then the next below are involved in the reflex action. The combination consists in the orderly sequence. The manner in which this is secured in this class of reflex action has been lumin- ously discussed by Sherrington,* but details cannot be given here. While only allied reflexes — i.e., such as mutually reinforce and therefore harmonize with each other — can be simultaneously com- bined, and antagonistic reflexes cannot, both allied and antagonistic reflexes can be successively combined. An example of the succes- sive combination of allied reflexes is the series of scratch reflexes caused by a parasite travelling across the receptive field of the reflex. An example of the successive combination of antagonistic * ' Integrative Action of the Nervous System, ' to which work the advanced student is referred. 910 TH£ CENfKAL tiEttVOtiS reflexes is afforded when either the scratch reflex or the flexion reflex is induced and caused to interrupt the other while it is pro- ceeding. The transition — e.g., from flexion to scratch reflex — is made without any period of confusion. Thus, if the scratch reflex has been induced and is being executed, and the foot is then pain- fully stimulated, the scratch reflex immediately ceases, and the flexor reflex takes its place. When the flexor reflex has termin- ated, the scratch reflex may be resumed. The same holds good for other antagonistic reflexes. In many cases the avoidance of con- fusion is due to the inhibition of the first reflex, or often to inhibition of the set of muscles which were active in the first reflex combined with excitation of their antagonists (so-called interference). It is obvious that this is an adaptation of great importance. Influence of the Brain on the Spinal Reflexes. — The spinal reflexes can be influenced by impulses descending from the higher centres. For (a) it is a matter of common experience that a reflex movement may be to a certain extent controlled, or prevented altogether by an effort of the will, and it is worthy of remark that only movements which can be voluntarily produced can be voluntarily inhibited. (b) Long-continued muscular contractions may be caused in animals after removal of the cerebral hemispheres by stimulation of afferent nerves — for example, by scratching the mucous membrane of the mouth in a ' brainless ' frog or Necturus. (c) By stimulation of certain of the higher centres reflex movements which would other- wise be elicited may be suppressed or greatly delayed. If the cerebral hemispheres are removed from a frog, and one leg of the animal dipped into dilute acid, a certain interval, the (uncorrected) reflex time, will elapse before the foot is drawn up (p. 996). If, now, a crystal of common salt be applied to the optic lobes, or corpora bigemina.or the upper part of the spinal cord, and the experi- ment repeated, it will be found either that the interval is much lengthened or that the reflex disappears altogether. Stimulation of the optic lobes relaxes the reflex sexual embrace of the male frog when it is present. From such experiments it has been concluded that centres which can inhibit the spinal reflexes are situated in the thalamus, the corpora bigemina and the medulla oblongata of the frog. In mammals, also, there is evidence of the existence of mechanisms in the brain, the excitation of which diminishes the reflex excitability of the cord. For example, stimulation of the frontal convolutions in the dog causes a diminution in the height of reflex contractions of the limbs. Strong stimulation of an afferent nerve may abolish or delay a reflex movement which is being elicited through other receptors, (d) If such, inhibitory mechanisms exist, it is to be supposed that elimination of the brain will render it easier to elicit reflexes from the cord. Experiment shows that this is actually the case. An animal like a frog responds FUNCTIONS OF ftiE SPINAL CORD 911 to stimuli by reflex movements more readily after the medulla oblongata has been divided from the spinal cord or the brain re- moved. In the dog the scratch reflex is elicited more easily after removal of the cerebral cortex or its elimination by cerebral anaemia. In the guinea-pig, after extirpation of the cortex of one hemisphere, the scratch reflex is more readily evoked on the side of the lesion (Brown). That the brain exerts more than a merely inhibitory influence on the production of reflex movements is suggested by many facts. The knee-jerk, for example, is increased or ' reinforced ' if an instant before the tendon is struck the patient makes a voluntary movement or is acted on by a sensory stimulus (Bowditch and Warren). In health it varies in strength with many circumstances which affect the activity of the central nervous system as a whole (Lombard, etc.). It often disppears in pathological lesions, situated high up in the cord in man, and is markedly impaired after high section of the cord in dogs. In hemiplegia (paralysis of one side of the body, caused by disease in the brain) the cutaneous reflexes on the para- lyzed side may sometimes be absent for years. Some observers have even gone so far as to say that under normal conditions the so-called spinal reflexes are really cerebral — in other words, that the afferent impulses run up to the brain and there discharge efferent impulses, which pass down to the motor cells of the anterior horn and cause their discharge. It may be admitted that there is no physiological ground for supposing that the afferent impulses which have to do with the reflex contraction of the muscles of the leg when the sole is tickled, stop short at the motor cells of those spinal segments from which the efferent nerves come off, while the af- ferent impulses which have to do with the sensation of tickling pass up to the brain. The probability is that under ordinary circum- stances such afferent impulses pass up the cord in long afferent paths, as well as directly towards the motor cells along those fibres of the posterior roots and their collaterals which bend forward into the anterior horn at the level of their entrance into the cord. And the only question is whether, as a matter of fact, the spinal motor cells are most easily discharged by the impulses that reach them directly, or by the impulses that come down to them by the round- about way of the brain, and the efferent fibres tiiat connect it with the cord. It is evident that the answer to this question need not be the same for all kinds of animals. It may well be that in the higher animals, in which the cortex has undergone a relatively great development, the spinal motor mechanisms are more easily dis- charged from above than from below, while in lower animals the opposite may be the case. When the cord is cut off from the brain the afferent impulses may overflow more easily into the spinal motor cells since their alternative path is blocked. In the frog, where 912 THE CENTRAL NERVOUS SYSTEM there is already a beaten track between the posterior root-fibres and the cells of the anterior horn, this overflow may be established immediately after section of the cord, and may of itself lead to an exaggeration of the reflexes. In animals like the dog a longer time may be necessary before the unaccustomed route from the end arborizations of the afferent axons and their collaterals to the dendrites or the bodies of the motor cells becomes natural and easy ; in man a still longer interval may be required. Moore and Oertel have made a careful comparative study of reflex action after com- plete section of the cord in the cervical or upper dorsal region, and conclude that the spinal reflexes in the higher animals are far more dependent on the upper portions of the central nervous system than in the frog. Spinal Shock. — The phenomena of spinal shock and its varying severity in different animals may be accounted for by the rupture of the paths normally used in the reflexes. The theory that the shock is due to an inhibition set up by the mechanical injury is untenable. For shock affects only the portion of the central nervous system distal (or aboral) to the lesion. When a dog is allowed to live after transection of the cord in the lower cervical region till shock has been recovered from, a second transection distal to the first is followed by only slight and very transient depression of the reflex power, although the direct effect of the second injury ought, of course, to be as great as that of the first. Finally, according to Sherrington, the condition of the spinal reflex arcs in shock differs from the condition caused by inhibition, and resembles rather a general spinal fatigue in which conduction along the arc and especially across the synapses is difficult and uncertain. This condition is supposed to be due to the loss of a ' tonic ' influence of higher centres, assumed to be necessary for the maintenance of the normal conductivity of the arc. These cranial centres, if they exist, or, at least, the most efficient of them, must be assumed to be situated distal to the cerebral cortex, probably in the pons or mid-brain. For section just behind the pons causes much more severe shock than removal of the cerebral hemispheres. Peripheral Reflex Centres. — The question whether any reflex centres exist outside of the spinal cord and brain, and especially in the sympa- thetic ganglia, has been the subject of a lengthy controversy. That the spinal ganglia cannot act as reflex centres is generally acknowledged, and it is not difficult to see that, for anatomical reasons, this must be so. A reflex arc must, so far as we know, in all highly -organized animals include at least two neurons. There is no proof that an afferent impulse can ascend an axon to a cell-body and there excite an efferent impulse, which, descending the same axon in a separate set of fibrils, gives rise to a reflex contraction, or a reflex secretion. Now, the cells of a spinal ganglion represent the original neuroblasts from which the posterior root-fibres grew out as processes towards the cord on the one side and the periphery on the other. A sensory fibre passing into the ganglion makes connection with a cell by a T-shaped junction and passes on its course again. No afferent fibres run from the nerve-trunk into the ganglion, to end in arborizations around the ganglion cells, and no efferent fibres arise from nerve-cells in the ganglion to pass out into the trunk. For although a slightly greater number of medullatcd fibres of small calibre is found in a spinal nerve-trunk immediately FUNCTIONS OF THE SPINAL CORD gi<5 distal to the junction of the roots than in both roots taken together, this appears to be due to the passage into the nerve (from the grey ramus communicans) of medullated fibres which end in the bloodvessels or other tissue of the ganglion (Dale). Here it is evident that there is no possibility of a complete reflex arc. Indeed, it is not certain that the normal afferent impulses pass through the bodies of the spinal ganglion cells at all. For (i) a negative variation can be observed in the posterior roots above the ganglia on stimulation of the trunk of a fr jg's sciatic nerve more than two days after the death of the animal, when the ganglion cells may be supposed to have completely lost their vitality, and when no reflex negative variation can be detected in the central stump of a severed anterior root on excitation of the sciatic or the corresponding posterior root. Such a reflex action current is normally obtainable from a fresh preparation. (2) When the blood-supply of the posterior root-fibres and the ganglion is cut off without killing the frog, the nerve impulse is still conducted by the fibres, as is shown by the reflex movements elicited on stimulation of the central end of the sciatic, at a time when the nerve-cells show marked histological alterations. (3) Prolonged excitation of the posterior roots or the mixed nerve causes no noticeable microscopical changes in the ganglion cells (Stcinach).* (4) The application of nicotine to a spinal ganglion does not hinder the passage of impulses through the corresponding afferent fibres, if it acts on spinal ganglion cells as it does on sympathetic ganglion cells (p. 182), this must be because the impulses do not require to traverse the ganglion. Axon-Reflexes. — In the ordinary sympathetic ganglia,! also, it is doubtful whether the anatomical foundation for a reflex arc exists, and the most careful physiological experiments have failed to connect them with any reflex function. Sokownin, indeed, observed that stimulation of the central end of the hypogastric nerve caused contractions of the bladder, and he considered these movements to be reflex, the centre being the inferior mesenteric ganglion. Langley and Anderson have also found that when all the nervous connections of the inferior mesenteric ganglion, except the hypogastric nerves, are cut, stimulation of the central end of one hypogastric causes contraction of the bladder, the efferent path being the other hypogastric. In addition, they have observed an apparent reflex excitation of the nerves which supply the erector muscles of the hairs (pilo-motor nerves) through other sympa- thetic ganglia. They believe, however, that in neither case is the action truly reflex, but that it is caused by stimulation of the central ends of motor fibres, which come off from the spinal cord, and in passing through the ganglion give off collateral branches to some of its cells. In the case of the inferior mesenteric ganglion the spinal fibres passing down in the left hypogastric would send branches to arborize around ganglion cells which give origin to fibres of the right hypogastric, and vice versa. When the central end of the left hypogastric is stimulated the excitation is conducted up the spinal fibres, and so reaches their branches, and, through the ganglion cells, the sympathetic fibres of the right hypogastric, which convey it to the muscles of the bladder (see sartorius or gracilis ex- periment of kuhne, p. 792) . Other examples of such axon-reflexes exist. * Hodge obtained changes. In such experiments it is necessary that the ganglion should not be directly excited by electrotonic currents or escape of the stimulating current. f The ganglion cells of Auerbach's and Meissner's plexus in the intestine are not of ordinary sympathetic type, and, as has been previously pointed out, it is probable that they, or some of them, are true reflex centres for the stomach and intestines. 58 9M THE CENTRAL NERVOUS SYSTEM Reflex Time. — When a reflex movement is evoked, a measurable period elapses between the application of the stimulus and the commencement of the movement. This interval may be called the uncorrected reflex time or the latent period of the reflex. A part of the interval is taken up in the transmission of the afferent impulse to the reflex centre, a part in the transmission of the efferent impulse to the muscles, a part represents the latent period of muscular contraction, and the remainder is the time spent in the centre, or the true reflex time. Ordinarily this time, though absolutely short, is relatively so great that the total latent period of a reflex is much longer than when a similar length of nerve-trunk is interposed be- tween the point of application of the stimulus and the muscle. When the conjunctiva or eyelid is stimulated on one side both eye- lids blink. This is a typical reflex action reduced to its simplest expression, and the true reflex time is correspondingly short — only about Jff second (50 a*). An additional T^y second (10 a-) is con- sumed in the passage of the afferent impulse along the fifth nerve to the medulla oblongata, of the efferent impulse from the medulla to the orbicularis palpebrarum along the seventh nerve, and in the latent period of the muscle. When a naked nerve, like the sciatic, is stimulated, the true reflex time is reduced to y^ to -^ second. As estimated by Tiirck's method (p. 996), the uncorrected reflex time is greatly lengthened, it may be to several, or even many, seconds. For here it is evident that the time taken by the acid to soak through the skin and reach the nerve-endings in strength sufficient to stimulate them is included. But even when the peripheral factors remain constant, the central factor may vary. With strong stimulation, e.g., the reflex time is shorter than with weak stimula- tion. With weak stimuli the latent period of the flexion reflex in the dog is usually 60 a- or 120 a-. It may even be as long as 200 a. With strong stimuli it may be as little as 30 cr. Even 22 a has been seen, which is little more than for nerve-trunk conduction. Fatigue of the nerve-centres delays the passage of impulses through them ; and strychnine, while it increases the excitability of the cord, also lengthens the reflex time. Reflexes in Disease. — In order that a reflex action may take place, the reflex arc — afferent nerve, central mechanism, and efferent nerve — must be complete; and, in fact, a whole series of simple reflex move- ments exists, the suppression, diminution, or exaggeration of which can be used in diagnosis as tests of the condition of the reflex arc. It is customary to divide these into superficial reflexes, elicited from receptive fields on the, surface of the body (extero-ceptive fields), and deep reflexes, elicited from receptors in the depth of the organism (proprio- ceptiue fields), especially in the muscles and the tendons and joints con- nected with them. The extero-ceptive reflexes are normally excited by extraneous stimuli acting on the surface from the environment. The proprio-ceptive reflexes are normally excited by changes (muscular *